Besides weekly updates of this guide, I have done a huge revision on 8/13/2012 on Immortal/Sentry all-ins, cleaning up and adding of content to How To React and Scouting, and the Replay section. I know a lot of people are asking about Immortal/Sentry, and this guide now has a ton of content on that!
This is a comprehensive guide to ZvP - it's all of the standard stuff, on everything. The guide is mostly focused on FFE, but I cover all other openings as well (you can find other types of expands in the 1 base section at the bottom).
I think the most noteworthy section is the benchmarking section, and I hope that, along with everything else, helps give people direction in their play.
- Zerg rules out 1 base all-in - Zerg takes third and defends while Toss has map control - Zerg gets map control - Toss tries to take third against Zerg, defends - Zerg tries to prevent Toss from moving out before Hive Tech by either harass, trading/aggression, or turtling. - Toss pushes out with 3+ Colossi to kill Zerg before Broodlords
Standard ZvP opening these days is gasless 14 pool. There are a few reasons why this is standard:
1. It's safe against any cheese. Well, any pool or gas first is, but we rule out hatch first because a good cannon rush with double pylons in a good nook can occur and by the time the pool pops from hatch first, even if you get the pool quite quickly afterwards, Toss will just bank so much money that he can put down so many damn pylons and cannons that 6 lings and all your drones just can't deal with it. You can maybe try patrolling like 4 drones if you really want to go hatch first, but that would be behind 14 pool in terms of econ (also note, a Gateway Opener that reactively gets a Forge, will autowin against Hatch First too).
2. 11 pool is just as vulnerable to cannon rushes. You have to go 10 pool before overlord to really possibly get lings out in time to stop a cannon rush. Even if you go 11 pool, if Toss gets those 3 pylons up, he gets the cannon rush. So if you have to patrol drones, or pull drones, the same if you go 11 pool or 14 pool, you might as well go the more economic 14 pool. I repeat, an 11 pool will not get lings out to deal with a cannon rush or ramp block in time.
3. Speedlings are kind of useless against FFE. He's walled off. Yes, you might be able to pick off a probe easier, but you have total map control for 8 minutes with just 4 lings, so you don't need speed. If toss wants to get that pylon up, he's going to get it up anyways. And it just hurts your drone count too much.
4. Gateway openers just aren't popular. Zerg can generally go fast third as long as they rule out 4 gate, and then play as if it was FFE for the most part, so FFE is the more popular choice unless you know they are going gateway expand (eg close air spawn, crossfire). Even if they go gateway, you just grab gas asap and you'll be fine, or just plant a spine. Besides, even if Toss goes Gateway Expand, he will be so far behind that you will be way ahead anyways. Based on the odds, Pool First is the 'best' opener to flip your coin on
Note: I discuss 14/14 being the best build against 1 base Toss/Gateway openers, in the 1 Base Toss section
After going 14 pool, you'll want to patrol a drone at the ramp to stop ramp blocks. Toss throws down 3 pylons at 17 supply, so pull the drone at 16 supply.
You'll want to get that hatch at 16 supply, ideally. Since you will be pulling 2 drones to deal with the probe anyways (one to patrol ramp, one to follow probe), try micro'ing them to get the hatch down Idra style (if you ever saw his games before, he would do this and try to go hatch first). If not, just get those 4 lings out (you need 4 to deal with the pylon, if you just make 2, he can throw a pylon down and then you need to make those 2 lings and they just come out later to deal with it).
You can also take it at your third, but I don't like doing that because maynarding that far can really screw your econ. I would strongly advise against that - just wait for the 4 lings, Toss hurt himself just as much as your hurt by the delayed hatch by putting down that pylon. This is just my own personal preference though - many pros don't even bother with the natural and go straight for the third, and Toss can do something like add a Cybernetics Core to the pylon that would have made taking the Third a better choice to begin with. I'd say it really depends on the map, what Toss does or seems to be trying to do, and your comfort with it is.
You cannot let Toss get the 3 pylons to block the ramp or the choke point. If he gets those up, it's basically GG - there's no way to win without a major fuck-up by Toss (like not seeing a hatchery in their base, or not realizing that the only thing Zerg can do is Nydus all-in and just not be active to deny Nydus). On most maps you can't 'drone drill', and even the maps you can drone drill, you can't drone drill on every spawn (like bottom right Shakuras). If Toss somehow gets them up, take all but 2-4 drones and try to drone drill, usually you'll have to use a really far away mineral patch that forces you to not watch, but by using shift+right click, it forces the drones to stack up and you can just time it out for when the drones attack. Hopefully Toss is stupid enough to not reinforce, and you can come out ahead, but it's highly unlikely.
So that's why at 16 supply, you send out a drone to block the choke point.
When Toss puts down that first pylon, pull 2-3 drones to nip at it. They will kill it before it finishes, and just a single pylon is no means to panic. If he puts it to block the hatch, or anywhere else where it's very open, don't worry about it - you only need to worry if it's in a blocked off position like behind the minerals.
The more Toss puts down, the more drones you pull. It's important to always be producing more drones and not get supply blocked. By pulling 4-5 drones per building placed down, you should be good. The most important thing to do against a cannon rush, is to block any sort of chokes. Have 1-2 drones behind the mineral lines, patrolling, to make sure Toss doesn't create a shitty situation for you. If Toss gets 2 pylons up behind the natural on Shakuras with his probe in between, you are pretty screwed, so the most important thing is to make sure something like that doesn't happen. With a 14 pool, you will have lings come out right when Toss banks a bunch of money to place a shitton of stuff down, and end the rush.
The important thing is just to pull 4-5 drones per building, and make sure Toss doesn't get a building put in a location that's impossible to hit with 4+ drones at once.
15 Pool can be more economic, but if Toss pylon blocks you, it will put you behind moreso than 14 Pool due to the delayed lings. I prefer 15 pools on larger, 4 player maps.
Standard Third timing is usually around 21-23 supply.
Also, if the map is imba enough to have gold minerals, consider taking it as your third (Antiga close spawn, maybe not, but Metal, always). In my very strong opinion, there's no way for Toss to win if you grab gold on that map vs FFE. They can't do any pressure that would be stronger against a gold than it would be against a standard hatch. Making a precautionary spine would be more than compensated by the huge income boost - it's more than just 20% more income, but an exponential increase due to how zerg's inject mechanics work, as the higher rate of income and drone production in early game quickly snowballs, and with the gold, it snowballs even harder (side note: MULEs beat zerg in short term, but over long term, inject completely surpasses it).
It's extremely important that you evenly distribute your workers amongst your bases. Literally, evenly distributing your workers amongst your bases is like getting a free few additional drones, and not doing it, limits you severely.
For example, 24+24 workers on 2 bases, is less income than 14+14+14 workers, even though it's full 2 base saturation with 48 workers vs 42 workers on 3 bases. And a good, strong, gateway timing, can really happen when you are practically even on workers with Toss. So the only way to win such a battle, is to have a higher income with equal harvester counts, by saturating properly. And if you don't saturate properly, you might as well just go 2 base lair because your income isn't much better anyways and just benefit from the lair. Or, you could learn to improve your game!
If you ever watch 1st person perspective of a pro player - you can see it in replays, but better seen watching streams - you will notice they obsessively box over, select all the drones, and look at, their mining bases. To the uninitiated, you might say "Wow Nestea spams his APM even in the lategame" or "No wonder MVP has such high APM, he just spams all game long and constantly is boxing over his workers". But no, that's not what they are doing. They know that it's extremely important to spread your workers evenly amongst your bases.
Okay, I'm sure you get it now. So how to do it? It depends on your 2nd base timing.
If you get pylon blocked at the natural, and wait until lings tear it down instead of taking another base, you should take half of your drones and send it to the natural. If you don't get pylon blocked, just rally all hatches to the newest hatch - you will actually have an increased income due to the drone's pairing up, and that maynarding in this game actually hurts your income for the most part (due to the mechanics of larva inject allowing you to quickly saturate a new base, and because you would prefer to have more money earlier, than more money long term, because the more drones you have earlier, the better for long term, more so than maynarding). You'll learn when to swap up the rally points, and you can balance things out by where you take gas or make buildings at. This is actually what most pros do, and you won't hit 80+ by 8 minute unless you actually rally correctly to new bases, as opposed to maynarding or not paying attention to worker saturation.
If you compare incomes of Z vs a P who maynards half his workers, you'll notice that you'll actually be ahead by about 200 minerals per minute if you rally instead of maynard. Maynarding in this game actually is not a good idea, unless you have more than 16 workers at the base you pull from. That's why you should rally instead of maynard, unless the new base is late (like if you get pylon blocked).
It's up to you how you maynard them specifically, but what's important, is that you OBSESS over this in the game. Seriously. You need to be balancing them out all of the first 9 minutes. Personally, I am constantly re-adjusting my worker saturations all throughout the game, and constantly re-moving 3-5 drones as I morph new buildings, take gases, et cetera.
As you play more and refine your build, you will get a feel for how to best rally your drones to get the most balance. But this is what I see the best macro pros do - rally drones, instead of taking workers from existing mineral lines.
When you take your gas is a big deal. If you notice you bank gas in your games, chances are you are taking gas too quickly, or too much gas. The standard gas timing tends to be 6:00, or right around when third finishes.
DRG, who seems to have the best macro I've ever seen of someone going fast third in tournament play, tends to take 3 gas after his third has been well established and droned up a bit. You can make 1 gas early, 1 gas late, but the idea is generally have 2 gas up by 7:00 at the latest (personally I feel the 1 gas early/1 late kind of cuts econ a bit too much at the critical early stage of the game, as it's what I did for the longest time).
If you like to go by supply, that means 2 gas is roughly taken around 50 supply.
Standard Roach Warren at 7:30 unless you see no gas (6:30) or 4 gas (7:30+), and Evo Chamber at 7:00+ unless you see no gas (not at all) or get a solid scout off telling you not to get it. These days Zergs don't seem to even get Roach Warren until Lair is done, or if Toss has no gas at the natural, and many will skip it if they see Toss isn't doing a gateway timing, or only make it if they see a possible gateway timing incoming. But, if you aren't so solid in your scouting and gamesense yet, you are better off making one always around 7:30. Make sure to get a 6:30 Roach Warren if Toss has no gas at his natural though, and make a few Roaches at least when it pops, as Toss is doing a quick warp gate pressure.
Evo chamber should be grabbed at 7:15 at earliest if you see gas at the Toss natural, and still aren't exactly sure what Toss is up to. Any earlier than 7:00 is too early. Upgrades won't finish in time to deal with any sort of gateway all-in (35evo+160upgrade= 3 minutes, 5 seconds, or 10:20 from a 7:15 evo). Initial stargate units are best handled with a queen or two, and additional spores added as necessary.
If Toss makes any sort of early zealot aggression, like double gate or gate/core/gate, you should get a Roach Warren at 6:30 (just like if you see no gas at his natural, as you likely won't see any gas at his natural if he does this), but rely on lings if he pressures earlier than that (this is because +1 won't be out on the zealots before 7:30).
If Toss does a hardcore, 6 gate +1/7gate/Huk style zealot heavy 8 gate, that cuts probes, and pushes out at 8:00-8:30 and potentially shooting your base at 8:50, you will pretty much autolose if you go for lair (obviously, if you are going 9:00+ lair, or basically lair after pumping units and knowing you can hold the push, that's fine, but the idea here is that a 6 gate +1 8:30 will kill you if you teched too hard). I'll talk about this more in-depth later, but basically, anything like roach speed or burrow will definitely not come in time to deal with such a push, if you simply don't die in the process, so this is sort of the benchmark we are looking at in regards to lair timing.
On a side note, it's great practice to play with a friend to keep 6 gate +1'ing you, as it's considered the heaviest all-in Toss can do from FFE, so if you can hold this all-in when going fast third, you can possibly hold anything (and thus, why fast third is standard vs FFE).
So lair timing is all about seeing gas at the Toss natural.
Zero gas = No lair (or rather, lair after you are sure you can hold the Toss push, and due to it's timing nature, you'll definitely know when you can be good to go lair) 1 gas = Generally, this is very standard play. Toss is going macro, so you'll want lair, but economy is the focus here. Standard Lair timing for Zerg going fast third is ~7:15 2 gas = Toss is doing an extremely tech heavy attack. So you can delay Lair, as you'll want to keep focusing on economy, as some Toss tech pushes will kill you if your economy is weak.
Getting lair at 50 supply or earlier (which means 42 mining drones basically) will hurt your economy too much. So even if you want to get lair very quickly, for whatever reason, make sure you get at least around 55 supply first. Going lair before 50 supply can hurt your drone count by as much as 10, so make no mistake it hurts your econ to get that lair.
Once you go lair, you generally want to grab gas to go with it. I would recommend you get a 3rd gas, as you will only be making roach/ling. Once you know what Toss is doing, or have enough to defend, get that 4th gas as you plan to get your lair tech of choice. Don't get your 5-6 gases until you absolutely know what's going, and have 60+ drones, and have run on 3-4 gas for a little while (getting 3-6 all at once, even if you are going straight to muta/infestor, will result in gas banking too quickly).
Generally, Toss warpgate finishes from FFE at 8:30+. So bear in mind, even if you see Toss go 4 gas, and getting a stargate and teching hard, you still need to make units at the 8:30 mark. How many units you make with how many drones you make, and how hard you tech up, is completely based on what you scout.
Standard Macro Hatch Timing is 8:00-8:30. You can delay this a little bit if pressure comes on right at 8:00 from a Zero Gas Toss, but otherwise you will need this quite quickly.
If Toss makes a stargate, that's a great reason to make a macro hatch (lings are better than roaches, and it means no super heavy pressure early on). Certain all-ins require the ling production from a Macro Hatch though, like Immortal/Sentry All-ins or Blink.
The most important thing to do when going Fast Third, besides making drones, and making that third hatch, is literally putting an overlord by the Toss natural's gas. Don't worry about Toss trying to fake you out - he needs to grab both gas in his main before his natural nexus finishes in order to start getting his cyber tech out (on a side note, if Toss doesn't grab doublegas in his main, that's usually indicative of FFE double gate before Core, which means heavy zealot pressure early on but much later warp gate and tech).
It's also extremely important to put at least 1 overlord by the Toss main. This is how I send my overlords:
Initial Overlord: Toss natural, or close air base. I check to make sure he doesn't go FFE 2 Gate and Core Timing, then to natural's gas. 9 Overlord: Spot my own natural for cannons, proxies, then to Toss main when nat pops or queen arrives (if he sends probe while you send out lings, he can sneak cannons there and that sucks). 15-17 Overlord: Directly above Third to spot for cannons 23-24 Overlord: Toss third/2nd Overlord to sac 32 Overlord: Where Toss tends to proxy pylons near your base 36 Overlord: Near Toss base to turn into Overseer if you sac 2 overlords into main with 3rd by gas.
Toss generally plants their tech direction at 5:45-6:30, so you want to see what's up at 7:00-7:30. At 7:00 is when you need to think to yourself "Okay, what is Toss doing?" Toss cannot simply just plant the tech after the 7:00 overlord that dies to a single sentry/stalker at 7:40 - by the time it finishes, you will have an overseer!
There are 2 things you are looking for when scouting the Protoss, and can basically identify 99% of what's going on - is his base empty or full of a bunch of gateways, and natural's gas. Additional factors like constant/not constant probe production, chrono on the forge or cybercore, stalker, sentry, or nothing made out of gateway at first, and intuition, can really nail everything else.
I find drone scouting extremely important in ZvP. Quite simply, you need to know if Toss is doing a FFE, or his empty base is actually because of double proxied gateways. If you don't realize it's a doublegate early enough (and he actually banks 3 zealots instead of sending first 1), those zealots can walk into your base while you just threw down the hatchery. A lot of pros don't drone scout, maybe they are being greedy, maybe that's why Nestea lost to Naniwa's doublegate at the last set. If you want to disagree about drone scouting, or be a bit greedy and 'know' that Toss won't doublegate you, while keeping that Overlord over the natural to make sure nothing is incoming, go for it, but if you drone scout, read on.
A 13 drone scout will get into Toss' base in time to block a Nexus First. A drone also beats a probe 1v1.
You just want to make sure Toss is going FFE - if he isn't, check the ramp for gateway openers. Then, just bring the drone home. Also note, FFE should have something thrown down, Nexus or Forge, before Pool pops, so if nothing is up still, he might be doing a proxy 2 gate.
This is really the biggest deal when it comes to scouting. Basically, you keep 1 overlord by the natural's gas, and at least 1 overlord by the main.
The overlord at the natural, is simply seeing how much gas he takes. You will know for sure that Toss took 2 gas at his natural right away, before the nexus finished, if he went gateway/cyber instead of 2 gate before cyber, so he can't fake you out here. Note, when I talk about gas counts at the natural, I am referring to gas taken before 6:30. All 2 base Toss will take 4xGas by 6:30+, it's his gas count up to 6:30 that tells you everything you need to know.
Secondly, you have the at least 1 overlord by the main. If you drone scouted, you know exactly where all his pylons are, and where to send them.
Now we are leaving nothing to chance here, so what the overlord sac's are doing is NOT to see that stargate, or twilight council. What his job is, is to get a look at the gateway count. To be blunt, "Is there a buttload of gateways in his main?". Of course, the overlord will eventually die, but even a stalker will take a while to kill an overlord. If Toss made 8 gateways, there is no way in hell that the overlord will see just an empty base. If Toss made nothing, and your overlord dies seeing nothing, chances are that there are no mass gateways, and he put down tech first instead. Seeing nothing tells you everything - Toss cannot deny overlord scouting.
With a simple yes/no on that, and his gas count, you can really figure out almost exactly what Toss is doing. But with vigilant overlord sac's, you should be able to see exactly everything. Also, Toss can proxy a DT shrine or SG, but they can't really proxy gateways (or rather, it's not worth it too) - an empty base should tell you everything.
As a Rule though - if you still aren't sure what Toss is doing by 10:00, freak out! Put down spores, make lots of units (chances are, any sort of army based attack is obvious by 10:00, with him having a ton of units and all...).
Yes, the above section was just how to scout, not how to interpret what you scout. Here, I will break it down.
So the most important things are: 1. How many geysers did Toss take (and, how did he take them?) by 6:30? 2. Is there a bunch of stuff in his base, or did I see an empty base (you don't make 4 robos, so 'stuff' is always gateways)?
Since seeing the Toss gas count is what is most important, I will break this section down by how much gas you see, and how it's taken.
Note: Toss obviously needs 4 gas on full 2 base saturation, so eventually he will get all 4 gas. What I am talking about is the opening, ie gas count before 6:30. Also, I am referring to gas taken at natural, so '1 gas' really is 3 gas total.
This is either a 4 gate +1 timing attack, or 6 gate +1/7gate/8gate zealot all-in. Either case, you need a 6:30 roach warren.
Against a 4 gate +1, you will really need roaches to deal with it, and need that 6:30 warren. If you 'forget' and get it as late as 7:00, that is okay, but you'll want a spine to buy that bit of time for you, at where you think he'll attack (usually your third).
Against a 6/7/8 gate all-in, you will still want a 6:30 warren, because some variations can hit at 8:00, like huk style zealot 8 gate. Against some variations you can get by with a 7:15 warren, but you won't know which variation he is doing for sure, and the early roaches allow you to push back any attempt to throw a proxy pylon down, which helps you crush him all the easier.
You will need to sac an overlord into the main to see if it's a 4 gate +1 (just a couple gateways, usually 3 around 1 pylon, maybe just see 1-2 since he spreads them out, or if there are a TON of gateways for a 6/7/ gateway all-in.
90% of the time, this is Single Stargate.Toss uses just 1-2 void rays to clear out overlords, maybe do mild harassment, maybe bank up some phoenixes to discourage muta play, and use vr/zealot to secure a third against roach/ling respectively. He can also be aggressive and go voidray/gateway. He will make a single stargate by 6:00, then take a single gas. Most of the time, this is followed up with 3 more gateways. Single stargate should not get 4 gas quickly (yes, maybe there is mind games, but 4 gas should always make you ask if you should make spores, so it's not a very good trick).
A lot can sort of come from this, but basically, you are safe from some super hardcore, early gateway all-in, and can go lair, upgrade, and a macro hatch. Just make units at the general time at about 8:30+ (a higher ratio of drones to units than against zero gas though) until you really figure out what's going on. That's where the overlord sac's come into play.
If you see an empty base, that completely means he is going SG/DT. Remember, by 7:00 he's banked at least 400 gas, it's got to go somewhere.
If the base isn't empty, it is generally just more macro oriented play, or tech oriented gateway. Also, as sort of guidelines: Zealot = banking gas, maybe using them as aggression to distract from heavy tech Sentry = defensive or macro oriented play, or a later gateway push Stalker = aggression
2 gas should be setting off alarm bells, that something extremely tech oriented is coming, and Toss is going to be doing pressure. Whatever it is, drone hard to 9:00, since he won't be able to attack with ground units before then, and that he isn't planning to take a third anytime soon.
2 gas quickly taken, like right when the nexus is done at natural, almost always means double stargate. Confirm this by sac'ing an overlord into the main and seeing nothing at all. A stalker instead of a sentry, or even more telling, absolutely nothing, will be a big tell for it too. If toss is running on 4 gas, remember, that means by 7:30 he's going to have at least 700 gas. It's got to be going somewhere.
Don't rely on seeing the stargates. He could put them in separate locations, maybe to fake you out into thinking it's a single stargate, and double stargate means his entire base and economy is focused onto just 2 buildings. Compare that to, say, 6 gate +1, which has 6 buildings. You can't exactly hide 6 buildings, but you can hide 2. All you need to do, is confirm with the overlords, that the base is empty.
If your overlords see a bunch of gateway in the main though, that means it's not double stargate, but a tech heavy gateway push.
1. Mass blink stalker (lots of stalkers, no sentries) 2. Sentry/Immortal push (lots of sentries, no stalkers). This build tends to grab 2xGas at the same time, usually 6:00.
These are all the same in how you react - you need lair tech, and a stronger economy than you'd have against 6 gate +1.
Finally, he could also be going DTs. He has to have at least 1 gas taken, if not 2. Figuring out DTs is about putting multiple clues together.
1. Gateways, but not mass gateways. DT does require heavy gas, but they do benefit from additional gateways due to WG, of course. If you get a good scout off, and the base feels somewhat small, chances are, it's DT. Toss should have at least 5 gateways and tech planted, or 2 stargates, planted by 7:30. If you don't see any tech, and you see just a small amount of gateways and it isn't 4 gate +1 pressure, it may be DT play. FFE can support way more than just 4 plain gateways, so don't think that because Toss has only a few gates and no SG or robo, you are just safe and can chug along.
2. Zealots. A lot of DT play is preceded by lots of zealots, often as pressure too, so while you are preoccupied with zealots he just a-moved, he is actually setting up his real plan. If you see an oddly high number of zealots, and it's not 4 gate +1 pressure, consider DTs. Sometimes, a gate/core/gate or 2 gate opening is an opening that is meant to transition into DT's.
3. The Twilight Council is not upgrading. Blink takes a long time to finish, with your 7:00 overlord you should definitely see it upgrading, and being chrono'd. Hang around the twilight council if you find one - see if it's upgrading. The animation is actually very subtle, and hard to recognize, so I recommend you take the time to go into unit tester, and see what the differences are. Maybe it's the angle, but it's actually quite hard to distinguish if it's upgrading, but it DOES flash quicker.
4. Nothing has happened yet. You'e chugged along to maybe 9:00 or so. No SG, no mass gateway units at Toss base. DT takes a really long time. If Toss has gas at his nat, you should be asking if it's DT play if it's oddly calm.
Making an overseer when your lair is done if you see Toss take any gas, can be smart, especially if Toss did not go stargate or robo, you aren't sure what Toss did, or you saw a Twilight council (even if it's researching blink).
Finally, I strongly recommend making an Overseer when Lair is done, to see if Toss made 1-5 Gates for an Expand build, or 6+ Gates for a timing, so you can identify if you are safe to start teching immediately, and even cut corners like skipping roach speed.
If toss does not take a Third by 10:00, he's definitely doing some sort of all-in, although all all-ins will push out before 10:00, so if you aren't sure what Toss is doing by 10:00, I'd recommend throwing down some spores and making an Overseer or two at home! You should be able to tell what Toss is doing though, based on his posturing - much like 3 gate Sentry Expand will put a pylon on the low ground and have his units at his ramp, a Toss will have a probe, pylon, and his units near his third by 9:30 if he plans to expand. By 9:30 though, Toss should have taken his third or pushed out, so you should know what Toss is doing pretty quickly. If there is no third by 9:30, you should be very alarmed.
After 8:00-8:30, you will need to start pumping speedlings - even a Fast Third from Toss can do +1 Zealot pressure that can be threatening if you don't have a large chunk of speedlings. However, do NOT make any roaches until you know that Toss is all-inning or pressuring - they will seriously delay your tech, and you always want to be more Ling heavy, instead of Roach heavy. You really only need 10, maybe 20 roaches, at most, against the most dedicated, latest, and strongest of all-ins, and only after they've warped in multiple reinforcements - it's much better to be Ling heavy, not just to keep your economy stronger and tech quicker in the case of a macro game, but also because Lings are the real damage dealers.
As for engaging, always, always, always set up a flank with at least your lings, and grab some of the roaches during the battle to focus fire important targets that a few of them may be in range of (Sentries, Immortals, units with low HP). Toss all-ins also tend to hit before Lair is done.
Cut drone production at 8:00, or ~70 supply, or ~55 drones (if these aren't one and the same, you may have to reflect on your macro), and then going into 100% unit production of roach/ling. You should have started Roach Production around 7:30, as well.
This is largely a test of if you made that roach warren earlier than standard. If you make it later than 7:00, you can really take so much damage that Toss can do whatever he wants, like take his third uncontested, and just win the game one way or another. If you are a little late on your warren (maybe you were preoccupied, maybe the drone got stuck and then cancelled the command), put up a spine at the most likely location of attack (almost always the third). Even if you make that 6:30 warren, I recommend making 1 spine, to buy more time for you, since his initial push of maybe 2-3 gateway units won't kill a morphing spine fast enough that it doesn't buy you a good chunk of time.
All you really need to do, is get that 6:30 warren, and make roaches as soon as it pops. Pro-tip - click on your third (or where you think he is going to attack) and make the roaches at that hatchery. A good 5+ roaches made when warren pops, should push him back easily, and allow you to safely drone/tech up, and make units reactively if Toss continues to warp-in.
As long as you start Lair asap, and continue to drone up, and don't make more than ~5 roaches when Roach Warren finishes, you should come out far ahead of Toss, while being able to make reactive units if Toss decides to pressure more. If you overmake units, you can come out behind, especially if Toss just backs off.
A good sign of Toss' intentions is when he takes his gas at his natural - if Toss takes 2xGas at his natural, you know he plans to transition out soon.
A single stargate will just not have enough air units to do major damage. He will use it to poke and prod though. Don't overreact - if you put down 3 spores at 7:30 when you see that stargate, or even at 8:00 when that void ray kills an overlord, you will be seriously behind. Spores really hurt your econ - 1 drone lost, plus another the mineral count. A good ~7 drones 'lost' due to 3 spores is a real damper to your econ.
You need to 'know' the timing of stargate units. A Void ray takes an entire minute to make, phoenixes, 35. So a 6:00 stargate done at 7:00 completes a void ray at 8:00 which takes maybe 45 seconds to get to your base. So don't flip out when you see the stargate!
So that means when you first realize it's stargate, just maybe get the evo chamber up if you haven't already, and take it as an opportunity to drone up, and tech up, as you know no 9:30 gateway all-in is coming, and any gateway pressure will be light and/or delayed. You really only need to plant the spore as late as when the 1st stargate unit arrives at your base, or at around 8:00 so it finishes in time for the 2nd stargate unit. Just 1 extra queen would really be enough, and growing your AA as his fleet may grow.
It's possible that Toss can put on a lot of pressure with single stargate openers, as Zerg will not have anti-air besides queens and spores until after such a push would win the battle. If Toss makes a 2nd void ray, you'll want an extra queen made, if not 2. - If Toss adds on phoenixes after his first 2 units, it means you do not need to worry about warp gate units pressuring you, or a SG+WG all-in, and that you should add spores. - If Toss does not add on phoenixes after his first 2 units, it means you should worry about warp gate pressure, and a possible SG+WG all-in. The best way to defend such an all-in, is using queens as your AA, with a ling heavy roach/ling army.
If Toss goes for his third though, you will want to make a ton of lings and deny it though. Don't let Toss get his third until he's made a justifiable number of sentries. Do not let Toss 'cheat' and get his third when he shouldn't be allowed too. Void rays are great against roaches, but not lings, so keep an eye for when he tries to take his third.
Once you get lair up, how you respond is up to you. Many times Toss will pump out phoenix to discourage mutas, but you can 'open' with corruptors to end the phoenix harassment (lots of phoenix made to discourage mutas will also harass heavily until your AA arrives), and then pump out the mutas. You can also go hydras, pressure Toss and force him to use his air units to defend and lose some of them, then mutas. You can also just straight up go hydras. You can also just go pure roach/ling and hammer the Toss down, and use queens to defend any air + gateway push.
This is very all-in, in that Toss will have no sentries, and a late robo. If you are able to push Toss back, he will just die because he can't forcefield to protect himself. But it's also a very strong push.
However, sometimes Toss just is very careful, and masses a huge ball of stalkers. This can be extremely frustrating to lose to, Toss can just never attack until he has like 30 stalkers, and then pushes, and wipes you out, even if you have creep, a better concave, et cetera. This is why your macro needs to be good. It may seem like BS, but if you have this happen to you, chances are you aren't droning enough at the start, and when the blink stalker push comes, you are at less than 60 drones. It's very important you have a strong economy to deal with this push, and if you macro well, you should be able to have a force strong enough so that his army will slowly grow weaker and smaller, or he will have to retreat.
It's an extremely difficult push to deal with, but if you see 4 gas taken and lots of gateways, made more clear by lots of stalkers made at the start, probes constantly trying to set up proxy pylons, and chrono'd forge and cyber, you should be able to tell what's up.
Also, how you engage is important. When your army is somewhat even with his, and he blinks, you can give chase and nip away hurt stalkers, but when his cooldown is about to be reset, you should pull back because he'll blink away before you get a killing blow in, resulting in you losing units while he doesn't lose any. Don't press in when he warp's in either. You should have lots of lings, to nip away at him as he blinks - if you do it right, his push will slowly grow weaker as his stalkers all get red health. Think of the moment right when his blink cooldown resets, as invulnerability on his units - you generally don't want to give away units for free (also, a defensive spine can go a long way to making sure this invulnerability isn't used to get free potshots as you retreat to regroup with reinforcements).
There are 2 ways to hold this build - the more conventional straight up fighting with flanking, or base trading. I will cover both here.
Hydras, Infestors, and Drop Tech will always come too late to stop this (that isn't to say that they aren't great way to follow-up after holding). The key to holding this push is not about composition, but good macro - if you do not keep up with injects, even during the battles, you WILL lose as you cannot keep up in ling production.
This push can come as early as 10:20, and push out by 9:30, with 3 Immortals and 6+ Sentries.
Roach/ling You need to keep up with injects with this style, even during battles, as you want to be MUCH more ling heavy, and need to constantly flow out lings during the fight. You also simply cannot win against this if you do not have flanks going on, you need to engage in the open, on creep, and at least from 2 directions. Consider getting a dedicated creep queen on maps like Cloud Kingdom, where the third is a big choke, or laying tumors down at first with the queen there.
You really need to be pumping Lings right from 8:00 to 8:30, and units non-stop from there on out. Lots and lots of Lings. You do not want to make roaches until after Toss pushes out, in case he is doing a robo expand, to which making roaches against in a macro game would severely cut your tech (infestors, hive, bl - even roach speed can be too costly against a macro Toss). You also want to be much more Ling heavy against Immortal/Sentry, like all Toss gateway all-ins, and only need about 10-20 Roaches for this all-in, depending on how big the army has gotten and how many Immortals are present.
You NEED to flank against this type of push, since the main damage comes from the Sentries, especially on creep. You also want to bait forcefields with the lings you already have out when Toss pushes. You won't be able to straight up fight Toss, but you should be able to bait some forcefields with your lings (or, if he doesn't FF until too late, you can snag a few sentry kills). Don't worry too much about the pylons or probes, he will bring a warp prism eventually, and he will likely have a forward pylon out already. It's great if you can get a free probe or pylon, but don't lose lings in the process for anything less than killing sentries.
Eventually, you will need to fight on creep, and you MUST have your forces split at least once, so you can come from both sides. Your Lings will be the main damage dealer, but during the engagement it is important for you to select roaches within range to focus down the immortals. It is imperative you keep up your macro, as you need to be reinforcing with mostly lings. If you a-move straight into his army, you will not win - you HAVE to engage on creep, and you HAVE to flank.
It is important to have creep out against this build, and a macro hatch taken by 8:30, with this style of defense.
Base Trade or: Mutas There are 2 ways to base trade, with roach/ling or mutas. I'll cover the basics of both, first.
You will need a ton of lings out, again, pumped after 8:00-8:30. Try to delay Toss' push as much as you can, but don't lose any lings in the process (unless you snag an immortal or a couple sentries cost efficiently from a klutz). When Toss finally pushes out, you will need to cut unit production, and start massing spines in your natural. This is because with a base trade style, you will lose if Toss is able to go straight into your main, so by massing 8+ spines in the natural, you force Toss to wait for a few warp-ins first, and effectively force him to go to your third instead.
From there, simply base never engage Toss, and base trade. You can pump units at home and try to make the fight at the spines more costly for Toss, or just rush all units to base trade, which you should win easily. Have a drone or few out on the map to remake hatcheries and crucial tech under overlords spewing creep, and you should make all tech in the corner of your main, furthest from the reaches of Toss' push.
Mutas The great thing about Mutas against Robo Builds, is that they really shine against both Robo All-ins, and Robo Expands, which is why fast spire is my tech of choice against Toss who make a Robo.
The earlier you can identify that Toss is going Robo, the more greedy you can be - Leave out that Evolution Chamber, the Roach Warren, and take 4xGas by 7:30 and 6xGas by 8:00 (or to be simple, have all 6 gas taken by 8:00).
From there, it should be pretty easy. The quicker you can make Lair, Spire, the easier it will be to beat the Immortal/Sentry all-in with this style, but understand that doing this would lose against a 6/7/8 Gate All-in, so never cut out the Roach Warren or make a Spire immediately when Lair finishes blindly. You should have 15+ Mutas popping around the time Toss is killing your Third. This is okay, your Mutas will break down the wall-in at Toss' natural, and those 100+ Lings you have will stream in and evaporate his base. By the time he kills your main, you should have 24+ Mutas that will erase any sort of Pylon or Nexus he tries to throw up in a feable attempt to win the base trade, especially if you keep your lings alive.
If you made a Roach Warren or Evo before realizing Toss is doing a Robo Build, you can still pull this off, just make sure you grabbed 6 gas by maybe 8:15 at the latest, and that you started Spire immediately when Lair finished. If you don't realize it's a Robo build until about 20 seconds after Lair finishes, it will be too late for you to do this style.
If Toss goes Robo Expand, quick mutas can just win the game instantly, depending on how late his third is (the later the third, the more vulnerable Toss will be to mutas).
For more discussion on Mutas vs Immortal/Sentry, as well as all around discussion on dealing with Immortal/Sentry, especially with base trading, check out Nestea's Anti-Immortal/Sentry Build, complete with multiple replay examples and talk of pro strategies and results against Immortal/Sentry.
As Toss has no sentries, this build is completely all-in. Toss needs to do more damage than killing just a single base to come out ahead, and with no sentries, once you push back, you win the game.
As soon as you see quick 4 gas, alarm bells should be going off about how a Double stargate is possible, confirmed by sac'cing an overlord and seeing an empty base. You can't really hide (or proxy) 6 gateways, but you can hide (or proxy) 2 stargates.
A void ray takes a minute, stargate takes a minute, phoenixes, 35 seconds. So double stargate means Toss has 4 void rays at 10:00. If at 9:00 you aren't sure what's up and don't see massing of gateway units at Toss' base, it's definitely double stargate.
Basically, make queens to 'counter' void rays, and spores to 'counter' phoenixes. Don't plant them all immediately, but eventually you will need 2-3 spores at the place of contention with lots of queens. Void rays can avoid spores, but phoenixes can't, so balance it out, and drone fucking hard, and grab 4 gas and lair quickly.
I would recommend you get corruptors to push back, as he won't have more than 4 void rays and phoenixes just tickle corruptors. The problem with hydras, is that you need more than 1 hydra per phoenix, and he can pick them off easily while they rally out of different hatches, although if you have tons of queens, they can protect them. Since Toss won't have a significant ground army anytime soon, Colossi are usually the follow up anyways, so corruptors are better there. Corruptors also let you transition extremely quick into broodlords if he finally gets sentries, because no way in hell he'll be able to push on ground. Really, you can do whatever you want, if he doesn't kill a base, you win. He has no third against mass lings.
This is the only instance where infestors are okay against 2 base Toss - they do make sure he doesn't ever reach critical mass, and while he will take his third eventually, you can fast track to BL. I don't really like infestors as the answer, because they will have a hard time pushing back the air fleet until you get a few out with energy.
Also, an interesting follow up to the corruptors, is speedbanes, since there are no sentries. It's what Nestea does. Sometimes Toss throws the gateways down before committing too much though, so make sure he committed alot before going speedbanes. Deny third, win with whatever. Just make sure to kill off a lot of the void rays with corruptors first.
Finally: Make sure to trade with his void rays before he gets 6+ critical mass that will kill anything you could make. That's a stupid way to lose. You don't need to be cost efficient, but 8+ void rays with phoenix support and ever increasing numbers will destroy anything but mass infestor, and you will not get mass infestors in time when he already has 4+ void rays and lots of phoenixes by the time your lair is done, unless you went straight for infestors. This is fairly easy to do with Corruptors, as phoenix just tickle them and you can snipe the voids very quickly with them once you have 8+.
Toss can support triple stargate from FFE. The attack will hit a good 1-2 minutes later, and if you macro well, you don't even have to really prepare for it - you will have enough roach/ling to simply force him to go back home as he only kills 1 base as you save the drones and mass up spores at your other bases to keep them alive during the base trade, and have corruptors soon on their way when you noticed Toss was on double gas and not going for a gateway push. It's really just a bad build - it's the same as double stargate, and although it's slightly stronger, it hits wayyyy later (a double stargate into triple stargate, if anything, is 'better' as an all-in).
But just because no double stargate hits you at 10:00, doesn't mean you should simply forget the damning fact you saw quick 4 gas, and a fairly empty base, with your overlord scout at 7:00. Maybe you even just thought it was a single stargate, but single stargate never gets 4 gas quickly at the same time, and even if he did do that, it's okay you made a spore and extra queen at your main and third just to be precautionary. This is really about the rule I stated above: If you are still clueless at 10:00 on what's going on, freak the fuck out!
Also, once you survive, you really need to make sure you trade with those voids right away, as he can build up a huge amount of VR's really quickly.
Anyways that's most of the scary stuff. If Toss is on 3 gas, there's lots of standard plays he can make - 5 gate robo, 5 gate twilight, et cetera. This guide is just about the opening!
The rest of the game basically will go like this: - Get map control, delay Toss third - Keep Toss from moving out with a deathball before you have hive by harassing/trading/turtling - Get Bl/infestor, win
If you go roaches, you will need to smash Toss' army as soon as maxed to lower his deathball size so you can get Broodlord safely (my preferred way to play this style is roach/banerain/infestor, but this style is what roach/hydra basically was). If you go the muta route, you use mutas to contain Toss on 3 base while teching to BL/Infestor once Toss has his third up and a sizeable army. If you go roach/hydra/corruptor, you need to be hyper-aggressive and trade over and over and deny the third forever (works less on maps with chokes). Finally, you can turtle Stephano-style by splitting the map with 40+ spines and quick mass infestors to fast-track to hive.
Since going Fast Third is all about how many drones you can make by 8:00, it's important to benchmark yourself if you want to improve. You'd be surprised to learn, that anyone who isn't High Masters, just does not make drones. They don't. They're macro is complete garbage.
Don't believe me? What your replay. Do you *ever* have idle larva before the 8:00 mark? Do you ever have more than 200 minerals? Yes? Then that's a problem. Also, please tell me what supply you hit. Because unless you hit at least 65 supply, then your macro is complete garbage. You have to hit at least 70 supply. Note - I said 70 supply, not drones (that's impossible, you will have at least 10 of that supply tied up in larva at 8:00). Even if Toss pressures you and you had to make a lot of units, you should still hit 70+ supply (only excuse is a cannon rush attempt at the start).
So. What do people get?
DRG, who has the best macro in the world, from what I've observed with this in mind, consistently hits 80+ supply!. That's just beast. Bear in mind most masters struggle to hit past 60, and diamonds barely leave 50. It's rare to see any other pro hit 70+ in tournament play, and I consistently see Code A 'scrubs' below 70. The ones who hit 70+, are almost always the usual suspects of top tier players, or macro players - Idra, DRG, Nestea, Stephano. DRG is the only person I've seen hit 80+, and as high as 83, in a tournament game.
Nestea usually hits 65, although he has hit 70 plenty of times.
You need to hit at least 60 to be 'in the game'. If you can't get past 60, you have serious macro deficiencies, and you really want to strive to hit at least 70.
If you are having trouble, first, check your gas timing. Are you banking more than 200 gas at any point? You should only have about 100 gas total mined at the 7:00 mark, at most. If you are getting more than that, you need to work on your macro.
If you get supply blocked a single time, that usually cuts your supply count by about 10. Literally, a single supply block is like losing 10 drones for free. It hurts that much. If you make an overlord at 46/52 instead of 50/52, that cuts your drone count by about 2-3 each time. Don't make overlords too early, and until you hit around 50+ supply, you should *never* have 2 overlords being made at the same time unless one is about to die (eg sac'd overlords). Making 2 overlords at once, costs you about 3-4 drones by the 8:00 mark. So it's not just supply blocks that you need to avoid, but you need to learn to make overlords on time, not too early not too late (I find that most people make the 52 overlord way too early, like at 46).
After 50+ supply, a lot tends to go on, and you are probably losing an overlord as a sacrificial scout. A lot of pros tend to make 2-3 overlords at once at 50+ supply, because you will have 3x injects popping off. 54+3x8=78 total. The way with it working with 3x larva injects popping off, making 2-3 overlords at once at 50+ supply/7:00+, is a great way to go from the 50's to 70's from 7 to 8 minute.
It's a hell of a lot easier to hold a blink stalker all-in with 65 drones instead of 55 drones. Chances are, if you have games where you go "wtf Toss just masses a huge derp of stalkers/units/etc and wins", check your supply at 8:00. If you aren't past 60 supply, you've pretty much already lost the game, really. If it's in the 50's at the 8:00 mark, I don't think it's possible for you to win unless the opponent is just as bad or is similarly behind (eg failed cannon rush).
After about 30 supply, you need to be extremely aware of when you ever start an overlord, and how many overlords you have morphing. Making an overlord when you already have an overlord making can cost you about 2-4 supply, so it's really, really, important that you know whenever you start that overlord. By doing this, you will also be aware when you need to make overlords. You should also have your overlord timings mapped out to at least 60 supply (17, 23, 32, 36, 51, 55!), and understand what each additional overlord does (so at 54/54, 2 overlords is +16, so that would put you at 70!).
It is definitely possible though, to hit 70+ supply as a non-pro player, and I've even hit 80+ supply a couple times on ladder (it's rare, I can assure you that much). However, I consistently hit 70+ these days, with a good mind on macro, and I can assure you, everything from all-ins, to pressuring, to macro games, are much, sooo much easier, when you hit 70+ supply than if you are muddling at 62.
As I stated above, you never want to start Roach production until Toss pushes out with pressure. Lings are the core for dealing with Toss all-ins, and are 'free' in the sense that they will not delay your tech. Doing something like making 20 +1 Speedroaches against a Toss who actually going for a Fast Third, will get you killed, in that your Infestors, Hive, will just be too late. If Toss is going for a macro game, you may even want to totally skip Roach tech, if you can identify it early enough. Against Quick Third Toss, you should have Broodlords in time if you skip Roach Tech, and fast track to Hive, and 30+ Spines asap, if you want to play the macro game.
Once you've defended your third, you should regain map control as your economy has kicked, you have enough tech (roach speed, upgrades, etc), and your exponentially increasing rate of production has surpassed Toss. What you goal should be, is making the least amount of roaches necessary. ~20 Roaches should be enough to defend most pushes (of course, with ling support, but lings cost no gas, meaning faster hive, so they are 'free' in the sense that they don't cost you time).
Of course, you can be aggressive and make more roaches to deny the Toss third, but your plan should never be to straight up win the game with roaches alone (however, it's okay to be opportunistic). It's extremely all-in to max out on roaches, although there are definitely great times to do this, such as holding early pressure or knowing Toss is far behind or being too greedy, or the map favors it (wide open natural, for example). You generally want to avoid going max roaches, unless you see a sure-fire vulnerability in Toss. If you max roaches, and Toss doesn't die, and gets his third in reasonable time, you may die to a colossus based push simply because you won't have the tech to deal with it.
However, it's not to say that maxing roaches is easy for Toss to deal with, and many pros have made a name for themselves by maxing roaches and ending the game there (or at least delaying the third long enough to get a huge lead).
Eventually, Toss will get either enough blink stalkers, sentries with stargate support, or immortal/sentry, to take their third, so it's important you understand when making more roaches won't delay the third any longer. All you want to do is keep Toss honest. If you want to continue aggression, consider going with lings instead of roaches so you aren't costing yourself time.
So you really want to throw down your 'tech of choice' as soon as you know you are safe to hold Toss aggression, and have map control, to move towards the late-game.
I have talked largely about defending your third, but now I will move on later into the game. The following is how to play the game once you have defended against Toss or see him posturing for his third. You can also switch between the following playstyles (mass roach into muta, or hydra into muta, or muta into infestor).
The point of this is to keep Toss stuck in his base for so long, that by the time he can push out, you have your hive tech.
This is muta play. Mutalisks are a great option no matter what Toss does, even if he opens stargate or blink. These are best played on maps that have lots of airspace for mutas, and counterattack routes for your lings, and the larger maps.
Basically, you just make lots of mutas, don't make any more roaches (it's important you keep your initial roaches alive though, they afford you enough map control to keep Toss in his base), and harass, harass, harass, and expand, expand, expand. Toss will have 2 options:
1. Base trade. If you lose too many mutas by being too aggressive, or if you stay on muta tech too long and Toss finally has mass cannons on his bases that forces you to lose too many mutas to base trade quick enough and/or has storm tech out, you will lose. Pump lots of lings, and takes bases elsewhere, and run your queens and drones away from any base he starts to attack - you really want to keep as many drones as possible. Making a 2nd spire far away, before he kills your original spire, is helpful as well, and your spawning pool is also important, as lings are useful in this scenario. Spines are okay to buy time, but he will likely steamroll spines without support.
But simply put, you are killing his base way faster than he will kill yours. Focus on cannons or pylons powering cannons, cybernetics core, and nexii once his army is far enough away, so his buildings are all revealed. With 20+ mutas, you will always do more damage in the base trade then he can, unless you lost too many and/or toss has tons of blink stalkers and that storm tech. You can find a great guide on ZvP muta play in Decaf's ZvP Muta guide here: http://www.teamliquid.net/forum/viewmessage.php?topic_id=277693
2. Toss doesn't base trade. The problem with mutas, is that they can't really engage the Toss army. If you never move up to hive tech, and stay on mutas too long, eventually Toss will have storm, archons, mothership, just a really nasty army, and because you are never killing any army, it will just slowly grow until even a pure 3/3 broodlord army can't handle his 3/3/3 mass archon/colossi/mothership army, even if he only has had 1 mining base for the last 30 minutes at any one time.
You really want to move towards hive tech as soon as Toss starts to stabilize and has his third set up, an army that starts to become intimidating, or throws down his HT tech.
Also, many Zergs stay on lairtech instead of going toward hive. If you would prefer to ramp up the aggression and try to end the game, instead of going for hive, that's up to you (going for baneling+speed, maxing out on mutas and encouraging the base trade, etc).
Note: If Toss goes for colossi on 2 base (not as a timing, just gets them like deathball derp) and just masses a 2 base colossi deathball, go for mutas (if he goes fast colossi push, mass roach will own it, I'm saying if they are derping). And in general, if Toss is staying on 2 base because he can't take his third because you have total map control, mutas are a great way to end the game and avoid any BS, and just base trade if he ever pushes out.
In regards to stargate openers: There are two choices: 1. Go for hydras before mutas, push, and force Toss to use his air units to defend. You can kill enough phoenix with the hydras to secure clear airspace for the mutas. 2. Add a few corruptors to your muta ball. They aren't that much slower, and don't be afraid of a few phoenix. If toss tries to go for phoenix range, his army will be much smaller, and his colossi later - identify this double stargate phoenix range move, and cut muta production and switch to infestors and a quicker hivetech (Stephano said that when he lost 20+ mutas to White-ra's ~6 +range phoenixes all in 10 seconds, he knew he won the game because he just go to infestors quicker).
This is about trading armies with Toss aggressively, so that he can never move out until you have a huge hive tech army.
You could go: 1. Roach/Hydra/Corruptor. Somewhat 'outdated' nowadays 2. Mass Roach followed by Infestors 3. Mass Roach (you should max out 200/200 before 12 minute) 4. Roach/Banelingrain/Infestor (this was my preferred style for over a year)
When going roach/bane/infestor, you should get infestors at about 160 supply. You don't need many roaches - this is a supply intensive build, and banelings are the most gas costly per supply units in the game (more so than motherships, carriers, BCs, ghosts, etc). You really just make those 20-30 roaches as necessary at the start, and the rest is lings. You use a ling heavy army with roaches to threaten Toss and delay his third as much as possible, and use that 6 gas for those ~160 infestors, and then droptech + baneling to finish exactly at 200/200. You only need like 20-30 roaches - your army will look rather puny, as so much supply will be tied up in ~8+ overlords filled with banes, and you only need 2-3 infestors (they will gain energy, you don't want to waste time getting lots of infestors or use up all your gas in infestors, you just need 4 FG to FG the toss army twice to hold them in place).
You then turn all the lings into banes. For hotkeys: 1. Army + Overlords 2. Infestors 3. Overlords.
Basically, 1a, FG, and then right before selecting overlords, you FG again, and then select the overlords, and redirect them to bane rain all over. After this big battle, which should be done in a tight area aggressively at Toss' base, you can shift+ctrl+click the overlords in 1 to deselect all of them, and then hit ctrl+1 to remap the control group so no overlords are in it (they can be really awkward and dead weight after the drops). You will always demolish the Toss army, but if you keep your infestors alive and lots of the roaches alive, you should be able to remax on 10+ infestors, pressure again, and during this pressure, be going quick for hive tech, so then after this attack, you remax on broodlords. If you don't keep much of your army (20 roaches and all those banes will usually all die) and it obviously didn't go great, remax on roach/banerain.
You should definitely have burrow to get those infestors out of there. It's really important. Also, when banedrop is ready, I like to send 3 overlords out for banerain - 1 to each base of Toss, and you can shift command it so they go around the edges, and drop automatically (obviously, manually dropping is better, but I find being active with the army and just telling the overlords to shift move and then shift "d" at the middle of a mineral line, all 3 at the same time, much better).
The biggest counter to this, in my experience, is not air play (you just go heavier on infestors), not stalkers (few more infestors, more bane rain), but double robo colossi. But even with double robo, you should be 200/200 and ready to attack before he has 5+ colossi. Also, you can't use ling/banerain/infestor, because once Toss has 3+ colossi, all of the lings will just fry instantly, and you can't NP to prevent that like you could pre-patch.
Personally, I stopped playing this way because the tech cost is really the same as pure broodlord army. The difference, is that the tech time is a bit quicker, but I've been liking muta or mass spine play instead as a means to delay Toss for that little bit more of tech time.
In a fair game, your 200/200 roach/banerain/infestor army will always demolish the Toss army. Toss can definitely kill more of your shit or retreat with a little more of his gas costly units if the engagement goes well (however, if you mismicro, which is definitely possible with an attack that requires 150+ apm to pull off, you can lose everything, so practice in unit tester first) so that you can't just go to broodlords though. But the benefit of this style, is that it will hit right before Toss is at critical mass, whereas when going hive tech, the major concern is that Toss will always hit before you have broodlords ready to go (unless you delay him or turtle hard, with the other playstyles, to buy that time).
If you trade well, go to hive tech. If you don't trade well, you will need to remax on your t2 army. Better on maps that are wide open at both the natural and third. This style is somewhat riskier to play because if you don't do enough damage, you will lose because your hive tech WILL be late. But, sometimes noticing that Toss is being too greedy or is far behind, and taking advantage of that with aggression, is the correct strategic choice.
You can also play this style, and then transition into another (going roach/hydra to snipe stargate units, into muta, for example), or be more or less all-in with it (going mass roach, but grabbing 6 gas instead of the usual 4, to transition into muta/infestor afterwards).
This is about massing tons of spines with infestor support, so that Toss can't push you until you have a huge hive tech army. This is different from roach/infestor, in that you aren't trying to be aggressive with Roach+FG or Roach+IT, as you are trying to turtle behind mass spines, and make as few roaches as possible, only making them in times of panic, so that you may get more gas for hive and infestors instead.
This style is best used on maps mutas are bad on - single attack routes, lack of airspace (means easier to defend with infestors), and maps trading is bad on - Toss can get a very easy third (entombed valley, antiga), lack of open space.
But basically, you rush for infestors as soon as you realize Toss is going for his third, and you have established map control with roach/ling. You rush towards 100+ drones right away, expand a ton when Toss goes for his third, and lay down 40+ spines, replacing the drones constantly. Your infestors are very cost efficient, and mass infestor+spine is extremely deadly. As Toss gets colossi tech, you get spire, and when he has 3+ colossi and pushes out, you should have 15+ corruptors. If he a-moves into the spines, mass FG will destroy his army. If he moves colossi forward to snipe the spines, you FG them and then snipe them with the corruptors.
As long as you get those mass infestors out, and let them pool energy, early enough, and get enough drones and keep replacing them, you should definitely be able to buy enough time to get those 15+ corruptors into mass broodlords.
I would say the problem with this style, is because you are turtling so hard, Toss can get 4+ bases, and be able to compete with your pure broodlord army with tons of gateways for archons, and a mothership. But if you really believe that bl/infestor is superior to Toss, go for it. It's definitely an extremely strong style of play, especially if Toss opens stargate or some sort of opener that allows you to skimp on roaches.
If toss goes for a Third, you can instantly take the game to lategame! A Toss taking his third gives you free pass to fourth, infestation pit, and eventually hive, as well as skipping Roach tech if Toss did not do a 2 base opening. You can also play mid-game aggression as well.
The checklist for Zerg lategame is as follows:
Fourth (and soon, 5th)
Infestation Pit, Hive
Double Evo
The end-game goal for Zerg, is pure Broodlord/Infestor (gone are the day of a huge roach army with just 5 broodlords, or broodlords with hydralisk support). I'd recommend at minimum, 5 bases for BL tech, so as soon as Toss takes his third, you will want to take a fourth and a fifth quickly afterwards.
I would recommend double evo melee/carapace upgrades, as broodlings are the real damage dealer. The many queens you made to inject your hatches are great for support, and if you didn't mass spines before, now is your time. Just make sure you ALWAYS have 70+ drones - if you cut too many drones for spines or more broodlords, you will just get crushed by a counterpush if Toss can push you back, like a threatening mothership.
Tactics Get NP later on, really just for motherships. If you can be lucky enough to snag a NP, you win instantly. However, my preferred method of dealing with Motherships, is having 6+ Corruptors with upgraded air - Corruptors take down motherships insanely fast with Corruption (which is far from a useless spell, go test it yourself and see), especially with FG keeping the Mothership from getting in range of the Broods, and IT spam under the broods as well to keep the mothership at bay. Ground units will focus on the Broodlings and IT/Broodlords before the Corruptors.
I don't find spreading broodlords that useful - you only have what, 10-15 broodlords at most? So getting a vortex on just 5 broodlords is enough to completely screw you over. But of course, you should spread. I like to move command my broodlord army past where I'm headed, and use hold position when I want to attack, to keep them spread out (like siege tanks), so if he vortexes the front, the back 10 broodlords will rain down the pain.
Whatever you do though, you need to be extremely careful against motherships. Always have some energy on your infestors, and always pull back before the mothership gets in range to vortex you (9 range, by the way!).
You also need to make a good 5+ overseers when mothership comes out. I like to add a few of them to my infestor control group, and a few to my broodlord control group. You should be spamming changelings all the time as well - not to put in his army, but to spot where his army position is. Put them all over the place in front of his bases, so you can know where his army is located at all times (or rather, know where his army is *not* at, at all times). Changelings are sooooo awesome in lategame, they really are. It's like a ling burrowed in front of their base, but free, and you can make a million of them.
Make sure to add spines and spores in a Sim City around your most important piece of tech, the Greater Spire - adding Evo's around it also is not a bad idea. If you add a 2nd Spire for upgrades, make sure to put it far away from the original Spire, for insurance reasons.
Ultralisks in ZvP Symbol and BbongBbong have been 2 Zergs who are big proponents of using Ultra/bane in ZvP. The idea with Ultras is, if Toss goes for a third, instead of massing spines, you get Baneling Speed, NP, and Ultras asap. If Toss does a 3 base 3+ Colossus push asap, you should be able to crush it with the Ultras breaking forcefields for the banes to wreck everything. This is perfectly viable, and the ultras are a sort of 'tier 2.5' unit before broodlords. The upside is this gives you a chance to break Toss with aggression, but the downside is that if Toss goes fourth, templar tech, he will be much more prepared against the eventual broodlords. Check out this thread on Ultras in ZvP (Symbol Style) for further discussion.
Extreme Lategame In Extreme lategame, you will need a second spire, and you should get +3 Missiles for Infested Terrans as well. This stage of the game will focus a lot on Toss harassing you with SpeedPrisms, and how well you can Sim City/Spine/Spore your tech and future bases. If you massed spines in the mid-game, re-root them to new bases as frontal attacks are less of a concern once a large BL fleet is up. If Toss goes for Carriers or Void Rays, its very important that you focus on upgrades, and have a few Broodlords to deal with Storm so that Corruptor/IT Army can trade efficiently.
Avoiding Ridiculous Losses or: How to translate a Lead into The Win + Show Spoiler +
We've all played games where we were way ahead of Toss, held their stupid crap, BM'd the crap out of them, and then lost to some derp like mass void rays or pure colossi or an archon toilet or DTs. So how to avoid that?
DTs When Toss is on 3 base, get a spore and 2 spines at each base. If you are ahead, you should do this quicker. Desperate Toss will always try DTs before leaving the game.
Warp Prism Harass 2-5 spines in smart locations, like next to buildings or in mineral lines, should stop Warp Prism play. Add on spines and spores as the game goes on or warp prism play increases.
Mass Void Rays Well you should have queens early on. A single stargate won't get enough out quickly, to kill you, and should be obvious by lack of gateway units or gateways in his base, and lots of gas taken. However, your goal should be to eventually kill those goddamn void rays - especially if you defended a double stargate all-in. If you let him build up a critical mass, you can lose. Be aggressive once you have that macro lead and your anti-air units out, and kill them off if he's massing them. If you have a lead, it's okay to be cost inefficient.
This is not some ambiguous thread. It's extremely useful. If you are on 3 bases vs 2, or 5 bases vs 3, your "E" is good. Now, work on on Tech and Army. If you lost a stupid game even though you were way ahead in econ, check the army values when you lost the game. Chances are, you were way behind in army value, and 1k less in roaches will never beat a higher value army of colossi, +3 upgrades, etc.
Basically, once you've secured your third, and held off that Toss aggression with such a huge lead, you no longer need to focus on your econ. You already have more bases. You don't need more drones, especially once you've reached 70. You need to match what Toss is doing. Match your tech, to his tech. Match your army, to his army. With a higher econ, but even army and tech, there is no way to lose. Too many times, people lose because of their 6 bases vs a 2 or 3 base Toss, because they only have roaches against sentry/colossi/blink/upgraded deathball. Unless you have lots of mutas and Toss doesn't have 3 bases with HT and lots of blink stalkers, you will never beat colossi without broodlord tech.
Once Toss makes the move toward more econ, then you can get that 4th, or 5-6-7 bases. Just because you aren't growing your econ, doesn't mean you aren't growing. Tech and army are also great ways to 'expand'.
Never go below 70 drones in lategame A pure broodlord/infestor army, with 70 drones, and a few queens, really ends up to be like what, 10-15 broodlords, and 10 infestors? It's kind of tiny. But when you start making spines, and not replacing them with drones but instead more bl/infestor, you leave yourself vulnerable. 70+ drones is good, but if you are at 60ish drones, you will just get rolled by a counterpush if you don't win, and it's simply not worth 3 more broodlords for a much lower income that won't be sufficient to remax on t3 units. The easiest way to lose, is to max out on bl/infestor, and have only 60 drones because you wanted more broodlords. You won't always win with the broodlords right away - the game may drag on, toss may push back because of vortex or you not having infestor energy or sniping bases with mobile units. So ALWAYS make sure to replace those drones, so you always have 70+ drones!
There's nothing wrong with having 90 drones and losing your smaller bl/infestor army, and then remaxing very quickly behind mass spines to buy that bit of time, but it will cause you to lose way too many games if you have only 60 drones. It takes a long time, and a ton of money to remake bl/infestor. Don't let the super supply efficient Toss army get a chance to win if he gets a slightly lucky engagement. Leave nothing to luck. With 70+ drones, you can always trick out a 220/200 army anyways. This is a huge mistake of lower level players.
Respect the Mothership Yea, so you were on 4 base while toss was on 2. You crushed his army multiple times. But eventually he got 3 bases and a mothership. Be careful. I've already explained how to deal with motherships, but you really need to respect it. It's okay to lose 5 infestors to get that NP+Vortex off, or snipe it with mass corruptor. It's not okay to lose a 1/3 of your broodlords to a vortex because you got too aggressive or didn't manage your infestor energy.
Don't Max on Roaches when you have an Advantage, Don't Remax on Roaches You just got your army crushed without making a dent in his army. Whatever you do, don't remake an army that is going to be slightly weaker or smaller than what was exactly just gotten crushed.
Or, you have a lead and total map control. Don't max out on roaches, it's very all-in. Your hive tech will be too late, and your max roach army will just be dead weight for 20 minutes because Toss secured his third, finally, and you can't just attack hard with them because if you lose them, you lose your map control. Be smart on remaxing. If you lost the game, you lost the game, but don't max out on roaches. They are a tier 1 unit. Robotech rapes roaches, and you can't break down a brick wall - eventually, Toss will be able to secure his next base. Just as you can't always defend a base and it may be smarter to just let the base go and keep your drones and units, you can't always deny a base. The game will be even, secure your lead without resorting to desperate play.
I get that max roach can bust a toss being too greedy. I'm not saying don't go mass roach as a strategy. What I'm saying, is if you have gained a lead, maxing roaches is the easiest way to throw it away and, remaxing on roaches after you failed to make a dent, is a terrible idea.
Deathballs This stuff is obnoxious, I know. But it simply doesn't work at the higher levels, largely because of macro. If you are losing to deathballs, a big reason may be because of your macro. Check on it. Maybe your macro is better than the opponent, but is your macro good enough is the real question. But hey, we all don't macro well... so. The ETA concept is also a large part of this - if you maxed on roaches, any sort of deathball play will crush you if you don't kill him before he gets 2+ colossi.
Mutas are a really good hard counter to deathball play. Seriously consider mutas if Toss is going for deathball play. If he opened stargate, it's okay - mutas are better than colossi more so than phoenixes are better than muta/corruptor. Just have a few corruptors in your muta control group, or if Toss has only 4 phoenixes, don't be scared to stomp it with your 15+ mutas. If he has colossi, he won't have many phoenixes.
If Toss is doing a 2 base deathball play, mutas really are the best answer (after using corruptors, or maybe hydras, to take down mass phoenix). If he is doing a 3 base turtle deathball, you can jack up your roach/ling army you made to deny the third long enough to get mutas out, or mass roach and just trade with his colossi right when the 1st or 2nd one pops and sentries, hard enough that you can fast track to broodlord tech.
But really, the 2 answers to deathball derping, is mutas, and macro'ing well (as in, having the econ to be able to mass roaches to stomp his 4 unit army).
Massive Stalker balls, Immortal/Sentry balls These armies need to be nipped in the bud early on. You can't just drone up, and then let him come to you and defend - you need to build up your army, and attack his army while it's small, or push back when it's still small. Eventually your rate of production will overwhelm such armies. If you are losing to this kind of play, chances are your macro is weak, or you waited far too long to make units.
Secondly, don't be afraid to go mutas (as long as you know for sure his initial pressure is over and he has to back off because your roach/ling/hydra army is large enough). Blink is okay, but he'll need storm eventually, and mutas will always beat stalkers in a base trade. As for immortals, they are costly. While they aren't as costly as colossi, they are costly. Mutas will do fine against them.
Toss may go with a build other than FFE. I will cover that here, from 1/3g expands to cheeses.
Against any sort of gateway opening though, you will want to get ling speed asap. Throw down a gas as soon as you realize he's going gateway instead of forge. You need speed to check his army, his expo timing (or lack thereof), and to defend any sort of pressure/all-in, and maintain map control to take your third against gateway expand. I recommend you remove all 3 drones off gas after speed (you want minerals to bank up for that third, and you can get enough roaches quickly by just putting 3 in gas when you place down warren).
As long as he makes a core instead of 2 gate, you will want to get drone up hard to 30-35, and check with a speedling to see if he expands, and an overlord into his main unless it's completely obvious with your ling that he's going 1/3g expand (pylon on low ground, signs he wants to take an expand, and 1 zealot and 2-3 sentries).
Now, we already know that Toss almost always goes FFE, so therefore we should always go Gasless Pool. But, sometimes you are playing a map where FFE is unheard of (Crossfire, Dual Sight), it's a Best Of X series and mindgames are going on (Nestea vs Naniwa MLG), or who the opponent is (LiquidNony is known for loving Gateway Expands over FFE).
In these circumstances, 14/14 is a smart strategic choice to make.
A fast chrono'd stalker after zealot (or no zealot made at all) by Toss (like if he scouts no gas when pool completed) can do a lot of pressure if you don't go 14/14. You will need a spine, and if the natural is wide open, a single spine alone isn't enough, but will need queen micro + slowlings. With speedlings, you can make sure no pressure like this occurs, or if it does, get a stalker for free and 1 less sentry for Toss was made.
Mainly though, Toss cannot go 4:30 Nexus (1 gate expand). If he does, you can make a round of reactionary lings to force a cancel, and make Toss go 3 gate Expand at 5:30 (it will be a little later due to him losing 100 minerals to cancel). A good Toss will not go 1 Gate Expand against 14/14, and will hold an infinite number of speedlings with 3 Gate Expand (even if you force a cancel, 100 minerals is not worth over 10+ less drones). But that's the point - you force Toss to go 3g Expand instead.
However, it's not a huge deal if you went gas after pool, but it seems that 14/14 is standard if you suspect Gateway Openers. So if you suspect Toss will open 1 base, go for 14/14.
You should see no pylon on the low ground at his natural, meaning he either opened gateway at his ramp (very rarely, sim-citied), or proxied buildings. Make sure his base isn't empty once you realize it's not FFE. I recommend drone scouting, especially to check for 2nd gas timing on gateway builds and to block the nexus first of FFE. If you don't believe in drone scouting in ZvP, or want to be greedy, that is fine.
If there is an empty base, it's clearly proxy 2 gate. If he's going Nexus First, he has to make a pylon on the lowground so make probes after 11 supply, and to eventually get his forge.
So now you realize he's doing a gateway opening, and the first unit is a zealot. With your drone scout, make sure he takes his second gas (or steal it after he makes his core to make sure he can't do any fast DT/SG or get lots of sentries - I recommend gas stealing gateway openings!). If he doesn't get his 2nd gas, especially if he is saving up chrono energy and is not chrono'ing his nexus, worry about a 4 gate (confirmed with your overlord scout later).
Understand that if he has 2 gas, that should translate to about 200 gas when his 2nd unit pops out. If it's a zealot, question where that gas is going. If the 3rd unit is not a sentry, alarm bells should be going off.
If he is chrono'ing his gateway hard, he may be doing zealot/stalker pressure. Make a spine in your main when nat is nearing completion, and bring it down to your nat. If the natural is wide open, be aware you will need to make some lings because without speed, a single stalker can torture you.
If he makes zealot/sentry/sentry, you can be assured he is going sentry expand from 1/3 gates. It's when he makes anything else that you should be asking serious questions and sac'ing an overlord....
Send that overlord in at 30-35 supply. Check to see if he took his 2nd gas (2nd gas taken immediately after core, is signs of DT/SG, he should be taking it around 21), and with that drone scout you should know where his pylons are, to see if he put down DT/SG play or 4 gate, robo, or just has 1/3 gates.
A 1gate expand, expands at 4:30. A 3 gate expand, expands at 5:30. The only reason a nexus isn't going down as late as 5:30, is because he's distracted by your ling and he's obviously making a very clear effort to throw down that nexus immediately, or he is all-inning. No nexus by 5:40 should be setting off red alarm bells, no nexus by 5:50 can't be anything but an all-in.
Finally: At 5:30, Toss with 2 gas has mined ~400 gas (a 3g expand will have 2-3 sentries and wg tech). So you need to ask at 5:30 - where is his 400 gas?
If you just overlord scout, it can be deceiving when you see an empty base with maybe no gas being taken, or no wall-in to the ramp at his main, and you haven't seen yet that he has no pylon on the low ground at his natural. While seeing no pylon on the low ground (after seeing he didn't wall-in at ramp) is a dead giveaway to double 10/10 proxy gate, on some maps you just may not see that in time. It's okay, your 14 pool build will be fine, even if you made a hatch at 16.
As soon as your pool pops, make 2 spines in 'uncomfortable', choked up locations like by the geysers, next to your pool that's by a wall/object, or in your mineral line. If you only make 1, Toss can kill it and you will be screwed. Also, grab your gas asap - get ling speed, and you will be able to crush any remaining forces and just run into his base and autowin. I strongly recommend against roaches - he can go for core, stalkers, and torture your slow roaches, especially if the gates aren't proxied.
You should really cancel your hatch if you planted one, but if he's being an idiot and attacks it, let him attack it and just cancel it at the absolute last second. Even if he kills it, you should be fine. However, you don't NEED to cancel it, you will have enough resources to make enough to hold and make mass speedlings and autowin, and if he kills it you can still end up ahead if you make those 2 spines and keep your drones alive and get ling speed. It's up to you - I'd say it's safer to cancel, it's greedier to keep it but really you'll be ahead whether he kills it or doesn't. Or, you could keep or cancel it based on if he attacks it or not.
If he does not proxy the 2 gate, you can definitely keep your hatchery. It will take damage, but you will uproot your spines and put them down in the nat. There are macro oriented 2 gate openings where they expand right afterwards - just make sure he expands (he won't have stalkers) instead of going core, gas, into 4 gate follow-up, and take your third and threaten with your lings and just be so far ahead of him...
You can either go ling/queen/spine gasless, or get gas and roaches. I would recommend you go roaches, unless the map is like Shakuras, where you have a relatively small ramp to the natural that makes static defense extremely strong (3 spines, an extra queen, no way a 4 gate will ever get up, and if he tries to rush past it, you make a ton of lings quickly and finish off his army).
The reason I recommend roach/ling as defense, is because eventually, you will defend this 4 gate. If you go gasless, meaning you go mostly with spines and queens (which isn't viable on maps that are super wide open, at higher levels Toss can just react by expanding if you put down the necessary amount of spines on the wider natural maps, and you will need a lot), you can't deny Toss from putting down his natural. With roach/ling, you can eventually counter-push and delay Toss' natural for so long, that by the time he gets it, you will have followed up by droning up to ~44 drones, gotten 3 gas lair, and roach speed, and just won with an unbeatable mass roach army.
An all-in that hits a little later than 4 gate, but harder, with immortals, sentries, and sometimes with a warp prism.
The key here, is knowing it's not a plain 4 gate, but a 4 gate robo. If you cut drone production at 30-35, you will have a much harder time defending. With that overlord scout, you will be looking for 4 gates, but a 2nd gas, and him obviously not pushing out for a proxy pylon with his first 3 units, maybe even the robo. But if you reacted thinking it was a plain 4 gate, or droned a bit more and didn't realize it was a 1 base all-in until you notice no expo by 5:40, it's okay.
So the first thing you want to do, besides drone up to 40, is queue up 2 more queens. You really want a good number of queens against this, to deal with the warp prism, and for transfuse and creep spread. Your primary concern at the start, is him dropping a sentry at the top of the ramp and FF'ing it, and then warping in reinforcements. You have the watch towers and a ling in front of his base, so there is no reason to keep your army at the front of your natural.
You will want to go heavier on lings, queens, and eventually spines. Making spines right away (except maybe a conservative single spine) isn't too great because of the warp prism and immortals, and the timing is biggest before your base is totally spread with creep to allow your queens to deny the warp prism.
As his harass continues and your ling army gets very large, you can start adding a few drones, and put back in gas and get a roach warren, and start adding lots of spines. Spread creep around your base to deny the warp prism. Your army should be primarily lings, but the goal is to survive for 50 drones, lair, and then mass speedroaches to push back and then deny his natural. It's a hard push to deal with, but you should have a couple more drones than against 4 gate, and go for queens right away, and speedlings, then spines, then roaches (roaches on their own are useless, and mass spines are only good after you've creeped up your base to deny warp prism and have lots of lings/queens to support them).
When you see he didn't expand with 3-4 units at 5:30, you need to be asking questions. This can be made obvious by a gas taken immediately after core as opposed to a little bit after core is taken (with intuition, you will just notice he took his gas a little too quickly), and no sentries made.
So you should be asking if DT/SG play is coming if you don't see P have 3 sentries at 5:30. Sometimes Toss will make a zealot/sentry/stalker, or do double stalker pressure at the start. That's fine, as long as he expanded at 4:30-5:30. Usually though, it's pretty obvious that DT/SG is coming - he only made zealots, there is only a zealot and stalker, there are no sentries. Sac'ing an overlord at around 5:30 when you don't see surefire signs of a 1/3g expand, and when you see he clearly isn't going with 3+ gates, you can figure out enough to know to make spores.
Also, DT/SG openers go with 2 gates usually.
Make that evo chamber at 5:30 when you realize something is fishy, get an extra queen. Put the extra queen on hold position at the ramp, and a single spore at the natural. You should have a handful of lings made - Toss didn't expand, you are ahead in econ right now, it's okay - and have them stay at the ramp too, so if he's going DT, you are blocking the ramp, and the spore in his natural will sight the DT and you can kill it off.
Whatever he's doing, will hit you at about 6:30. If it doesn't, he might be just making more air/dt's, so make some more queens, maybe a spore in your main. Remember, what he's doing doesn't quite look like a conventional 1/3 gate expand, and you know he isn't doing a gateway all-in (he may even take a 6:00 nexus), but unless you got a good overlord in, you are confused on what's going on! Don't get complacent - a 1 base opener that expands oddly at 6:00 and doesn't quite have 3-4 sentries taking his nexus, could just be a 1 base stargate that is massing a ton of void rays and phoenixes, that will rape you. If you are unsure of what's going on, and it's obviously not 1/3g expand, make more queens. Even if he's just an idiot doing a bad 6:00 3 gate expand, he's the one who is going to be further behind, not you.
If he is doing a DT/SG expand, you will want to spread creep rapidly to your third, and drone very hard. Against air play, you should be spreading creep and making 2-3 extra queens, and walk them over to protect the third until the creep gets there and you can put a spore down (some people like SG expand as the safest possible expand build, since void/sentry will stop any roach/ling all-in easily).
If he went DT expand, you can't get your third until you have creep all the way over to the third. It's okay, he is quite far behind in econ and army size. Drone hard to about 45 drones and get lair, and go for speedroaches on 2 base. Use an overseer to finally take that third, and use the speedroaches to prevent toss from doing anything but peeing his pants. You will deny any sort of goofy gateway/archon push (a very deadly follow up to DT expands, which will roll you if you simply don't mass enough speedroaches), and delay the third for a very, very long time, allowing you total control of the game.
If the natural is extremely wide open, where a single spine won't work against fast single stalker pressure, and Toss scouts you didn't take your gas up to when the lings popped out, Toss can just open 3 gate Sentry Expand, except instead of making sentries, he makes stalkers. This build usually starts with a quick stalker made after zealot, too or; this build is a great transition if Toss opens fast single stalker pressure after he probe scouted to see you didn't get gas before lings popping out.
Get your gas as soon as you see Toss went gateway instead of FFE, and avoid losing to this. An example of a tortuous death to this, is MC vs Nestea GSL S2 2012 Dual Sight. The stalker ball will just get larger and larger if you never took gas for speedlings, and if the natural is wide open and you don't make 2 spines right away (or 1 if it has a ramp), he'l harass for so long as his stalker ball gets larger, then warp gate finishes, and the you die.
I lump these together because they are pretty similar in how you play it out and they look similar until the moment Toss throws down his nexus. 1 Gate Expand is not viable if you opened 14/14, because you can react to the 4:30 nexus with a single round of 2xInject lings, and force a cancel and a late nexus made after going impromptu 3 gate expand. However, anything but FFE is rare these days, so gasless pool first is the 'better' choice in that based on the odds, it's the better build to open with, so Toss can see you didn't go Gas First, and respond by going 1 Gate Expand instead of 3 Gate Expand. If he sees you went Gas First, he can just go with 3 Gate Expand (a good Toss will know to react like this, but most aren't that smart).
If you did not go 14/14, don't make lings to attack (no matter if it's 1 gate or 3 gate expand). Against a good Toss, you will not do enough damage. Secondly, you are really hurting your econ. If Toss is going 3 gate expand, and you just mass speedlings from a 14/14, the best you can do is delay the Toss Nexus as his 3 gateways go online and he just has to wait for 1 warp-in. Not having a good 10 drones for a minute, hurts a lot more than a Nexus being delayed for a minute (a good Toss will not lose any of his units since he stays at the ramp and FF's quick enough). Also, if you do go 14/14 against a 1 Gate Expand, only make a single 2x round of lings - making anything more, will actually put you further behind as all you are doing is forcing Toss to go 3 gate expand with a nexus that is maybe 10 seconds later . Once 3 gateways are up, he WILL get his nexus up, and a cancel only costs 100 minerals, which is a lot less than not having a 2x injects worth of drones for a minute.
I repeat, anything but going 14/14 and making reactionary lings to 4:30 nexus, is either not going to work against a good Toss, or it won't do enough damage to justify it (however, against bad toss, it's awesome). Now, moving on.
That's the only way to tell. With a 3 gate expand, he will have that same 3-4 units, made from the single gateway. Now, moving on.
Build Order ~35 Third supply (some like 2 base lair and take third while getting speed, but this is my guide, this is how I see it). ~40 Roach Warren & back on gas (Depending on Toss zealot/sentry pressure and map size, you may either need to make 2 early overlords and bank for 7+ roaches, or drone up). ~ 55 2nd gas - Scout if Toss is teching or making lots of units. - Get Lair as soon as you safely, possibly can.
After you drone to 40, you will want to consider making roaches to defend. You can delay the roaches if it's a longer map, or you scout Toss teching rather than making more gateway units or going for pressure, but Toss can move out with his first warp-in of units and go straight to FF the ramp in your main, and it can really hurt if you didn't start roaches (having a single spine up by 50, to buy time for those roaches, can go a long way too), and hit you by about 50-55. Having a ling out front of his base here is critical, and being supply blocked from 40-68 can cost you the game. It's extremely important to have a ling in front of his base to see if he does this zealot/sentry move-out.
I'd also recommend making a round of lings after taking your third, for 3 reasons. If he pushes out, it can buy you critical time if you go run into his natural, to force him back to his base, when he's about 1/3rd across the map, to get those roaches out. Secondly, once you hold off his attack, and he starts retreating, if you only have roaches and no lings, he will be able to return home with every sentry intact, and not be punished at all for his push. Third, you can often discourage him from moving out at all if he sees that many lings out and about.
At about 50-60, you should try to overlord scout. If Toss has taken both gas at natural, or is getting a stargate, robo, or twilight, it means you can drone up to 65-70 before roach production, and react accordingly with quicker lair, an evo and/or queens & spores. If it's lots of gateways, you might want to get some roaches in before lair. You will need to make roaches around 70 supply at the latest, in order to get map control to delay Toss third or deny strong pushes from getting set up.
You will want Lair as soon as you safely can, 55 supply at the earliest (delay the lair if Toss did not take gas at natural, or Toss is clearly going to push out as he adds lots more sentry/zealot). If your lair is too late, Toss can get his third too easily as you won't have roach speed. Your goal here is to defend that third, and then tech up to lairtech as fast as possible.
It's really quite similar to FFE. The differences are just you'll need speed, in case of early pressure or if he doesn't expand, and you'll need to produce units after taking your third, in case he pushes out with sentry/zealot (after all, gateway expands do have the merit of having a larger number of units, early on).
From there, the game is up to you. If you can defend the sentry/zealot pressure, you should be wayyyy ahead of Toss. I almost always go for mutas, and I don't think I've lost to 1/3g expands in recent memory by going Third, Defend, Mutas. There is a reason many people consider this build outdated - you play basically the same as you are against FFE except get speed early on (slightly later lair is result of getting speed).
I think all of these all-ins are outdated by now, and Toss largely knows how to hold them.... In the lower levels I think you can always make these work though, and even at higher levels, you can work if Toss just goes one way as opposed to another or doesn't react perfectly. These are all of the all-ins I know of, there isn't really any all-ins that are standard in the metagame at the moment.
Many variations of 6-10 Pools that you can do in ZvP (it sounds to me that 7 Pool is most popular). While you can all-in with drones to try to end the game immediately, some variations will simply aim to autowin against Nexus First and force Forge First to reset into a defensive 3 gate expand. A very deadly, popular follow-up to Early Pools is 2 base Muta.
You are sacrificing kind of a lot of econ (later hatch or less drones, then less drones due to making an extra pair of lings), in the hopes that when you pool 3 larva at 15 supply, you run into Toss' base if he went Nexus First, and you do some damage. I think you need to kill at least 5, maybe 6+ probes, to make this worthwhile (I believe you need 5+ lings in the base in order to guarantee probe kills against a good Toss). A forge first should be able to get ahead if he reacts correctly. You are hurt econ-wise if Toss didn't go Nexus First, but not as far behind as with 6-10 pools, and you can still go for the macro game.
This is my preferred cheese against Toss, like when on close spawns. I would prefer this over the 3 roach rush, this can be strong in the lower levels. It doesn't work as well if Toss gets 2 gateways quickly. The reason I like this rush the most, is because you can 10 drone scout on maps with close spawn to see if you should do this or not, as the pool is placed on 13 instead of earlier, like it is with 7rr. You also get speed with it, unlike most quick 1 base roach all-ins. (I'm not sure how it works against FFE). http://wiki.teamliquid.net/starcraft2/3_Roach_Rush_Speedling_all-in_(vs._Protoss)
- Open 14/14, speed then bane nest, lings. Stream into FFE.
It takes 5 banelings to kill a cannon, and 5 for a pylon. Toss can beat this with good micro, or a 2nd cannon placed quick and well. The best part of this build is how well it works even on large maps, due to speedlings being quick, and how banelings are made, but it can be obvious if you went gas first, to which many Toss will make that 2nd cannon, knowing the possibility of all-ins. Unlike roach all-ins, this build doesn't care about ramps and wall-ins.
On maps that have a wider open natural, this is a very strong all-in. It's extremely strong on maps like XNC or Dual Sight if Toss dares to go FFE.
- Open Gas before hatch - 20 Roach Warren, Overlord, Pool Larva - Pull off gas and follow up with lings
This build can be executed with a hatch taken in front of the probe, and then cancelled before roach warren pops (which should be made at 22-23 instead of 20-21 to make up for making a hatch).
Toss can expect this if you have gas in front of the probe. It doesn't work if Toss gets 2+ cannons, which is why I don't like it and prefer...
35 Roach Warren Roach/Speedling All-in vs FFE/Gateway Expand + Show Spoiler +
This build can work great even if Toss makes 2+ cannons instead of just 1, and really stresses Toss' handling. While he can beat if he scouts and knows it's coming and throws down a ton of panic cannons, or opens quick single stargate or is good with a quick sentry and micro, it's a very strong build that, unlike the 20 roach warren, Toss must realize it's coming or he will lose.
~19 or earlier, gas, speed, 2x hatch&queen - 35, Roach Warren, 3 Overlords, save larva - 35-52, Roaches, remove from gas. - Speedling Reinforcements
On maps like XNC, it's near impossible to stop this when going FFE. Protoss needs to send out a zealot scout, in order to spot for this, but you can avoid the zealot, and avoid revealing you have speed by taking the watch towers and knowing where the zealot is coming from, which results in Toss having less to defend with. This build is also extremely strong against gateway expands. I would not recommend this on maps like Shakuras that have a tight ramp to the natural. This build works better than 20 roach warren on larger maps, and may even be better to do on large maps if you consider that Toss may feel safer on such maps.
Make sure to go for cannons, pylons, sentries, and focus zealots with the roaches so your speedlings can be the real damage dealers. The roaches are just to bust in.
- Single Queen only - Gas when Lings pop/probe leaves base - Normal Third Timing - Speed, Baneling Nest - 28+ Lings
This all-in abuses the fact that Toss rules out all-ins by scouting your third. It works less if Toss is less greedy and gets sentries right away, and if he can find out about you making so many zerglings, having speed so early on, mining gas before third, or seeing no drones at your natural. This build plays upon Toss' assumptions based on limited information, so do as much as you can to deny scouting (run lings around side path, morph banelings to the side, et cetera).
- Expand, no 2nd queen (or delay it), put overlord into position by Toss base - Quick 2 gas, put only 2 drones in 2nd geyser - Lair - Nydus, bank 100 gas for worm (speed is not gotten because worm is most important to get asap)
Unless Toss is vigilant about nydus, because it was obvious with 2 gas or no drones at nat, Toss should lose a base. If toss is good though, he will abandon his main completely, cannon up his natural (or even wall off the bottom of the ramp), and just play it as 1 base vs no drones. He can come out ahead, unfortunately, but if you can take out the natural first, where the cannons are located, and before a pylon is set up in main, you can win quite handedly.
This build works only when Toss uses his nexus as part of his wall-off (Antiga sometimes, Metal, Dual Sight)
- Expand, no 2nd queen (or delay it), put overlord into position in front of Toss base - Quick gas, remove after 100. - Lair - Spew creep with overlord right out of range, plant down 3 spines. The edge of the creep should just barely be in range of his buildings, so he sees it at the last second. Stay out of range of cannons as you work his buildings and nexus. - Drone up
A strong follow up after droning up, is hydras. He will be on 1 base and quite far behind if you work him down well. Hopefully, Toss doesn't realize what's going on and make lots of cannons right away. You have to be able to kill his nexus, or it won't work. If Toss makes a wall-in without the nexus, you can do damage, but all Toss has to do is just rebuild his cyber, forge, and gate further back, with cannons up front so you can't move your spines up to the new wall.
Pretty straightforward - get lair going ~35 supply, hydra den, range. You should keep droning all the way up to until the hydra den is done. You can go for either nydus and nydus into his base if you are ballsy, or you can nydus out front and bust his wall with the hydras and even add spines with the creep spread, or you can go with drops. This build is best executed against FFE stargate. Reinforce with lings, put in a queen for creep spread, and maybe even add spines.
This doesn't really work well if he doesn't open stargate, or opens stargate and makes sure to deny nydus, or cuts stargate units quickly for robotech and stays alive long enough with micro and sentries and cannons for colossi. If he goes gateway, he can do okay with micro.
On some maps, 3 base isn't an option, and/or the map has a lot of airspace and distance, making 2 base muta a decent macro oriented build as long as Toss doesn't counter it hard with a build like aggressive gateway/blink play or stargate with lots of phoenixes.
~35 2xgas ~45 lair, 2x gas - Speed, spire - Secure Third after mutas
Toss really should be paralyzed in his base, he won't be able to push as the mutas come before, say, his 6 gateways come online. Eventually he may try to push - be ready with mass spines and lots of lings. If he's stupidly aggressive, you can actually win 2 base vs 2 base by trapping his entire army in the spines with your lings, and your mutas adding damage. If he just defends, or doesn't run into the mass spine, you will have to keep adding mutas, and double expand. You really need to drone up hard after the mutas come out.
If you go lair too quickly, your econ will be dead. If you go lair too late, you will miss the timing for when 2 base muta can do a lot of damage. I find starting lair in the 30's is way too quick, but in the 40s, just fine. As long as Toss doesn't open with a mass gate or blink stalker all-in, you can really do a lot of damage. Hopefully, it isn't too predictable when Toss realizes you didn't take your third. Also, you don't really need to scout what Toss is doing, once you go lair at a fairly normal time rather than super early, you are committed, and you will hit before Toss' tech really comes out. Losing overlords to scout is not worth it, as you will just miss your timing, or shorten the timing you can do a lot of damage. Making an overseer is nothing short of stupid - you just reveal that Hey I have a lair right now, but I'm not doing a super fast roach/ling all-in, so therefore I must be going mutas!
2 base infestor timings can be great, especially if Toss tries to open with aggressive gateway timings, as well as against dedicated stargate play (the things 2 base muta are worst against). This build is better in cases where mutas are worse, while still putting on pressure against Toss. You will definitely be behind if you don't do damage (like if Toss does not dedicate much to his stargate, or even worse, he goes robo or macro oriented play), just like 2 base muta. However, if your attack fails to do damage, you can hope to stay in the game by having a huge tech advantage, and fast tracking to broodlords because you never made roaches (or, some really strong roach aggression with infestor support).
~35 2xgas ~45 lair, 2x gas - Infestation Pit, Speed, Pathogen Glands - Secure Third after Infestors
Your goal is to get a ton of infestors out quickly, and spam IT all over his wall in to decimate sentries, cannons, take down the wall, and then stream in speedlings. You don't make roaches for this attack.
The good part of this build, is if you don't outright kill Toss, you have strong infestors to keep you alive so you can secure your third. Toss can end up ahead and go right towards colossi, and just steamroll you with 2+ Colossi before you can get the necessary 3 bases to handle 2 base Toss, but with your infestors, you should be able to handle any sort of gateway play if Toss goes for mass gateway, blink, or stargate, because FG+Lings will decimate anything without colossi long enough for you to get your third up and hopefully enough roaches to withstand a smaller colossi army, and then you can move towards a quicker hive.
Fairly straightforward. Follow my guide on Fast Third vs FFE, and at 8:00-9:00 cut drone production and just pump pure +1 speed roaches on 3 gas (eventually, 4 gas) and 4 hatches 4 queens 3 base. You will still need to defend whatever pressure or opener Toss does, but with decent macro and a macro hatch, you should be able to hit 200/200 before the 12:00 mark. If you are hitting it after 12:00, or after 13:00 even, you are doing it wrong. The merit of this, is that any deficiencies in macro you may have are blanketed, and it largely looks the same no matter who is executing it! Don't do this on maps like Shakuras with a ramp to the natural, and you'll lose if Toss is conservative and doesn't lose his entire army stupidly, and gets 2+ colossi before taking his third.
There was a while this was considered a soft counter to FFE, but most Toss now know just make sentries and get a robo bay with immortals, and eventually colossi.
Scout to see if Zerg went Hatch First, or Gas First. If Zerg went Gas First, that's definitely a 2 base play (if not always an all-in). Unless you can 100% scout what is going on, you should make at least 1 extra cannon (and if no all-in yet, make add another as time passes and the possible all-in gets stronger) - you can always take comfort that Zerg hurt his econ more than you hurt yours. If he went Hatch First, cannon rush it. Learn to cannon rush well - it actually takes a lot of skill to cannon rush well, but eventually no one should be able to survive with Hatch First (if you opened Gateway, add a forge, then cannon rush with a zealot).
If Zerg went gasless, block the expansion, throw down a pylon if you can, and go back to see if Z took his gas or not by the time pool popped, or is close to finishing, before scurrying away. A gas before third is an all-in.
Personally, I feel Toss should just always 9 probe scout, and Nexus First if Zerg made a 14 Pool or later. A 14 Pool goes down at 2:00-2:15, so generally a 9 scout will arrive before it is planted. You should be able to make nexus, forge, cannon, maybe even gate before cannon, against 14 pool, but yet so many Toss make a cannon so early, or don't go Nexus First. I don't know what's up with certain GSL casters - it's not 'risky' to get a Nexus First or gate before cannon against 14 pool.
If you went 1/3g Expand, you should do pressure if Zerg took his third, and stay inside your base if Zerg didn't, with a scout sent out around 5:30 (when you take nat with 3g). You may not end the game, but you can force a lot of roaches, and then just recall back home if Z has a lot of roaches (oh wait... not yet). If Z seems to have roaches, just back off. Don't wait until you have zero energy to retreat.
If you went FFE, there isn't really much you can do to deny an overlord scout. However, you can make it costly. Spread your pylons out, spread your buildings out. Don't make it so a single overlord can see everything - you want to force Zerg to have to lose at least 2 overlords, and maybe even an overseer later on, to get a full scout off. If you plan to go DT or 2xSG, I'd recommend proxying it. The hints to scout it are pretty obvious with just an overlord by natural and your lack of army, but if you proxy it, you may catch zerg off guard (which is the whole point of those builds in the first place).
Understand the role of the game as well - if Zerg doesn't have a third, defend. If he does, you can attack, tech hard, or try to expand yourself so that you are forcing Zerg to pick a path in the same fork in the road that you did. From what I understand, standard third timing is 9:00+.
Also, during any pressure, killing a queen really screws me over. It's like a 1/3rd of my production for each queen when going 3 hatch. It's a huge deal, and I've lost many games because I was down drones and couldn't afford enough army later on, or couldn't match production with a hard attack.
On a side note, never FFE on Metalopolis. I always just take the gold as my natural or third (with both 2 and 3 base play), and you will never keep up since I can connect the creep quickly, and there isn't really any pressure you can do that wouldn't be any stronger against the Gold as it would against a regular third. Really, just veto that map, it's horrible for PvZ.
Zerg basically has 3 options - muta, roach, or infestor. Respectively, these styles are harass, aggression, or turtling. If you see Zerg take 6 gas before 10:00, that can indicate that it's not mass roach, but he can put his gases at his third if he wants (unlike Toss FFE has to get first 2 gas in main), and it doesn't tell you infestor vs muta. The only way to tell if he's going mass roach or not, is if you see Zerg still making roaches after he's defended any sort of aggression - ie he makes more than the absolute bare minimum.
From there, deciding muta vs infestor is really based on intuition: - Is Zerg setting up his army to counterattack, or defensively? - Is he taking far away bases, or double expanding to nearby bases around the same area or route? - Is he spending his minerals on lings, or drones and spines? - Is he gradually adding spines, or does he throw down a ton early on and continues to mass them? - Does the map have a lot of airspace, or not? - Does the map have multiple attack routes, or a single path (Wide open maps can be good for either)? - Can Zerg possibly pressure the third, or is it ridiculously easy for Toss to secure? - Is the map large, or is it small and forces a lot of spines anyways?
Note, you cannot just check his upgrades. Even if he's going mass roach, the majority of the time, he will not have 2 upgrades done by 12:00, and he won't even have a single upgrade done by 10:00 (upgrade takes 2 minutes when you don't have chrono, and +2 attack can be preferable to 1/1 for roaches). Some Zergs like missile no matter what tech route they plan for, because of it's huge marginal benefit (+2 damage per upgrade for roaches), or melee because they plan to eventually get to broodlords and it's amazing for broodlord damage (since broodlings add the most damage). Many Zergs don't even decide their tech route until the exact moment comes to throw down the tech - I know personally I always go +1 missile, and I base muta vs infestor vs roach on how the game is going.
Roach Made pretty obvious when Zerg makes more roaches than the bare minimum to defend (like your observer by his base seeing constant roaches rallied). It's okay if your third is later, because Zerg is delaying his hive and infestors significantly, as well as his macro - it's pretty much an all-in to go 200/200 roach. Just don't leave your army in the open - as long as your army is never completely out in the open, you should be fine. 200/200 roach will not bust up a ramp of sentries and immortals. I would not recommend Colossi - it's an overreaction, Zerg can jush tech switch to mutas, they cost way too much for you to afford on 2 base, and immortals do the same job for cheaper (making 2 base colossi no matter what is going on is horrible).
However, there are other types of Roach. He could go roach/hydra, that's pretty well figured out now with just making Colossi. If Zerg goes with infestors, I'd recommend a doublerobo colossi after third, colossi will just wreck them (roach/infestor is different from infestor turtling, in that he makes a ton of roaches, is aggressive, and doesn't make spines). If he goes roach/banerain/infestor, double robo colossi is extremely strong, with of course blink micro. If you get storm tech, you can just FF, run away, and storm so all the overlords just go into the red if he gives chase.
All variations of going roaches is about aggression, so you generally want to turtle, then push when he tries to get his hive (as long as you didn't die before then).
Muta First, don't make colossi until you know that Zerg is going either infestors, or a ton of roaches. If you are worried about roaches, immortals will work enough for you to secure your third against anything but an all-in max roach. Second, don't make HT on 2 base, you just can't afford it, and you'll get rolled by mass mutas in the ~4 minutes it take to get HT. Just get stalkers, defense upgrades, and blink, calmly. Many Toss pros add cannons in the mineral line of their main after zerg regains map control after the stage when Toss normally does aggression, so they can keep their stalkers in their natural in case of roaches or mutas in the natural, and buy enough time for their army to respond.
Also, don't think that going phoenix or blink will rule out muta play. Zerg can go hydras into mutas, or get corruptors, against phoenix, or simply not move out with their mutas until they get 15+, and blink won't do anything to save your base if you leave it (so mutas will still be effective in making sure you never push out until he wants you to). Phoenix are a great way to scout for possible muta play though.
I don't know if phoenixes are a great way to counter or discourage muta play. If a Zerg wants his mutas, he'll make corruptors, attack with a small amount of hydras with lots of roaches to force you to fight with them (and then he focuses them with hydras), or he can just go infestors and capitalize on you delaying your 3 base colossi timing so much. I know a lot of people say phoenixes are crap against mutas. I would just say, if phoenixes are possible, it's harder to use them to counter muta play than to go stalker/ht (I've seen HerO rip it up with well timed phoenix into third and fend off mutas with them amazingly), and that I've never had a problem with them. Stephano has also commented about how crappy the range upgrade is, since it's so expensive and requires such a large investment into phoenixes, that you really need to delay your third or colossi that Zerg just gets infestors and quicker hive.
Infestor The quicker you find out, the quicker you can be greedy and tech harder, and even take a quicker fourth. A problem with infestor play is how passive it is, so you can always force the game into extreme lategame with a mothership, and try to negate any disadvantage by time and a well placed vortex. If you can read the Zerg well, you can bust with 4+ colossi before he has mass corruptors with mass infestors behind it. I'd recommend once you start to get a larger army, maybe with 2+ colossi, that you constantly poke and prod around, to try to find a place where the spines are a little weaker. Maybe you can also catch him up-rooting too many spines at once, and kill enough of them when they are re-rooting that you can bust through. Once zerg starts morphing those corruptors though, back off until you can get mothership/archon. Warp prism harass can also go a long way if Z has no AA, especially no air units when he still doesn't have his spire and is making infestors.
You want to bust Zerg with 3+ Colossi, with 3 bases, before Zerg has hive tech. If you defend his aggression/harass well or push before his turtling got BL tech, you should be able to hit and do significant damage. If you are slightly behind, you may want to go for a 4th as well as tech (mothership, storm) while Zerg is getting broodlords instead of trying to push against broodlords, who will just push you back till you die if you are only on 3 bases and don't have mothership/storm.
Hey Belial88, who are you? I'm a Mid-Masters Zerg on NA who's trying to climb up the ladder. I actually try hard to not watch VODs, reddit, tournaments, or TL, because I've found if you really want to improve as a player, cutting down on this stuff and instead really putting forth an effort to play as much as you possibly can (oh you have 20 minutes before work? You're tired from work and want to surf the internet for a few minutes before bed? the girlfriend is still getting ready to go out? PLAY, PLAY, PLAY!). However, I work over 60 hours a week, have a girlfriend, and a little bit of a life, which often leaves me in a place where I can't play, but I can post.
I've already graduated university with a degree in Economics (you can't be an Econ major and not be a conservative - although I'm a proud Republican, I'd definitely say I'm libertarian [it's the Grand Ole Party because we have room for everyone] and an atheist [however I love christianity, I think it's a force of good and protestant countries always turn out well), love ice cream, and kind of a bit of a jerk, but in a fun way. I've traveled a lot, and definitely spent a lot of time in bad places, from living in the ghetto and seeing people die (I'm not claiming to be particularly poor, just where the really nice apartments for cheap were - jk, sort of), to having traveled to third world countries (I don't think you can hate capitalism, or really respect all cultures, when you've been to certain parts of Africa).
I don't think anyone dislikes me after they talk to me - I really take things impersonally, and never find anything offensive. However, my comments aren't always taken with the same attitude, so naturally a lot of people dislike me. That, and I am a lower level player, so naturally I say a lot of stuff that is just plain wrong, but I'm humble enough, don't take myself seriously at all, and completely aware that I'm a scrub, that I actually want to be called out, so I can improve.
I'm sure I come off as disrespectful, but unless I say a blatant ad-hom, I don't mean to be. I'm the kind of person to always ask why, no matter who it is that at the moment, I don't understand. I'm never satisfied with an answer without a reason (which is why my posts tend to be long-winded, I answer things as I would want them explained).
Hey Belial88, why did you write this awesome guide? Too much shitty content on TL Strat. In particular, a 'comprehensive guide to zvp fast third' was posted by swooozy, someone who is now permabanned. I called him out for having a completely terrible guide that had a lot of misinformation in it, and although most people agreed, even with the rough tone I used, some people actually thanked him, as if bad content was better than no content. Also, I tend to basically write guides as responses to people's questions, and if my posts were put together, they could be organized guides themselves.
So now I can just link this guide to answer questions (even advanced ones, or ones that are complex). I had always planned to write one, but the swoooozy guide pushed me to actually do it.
Hey Belial88, how long did the guide take? As it was a guide to Fast Third only, maybe 5 or 6 hours? Editing afterwards, another 3 or so hours. The big update that just happened, maybe 8. Replay/Vod section, which is arguably the most important part of a guide, is generally the longest, and also why I take so long to come out with new guides. I constantly save vods/replays that I think have value in a future guide.
Hey Belial88, I'm just starting out, any tips? Focus on your macro. When you are below Masters, you don't really realize how horrible your macro is, so you may want a higher level person to analyze it for you, and you can never underestimate how important it is, because if you just macro 'okay', not good, but 'okay (avoid ever being supply blocked, don't bank 400+ money ever, never have a lapse of more than a second in idle larva or worker/unit production), and understand basic scouting (did he expand or not, does he have a lot of basic army/macro or could he be teching?), you would just simply skyrocket to Diamond (with a slight improvement, maybe a minor understanding of the game as anyone would have by diamond, then you would be masters soon enough). All of the goofy, late strats and attacks you see below masters, all become unviable as good macro simply hard counters them (3 void rays at 11:00? u leik 170/200 roaches?).
When you are above Masters, you realize how shitty your macro is. Especially if you play against a high masters or GM - you really don't appreciate how good a pro player is until you play one once you are at a level where you can execute basic macro. The efficiency they make a single unit have, is ridiculous (below masters, it's unanimous that people just hurt themselves every time they macro or attack because of lack of multi-task, let alone macro).
Just... watch some basic pro games, and try to make your games look like that. There is something wrong with you, if you are playing a TvP, and you are going 3 rax with tech/tech/reactor, or if you are zerg and grab more than 1 gas before 45 supply, or you are Toss and do any opener besides 1/3gate expand or FFE (besides all-ins). If your game doesn't resemble what you see pros do today, something is wrong. You can at least execute a similar build that takes gas and expansions and production buildings at similar times.
Hey Belial88, where do you plan to do next? Probably a ZvT guide, then ZvZ guide. ZvT is very straightforward to me, I feel a guide on that would be really quick and easy to do (it's by far my best match-up, but I win it so often against people on ladder I rarely think about it, if that makes sense), and there are still some kinks I'm trying to sort out in regards to strategy in ZvZ. I may also start streaming soon, but the schedule would not be regular at all, and I just moved in and got internet that has more than .6mb up (it's about 3.8 with this xfinity).
Hey Belial88, anything else? Thanks NrGmonk to changing title. Thanks to UCSD SC2 for letting me know how shitty I am, I was quite unaware. I'll be high masters soon enough. You can post this on your facebook page too you know. Anyone else can copy this too, as long as you give credit to Belial88@TL. If you want to contact me, PM me. Don't message me on B.net except to say how much you like me. If you want to thank me, for anything I do, do it in public so everyone sees how awesome you think I am.
Myself vs Ladder Cannon Rush Note that I click on every warping building to see what it is, and I very deliberately, diligently, and slowly, put 4+ drones on each building. Even when lings pop, I still use drones because lings alone won't do enough if Toss dedicates. He follows with a 1 base all-in, note I get evo/warren ~35 to deal with 1 base.
Myself vs Ladder Proxy 2 Gate Despite low ground pylon and probe 'trying' to take nexus, I knew what was up when nothing went down by the time pool went up. I didn't need to cancel nat, I just make 2 spines, 2 queens, pump lings, speed.
Myself vs Ladder Gas'd Third vs 3 Gate Expand I got a quick 3 bases up, and the muta follow-up was too much
Myself vs Ladder Gas'd Third vs 3 Gate Expand, Mutas in a base trade
Myself vs Ladder Gas'd Third vs 3 Gate Expand into Double Stargate You can see I have no idea what he's doing but am still safe from everything, and the roach/ling I made in case of gateway all-in allows me to pressure him and win
Myself vs Ladder 4 gate +1 8:00 Timing Note the overlord sac to make sure it's not 6/7/8 gate, and me starting lair and droning/teching hard throughout, to make sure I end up far ahead, and just making ~5 roaches.
DRG vs Squirtle GSTL S1 Ro4 Set 1 4 gate +1 8:00 timing Although DRG loses in the end, he shows a picture perfect hold against +1 4 gate pressure
Myself vs Ladder 6 gate +1 All-in Notice despite 'knowing' I need a 6:30 Warren due to no gas yet, I still sac an overlord at 7:00 to see if it's a 4gate, in which case I'd need to drone and tech up after 5+ Roaches, or if it's a 6 gate, where I need to go all units. I see lots of gateways, so I don't even make an evo, lair, or 4th gas, and pump units. I hold easily. I also saw heavy chrono on his cyber, forge, low probe count, so I knew I needed early roaches Worst case scenario is I may overmake units against a 4 gate +1, so I only made a few roaches, and lots of lings, so I could hold the initial pressure, and then make units/drones/tech accordingly.
Tech Heavy Harass (3-4 Gas, Few or No Gateways)
Myself vs Ladder DT Rush I saw a 3rd gas, no stargate, and a twilight council that wasn't quickly blinking. It could have been blink, but I did NOT see 7 gateways indicating a blink all-in. Generally, fast third gets a robo, so I didn't think he'd expand. I didn't see 7 gateways in total, maybe they were somewhere else, but I made sure to get ready for DTs by just making 2 overseers in my base, and for blink by making roach/ling
Myself vs Ladder Immortal/Sentry all-in using Roach/ling It's as close as it was because I lose too many lings trying to deny pylons. Note I am ling heavy, and strive to flank. I was able to scout the build by seeing late expo, army movement, and early on I saw the observer sliver moving overlord toward main, but I saw the chrono and I could have seen the robo if I didn't see observer sliver.
Myself vs Ladder Immortal/Sentry using Muta Base Trade I massed spines in my natural when he finally pushes to make sure he goes to third first.
Myself vs Ladder 2 Base Deathball I made lots of roaches and was simply able to overwhelm his small army, and then the muta follow-up was simply too much for him
Macro Games
Myself vs Ladder Mutas into Base Trade I always go mutas if I see Toss get Colossi on 2 base
Myself vs Ladder Extreme Lategame ZvP, Dealing with Mothership Note when seeing the quick third, I completely skip roach tech, and go Hive, Mass Spines. I put more and more spines down as the game goes on, and have 3/3/3. During engagements, I Corrupt/Focus the Mothership, and lay down IT right in front of Broodlords to protect them from Stalkers and Mothership, despite storm + Colossi.
Myself vs Ladder Extreme Lategame ZvP, Dealing with Mothership and Speed Prism Harass I get a Roach Warren due to late third, but he takes one eventually so I get Hive. He does a 3+ Colossus 3 base push, but I hold with IT Spam, Spines, and Corruptors. I go double spire, 3/3/3, 3/3, and Mass spines to deal with SpeedPrism harass. The 6+ Corruptors force a bad vortex, and I win then and there.
Extremely well put together, good information, easy to take in. While it is not incredibly in depth on any single subject it gives a great overview of how to play this style as a whole. One of the better guides I have seen on tl, thanks!
Great guide. I've been using something similar with slightly different timings (Tang made a stream lesson about early third->roach style) and it's an extremely strong style vs FFE. I've gone back to using 2base muta because I'm a scrub and like easy wins, but I'd say this style is definitely the strongest vs Toss if you're good and I believe that if a zerg plays it well, it can't really be stopped by a toss no matter what unless they somehow get to mothership/archon.
On March 12 2012 18:01 dream-_- wrote: Extremely well put together, good information, easy to take in. While it is not incredibly in depth on any single subject it gives a great overview of how to play this style as a whole. One of the better guides I have seen on tl, thanks!
The guide is just meant for the basic opener, so by nature it's a basic guide, but if you have any suggestions to make it more in-depth, let me know. Thanks though!
On March 12 2012 18:04 Tobberoth wrote: Great guide. I've been using something similar with slightly different timings (Tang made a stream lesson about early third->roach style) and it's an extremely strong style vs FFE. I've gone back to using 2base muta because I'm a scrub and like easy wins, but I'd say this style is definitely the strongest vs Toss if you're good and I believe that if a zerg plays it well, it can't really be stopped by a toss no matter what unless they somehow get to mothership/archon.
You're welcome.
2 base muta can be good on TDA, but I veto that map because I hate being restricted in choice, and toss can very easily counter it if they know what's up (which they know basically automatically on TDA, they just go mass blink stalker anyways, either you die if you went fast third, or it counter the mutas).
I agree though, I feel that fast third puts zerg at a lead and if they respond perfectly (and perfectly is pretty much impossible for humans i suppose), toss is behind.
mothership/archon can be dealt with. I don't think it's hard to deal with anymore - you get NP. Toss has to be very careful with their vortex and mothership usage, while zerg just needs to 'sac' one infestor and then instantly cast vortex. They just burrow around the mothership, get into position, and cast vortex on the toss army super fast. The best way to position the broodlords, I feel, is to stagger them, like siege tanks, so they don't get screwed by vortex too.
<3 you. Some things I disagree on, such as hydras to hold the sentry immortal push, I can hold that off mass roach/ling and a macro hatch, but for the most part, spot on and really really helpful, thank you.
Drone drilling in response to 3 pylon block is the worse thing you can do. I think taking your gas and going fast roach expand is the best thing you can do at that point. I also think you're a bit too focused on the 8:30 timing for the warpgate attack, I know many variations of 4/5/6/7 gate +1 timings that hit at all different times between 7:30 and 8:30 and require pretty much different answers if you want to play optimally.
Other than that, good guide that should help zergs discover the standard way of playing (rather than all the 2 base all-ins I have to face on ladder T.T)
Also, the OP would look much better with CENTER tags on the picture and subtitle
<3 you. Some things I disagree on, such as hydras to hold the sentry immortal push, I can hold that off mass roach/ling and a macro hatch, but for the most part, spot on and really really helpful, thank you.
I would love to see a replay where that happened, but I've seen multiple high masters and some pros say you need hydras,and roach/ling alone isn't enough.
Drone drilling in response to 3 pylon block is the worse thing you can do. I think taking your gas and going fast roach expand is the best thing you can do at that point.
You just don't let the pylons go down. Once they go down you're behind, and you're pretty screwed. There's little you can do to win against an opponent who doesn't make a blunder.
I also think you're a bit too focused on the 8:30 timing for the warpgate attack, I know many variations of 4/5/6/7 gate +1 timings that hit at all different times between 7:30 and 8:30 and require pretty much different answers if you want to play optimally.
It's a basic guide, but I'd love to see replays of this.
Roach warren should be taken at 7:30. You can take it earlier if you see no gas at the Toss natural. If Toss makes any sort of early zealot aggression, like double gate, you are better off using lings - don't make a roach warren earlier just to deal with such pressure. However, if you see lots of gas at Toss natural, or a Stargate opener, you can take your roach warren later.
I strongly disagree. 2 gate +1 zealot pressure outright kills a Zerg who doesn't get roaches as defence. 6 zealots with +1 attack hitting your third at 7 minutes can't be defended with just lings, and if you try, as long as the Protoss doesn't mess up micro and gets good positioning, you will lose your third base, depending on how many lings you suicide you could lose your natural and hence the game also. The only way you can defend the 2 gate +1 Zealot pressure with lings is if the Protoss is silly and lets you get surrounds on single zealots before he reaches a critical mass.
Could you elaborate on the overlords coming too early (or 2 at a time) costing you 3-4 drones? If you're on 3 hatches with ~10 lava coming off each (lava that goes towards supply), surely at some point you need to produce more than 1 overlord at a time?
I also find it tough to do spinecrawler pushes with broods etc on maps with big centers (antiga), and will often lose to maneuverable toss armies. Is there a different late game composition that is viable? Or do you just have to spread creep and push painfully slowly.
Your 8:00 benchmarks just made me feel really good about my macro because I'm regularly able to get 75 by 8:00, as that's the same exact benchmark I use.
Unfortunately my early game macro is the only decent part of my play or I wouldn't be mid masters.
Roach warren should be taken at 7:30. You can take it earlier if you see no gas at the Toss natural. If Toss makes any sort of early zealot aggression, like double gate, you are better off using lings - don't make a roach warren earlier just to deal with such pressure. However, if you see lots of gas at Toss natural, or a Stargate opener, you can take your roach warren later.
I strongly disagree. 2 gate +1 zealot pressure outright kills a Zerg who doesn't get roaches as defence. 6 zealots with +1 attack hitting your third at 7 minutes can't be defended with just lings, and if you try, as long as the Protoss doesn't mess up micro and gets good positioning, you will lose your third base, depending on how many lings you suicide you could lose your natural and hence the game also. The only way you can defend the 2 gate +1 Zealot pressure with lings is if the Protoss is silly and lets you get surrounds on single zealots before he reaches a critical mass.
You should see the zealots coming, what Tang recommends is putting down spines. Depending on map, you might be forced to put spines on both your nat and your third, but it's worth it since 6 zealots with +1 is a huge investment.
2 spines at nat and third is 400 minerals (add the drones), which will easily hold 6 zealots when you add the fact that there's a queen. A small strikeforce of lings added to that and he will do 0 damage. It's fine for you to invest more since the toss not only wasted 600 minerals on zealots, he also wasted minerals AND gas on the upgrade, and he wasted a ton of chronoboost on forge and gateways.
I strongly disagree. 2 gate +1 zealot pressure outright kills a Zerg who doesn't get roaches as defence. 6 zealots with +1 attack hitting your third at 7 minutes can't be defended with just lings, and if you try, as long as the Protoss doesn't mess up micro and gets good positioning, you will lose your third base, depending on how many lings you suicide you could lose your natural and hence the game also. The only way you can defend the 2 gate +1 Zealot pressure with lings is if the Protoss is silly and lets you get surrounds on single zealots before he reaches a critical mass.
I disagree, but okay.
Could you elaborate on the overlords coming too early (or 2 at a time) costing you 3-4 drones? If you're on 3 hatches with ~10 lava coming off each (lava that goes towards supply), surely at some point you need to produce more than 1 overlord at a time?
I meant that by making that overlord early, at 8:00 you will be X short than normal.
When you hit about 60 supply, that's when you need to have overlapping overlords (1 started while one is almost done).
Your 8:00 benchmarks just made me feel really good about my macro because I'm regularly able to get 75 by 8:00, as that's the same exact benchmark I use.
Then you are doing good!
I also find it tough to do spinecrawler pushes with broods etc on maps with big centers (antiga), and will often lose to maneuverable toss armies. Is there a different late game composition that is viable? Or do you just have to spread creep and push painfully slowly.
You should have changelings all over the map so you know where toss is moving. Mass stalkers can be a pain but eventually he'll have to fight you.
Could you give an overview of how your supply evolves (awkward 2 gate proxy zealot or similir shenanigans not withstanding) in order to have +60 supply at the 8.00? I'm talking drones, zerglings, queens, spawn larva and overlords.
It's pretty much the "safe" PvZ macro build if your macro and gamesense is spot on and you know when you need colossi quickly. I don't think there's a way for Zerg to get cheap wins (via all-ins or eco cheesing) against that unless they start using aggressive 2/3 base timings to shift the metagame. In other words prepare yourself for a real macro war.
Also i dont understand how you can attempt to deny his third on most maps other than crapolopolis (correct me if i'm wrong) with mass lings, i mean they're cheap, but just not very good against VR + zealot/sentry, and the sentry count doesn't have to be very high. it would basically be a waste of minerals or a waste of larvae.
Could you give an overview of how your supply evolves (awkward 2 gate proxy zealot or similir shenanigans not withstanding) in order to have +60 supply at the 8.00? I'm talking drones, zerglings, queens, spawn larva and overlords.
That would be great.
I specified it in the guide, but to put it all in one place:
14 pool 16-21 hatch, 4 lings Queens as necessary 25-30 Third 2 x gas ~ when third is done 7:00 evo 7:30 roach warren, 3 x overlord sacs 8:00 60+ supply Making overlords is just being super careful. Like 34/36 instead of 31-33. Even at the 46 supply you dont need to start 2 overlords at once. You go to like 50/54. Then after that you need to be very careful, because of injects, but start an overlord when the 1st one is nearing completion. At about 60 you are doing like 1 overlord and starting the next very soon afterwards.
do you have any indicator on +1 4gate or do you just take roach warren at 7:00 if toss hasnt taken any gas at the natural at that time?
I'm not sure what you are referring to. With a 7:30 roach warren, maybe a bit sooner if you see no gas, you should be able to delay with lings or a spine until roaches pop.
One question I do have is how you feel about spine crawlers in the early-mid game of this matchup. I feel like one spine at the third is a good investment if you know they have a zealot timing coming, but is it better to just make earlier roaches and drone a bit harder to make up for them?
do you have any indicator on +1 4gate or do you just take roach warren at 7:00 if toss hasnt taken any gas at the natural at that time?
I'm not sure what you are referring to. With a 7:30 roach warren, maybe a bit sooner if you see no gas, you should be able to delay with lings or a spine until roaches pop.
7:30 is much too late i think. that means your roaches pop at 8:50 and +1 4 gate hits at 7:45 with zealots two-shotting your lings.
you´ll need your RW at 7:00 vs no gas natural. even if you hold it with RW at 7:30 you will need SO many lings that could be drones, its more economic to get RW at 7:00 vs no gas natural.
Great guide here, your advice (especially on pro-level macro) has really helped me refine my ZvP.
Honestly, I haven't seen any other zerg poster who puts the effort in that you do, keep up the great work.
shadymmj - I feel those builds are outdated for a reason. Getting collossus before the 3rd is up (I'm not mistaken?) is inviting a responsive zerg to saturate 3 bases, get up roach / hydra / corrupter, and roll over you when you try to move out. I feel like I've been put in such a great position against Protoss who trust a few collossus more than economy.
One question I do have is how you feel about spine crawlers in the early-mid game of this matchup. I feel like one spine at the third is a good investment if you know they have a zealot timing coming, but is it better to just make earlier roaches and drone a bit harder to make up for them?
I don't think spines are a good idea. I never see pros do it either. There's just no reason to. Any zealot or stalker pressure, you can see from a mile away, and make lings to deal with, much more cost effectively.
I don't think roaches are a good idea either. You make roaches to deal with warp gate, not gateway.
7:30 is much too late i think. that means your roaches pop at 8:50 and +1 4 gate hits at 7:45 with zealots two-shotting your lings.
You should have a spine and lings to deal with such a push. You can throw down the roach warren at 7:00 to help. I know +1 zealots are a pain, but I think taking advantage of 3 hatch larva inject and pumping lings, with the support of a spine and queen, would be better.
I just don't see zergs ever put down a warren before 7:30, and putting it down earlier kind of hurts your econ. Even if you go roaches, your econ is still developing, so I don't think you can really get enough roaches out to deal with just 6 zealots. It's great to get some roaches, but you are primarily using lings, and then using them to delay long enough for the spine or roaches to come. I mean, 3 roaches just isn't going to be enough. You gotta use lings i think.
But going 7:00 roach warren to deal with no gas toss, I think that's perfectly reasonable.
Honestly, I haven't seen any other zerg poster who puts the effort in that you do, keep up the great work.
Thanks. I think most people think I just 'argue' or something. I just brush people the wrong way because I actually believe in what I say and am perfectly happy to be proven wrong, and that kind of confidence just seems to piss a lot of people off. I appreciate it though, I know you are much higher level than me so I'm glad you could find some use.
On March 12 2012 20:41 ThomasHobbes wrote: Great guide here, your advice (especially on pro-level macro) has really helped me refine my ZvP.
Honestly, I haven't seen any other zerg poster who puts the effort in that you do, keep up the great work.
shadymmj - I feel those builds are outdated for a reason. Getting collossus before the 3rd is up (I'm not mistaken?) is inviting a responsive zerg to saturate 3 bases, get up roach / hydra / corrupter, and roll over you when you try to move out. I feel like I've been put in such a great position against Protoss who trust a few collossus more than economy.
no no its 1sg into # gate into 3rd base then a push with colossi...just read the link. it's not outdated. it has been done by genius and at the latest european tournament, what's it called. and possibly earlier.
One question I do have is how you feel about spine crawlers in the early-mid game of this matchup. I feel like one spine at the third is a good investment if you know they have a zealot timing coming, but is it better to just make earlier roaches and drone a bit harder to make up for them?
I don't think spines are a good idea. I never see pros do it either. There's just no reason to. Any zealot or stalker pressure, you can see from a mile away, and make lings to deal with, much more cost effectively.
well actually Snute does it; and it seems to help him to drone a bit more. Obviously it's not mainstream, but I would consider it as a reasonable possibility.
Just a quick question; how does the 7:30 roach warren hold against Hero's quick 4 gate +1 attack pressure off two bases (meaning, is it still okay to build it that late if you do believe this to be incoming)? I feel like it's a little late considering no spine crawlers; I understand you wrote to build it sooner if you don't scout gas at the expo, but I'm still curious.
Also, do you consider creep spread to your third a priority? 3rd queen right away for tumor, or later? Injects first? 4 queens total, or 3?
well actually Snute [makes early spines]; and it seems to help him to drone a bit more. Obviously it's not mainstream, but I would consider it as a reasonable possibility.
Of course. I could definitely be wrong, and a spine for early pressure is great. I was thinking like spine as soon as third finished, to deal with like single stalker or zealot.
Just a quick question; how does the 7:30 roach warren hold against Hero's quick 4 gate +1 attack pressure off two bases (meaning, is it still okay to build it that late if you do believe this to be incoming)? I feel like it's a little late considering no spine crawlers; I understand you wrote to build it sooner if you don't scout gas at the expo, but I'm still curious.
This is just a general way to play fast third. It's a very general guide. I think your question is a bit beyond the OP, and it's not even 'my build', it's just sort of how to play fast third, and how it's generally done.
I kind of feel 7:30 roach warren is put at the time where, before roaches can come out, you can defend sufficiently with lings or queen/spine, to buy enough time. But making it a bit earlier to deal with pressure is fine.
I'm not saying don't ever make it earlier, or no spines, if you scout something that you push you to do so. But I think that 7:30 roach warren is quite early as it is, and most pros seem to always get it then or later.
If you could post a replay where someone holds such pressure successfully, and how they did it, maybe I can say more. But I'm no pro, and none of here are either. I think such play is well beyond this thread, if not this forum.
I don't consider creep spread to my third a priority, but if you do, go for it. There's nothing wrong with that. Getting your 3rd queen early or a 4th queen after 3rd is up to spread creep in important places (conneting natural and third on Korhal, spreading outward to deal with timings at vulnerable third on Daybreak, etc) is completely fine. I do think injects are very very important though (in zvt after double queen at start, i used to go inject/creep, but then i found out that inject/inject gave about 3 more drones, it was way better, and i could just spread creep with a later 3rd queen just fine), so I'd prefer getting your 3rd queen at natural and earlier so it can creep then go to third, or getting a fourth queen. but whatever.
You don't need to direct the questions to me. I'm not the expert here, this is just a very general guide, and anyone who is comfortable with going fast third can answer. It's not even a build order here, it's just sort of how to play it - like how to scout, what you are looking for, what you are trying to do.
Its good overview but tehre are 2 things that come to mind:
1. Things like hatch between X and Y doesn't help people: people need either a reason or an ingame solution I.E. I take my third hatch after i spend my first injects worth of larvae.
2. While this guide is a good overview the detail isn't broad enough to be comprehensive both in scope and depth.
Edit: My post seemed a little jerky and almost hypocritical with the lack of what i meant by not being comprehensive so hopefully this will be easier to understand.
You don't talk about how to hold of popular timing I.E. 7 gate blink +2 allins. I feel to be comprehensive you should descibe how to hold these builds of with something like. "After scouting the twilight and second upgrade you should begin making units (all gas going to roach rest ling) because otherwise if you don't have enough blink stalkers will snowball out of control. And maybe if a tip like " If the fight is at you third make roaches at your third hatch and the rest lings for fast reinforcement"
wow, this is how a guide should be done. Benchmarking and timings are the real important things here and are really going to help me and others improve. I don't think i've seen any other guides with this attention to detail on timings and scouting tells
It's a very solid guide. I prefer 15 pool to 14 though, that's just preference. For me this means that I will be able to put down my third hatch at 4.30 with a second queen in production and larvae popping soon after the third hatch is down and I'll be able to spend all my money on drones.
For those who are wondering when to get a macro hatch, if you've droned without any pressure (VERY SITUATIONAL) but a few zealots you should be dropping a macro hatch at 9.00 to 9.30 which will help you keep your money low if you are going roach ling. Though on some maps you might aswell take a faster forth like Stephano would (Shakuras).
I think a lot of players will benefit from using macro benchmarks, as you mention. However this section could definitely be expanded upon greatly. Though it is very important, how many drones you have at the 8minute mark is not the only benchmark you should set. When I'm teaching a student macro ZvP, I encourage using multiple benchmarks at the 5 minute, 6 minute, 7 minute, and 8 minute marks. You can count more than just drones, you can check..
For example, seeing that they aren't at the required 60-70 drones by 8minutes sometimes isn't enough information, but if you check to make sure you're at 52 supply by 7min, you can quickly pinpoint where your mistakes lie.
Thanks for the guide, Belial. Keep being confident but do try not to fall into being arrogant!
well actually Snute [makes early spines]; and it seems to help him to drone a bit more. Obviously it's not mainstream, but I would consider it as a reasonable possibility.
Of course. I could definitely be wrong, and a spine for early pressure is great. I was thinking like spine as soon as third finished, to deal with like single stalker or zealot.
Just a quick question; how does the 7:30 roach warren hold against Hero's quick 4 gate +1 attack pressure off two bases (meaning, is it still okay to build it that late if you do believe this to be incoming)? I feel like it's a little late considering no spine crawlers; I understand you wrote to build it sooner if you don't scout gas at the expo, but I'm still curious.
This is just a general way to play fast third. It's a very general guide. I think your question is a bit beyond the OP, and it's not even 'my build', it's just sort of how to play fast third, and how it's generally done.
I kind of feel 7:30 roach warren is put at the time where, before roaches can come out, you can defend sufficiently with lings or queen/spine, to buy enough time. But making it a bit earlier to deal with pressure is fine.
I'm not saying don't ever make it earlier, or no spines, if you scout something that you push you to do so. But I think that 7:30 roach warren is quite early as it is, and most pros seem to always get it then or later.
If you could post a replay where someone holds such pressure successfully, and how they did it, maybe I can say more. But I'm no pro, and none of here are either. I think such play is well beyond this thread, if not this forum.
I don't consider creep spread to my third a priority, but if you do, go for it. There's nothing wrong with that. Getting your 3rd queen early or a 4th queen after 3rd is up to spread creep in important places (conneting natural and third on Korhal, spreading outward to deal with timings at vulnerable third on Daybreak, etc) is completely fine. I do think injects are very very important though (in zvt after double queen at start, i used to go inject/creep, but then i found out that inject/inject gave about 3 more drones, it was way better, and i could just spread creep with a later 3rd queen just fine), so I'd prefer getting your 3rd queen at natural and earlier so it can creep then go to third, or getting a fourth queen. but whatever.
You don't need to direct the questions to me. I'm not the expert here, this is just a very general guide, and anyone who is comfortable with going fast third can answer. It's not even a build order here, it's just sort of how to play it - like how to scout, what you are looking for, what you are trying to do.
As someone who has abused the liquid'hero style +1 4gate zealot push for quite some time I will tell you that 7:30 is a bit too late to be able to defend. Spines and zerglings just dont hold against +1 zealots, and if your roaches pop out after the 8:50 mark, you are almost guaranteed to lose your 3rd.
There is also the issue that if the toss doesn't go with his first 4 zealots and instead waits for a wave of 8, you can easily be caught off guard with 8 zealots, and unless you were blindly making a lot of roaches the instant your rw finishes, you will be stuck trying to pop out enough roaches to suddenly stop 8 zealots before they can take down your 3rd, and ill tell you from personal experience, that wont happen.
However, having no gas at the expansion is not a solid tell, as I take my natural gas directly after I put down my gateways. I have been playing a lot of zerg lately as I believe that style fits me better, and when I zvp I always drop my roach warren at 7 minutes, no questions asked. Anything later than that and you have a strong chance of losing your 3rd to a well executed +1 zealot push.
Another, extremely well put together, easy to read, and awesomely helpful guide from Belial. As someone who has read both this guide and your guide against early pools, I can say with confidence that you are amazing at making guides. I am looking forward to your next post.
While the benchmarking part is really awesome, I think people should not overthink it. I'm just high platinum and vs very easy ai focusing only on macro, I can quite easily get to 75+ supply by the 8 min mark, it's not hard mechanically. This is when focusing on nothing else, it's important to remember that DRG and such do this while also scouting and reacting. That said, it's quite insane to feel the economy you get when you do it, keeping money low becomes an issue quite quick. I basically got to 8 minutes, started massing roaches like crazy, a-moved when I was maxed and checked stats when I won, my SQ was through the roof.
BTW, I have a question to you Belial: You cover scouting pretty well, it's very nice to see someone explain how to minimally scout everything you need to know. However, you never give any timings... When should I be on the lookout for the gases at the natural? On most maps, I can probably place the overlords so I have constant coverage, but even so, when should gases usually go down for the toss?
1. Things like hatch between X and Y doesn't help people: people need either a reason or an ingame solution I.E. I take my third hatch after i spend my first injects worth of larvae.
You are just taking it when you have enough drones. If you make it too early, it hurts too much, take it too late, your econ doesn't grow as much later on. I think saying 25-30 supply is pretty straightforward to everyone.
2. While this guide is a good overview the detail isn't broad enough to be comprehensive both in scope and depth.
It's just about the opener, but if you have any suggestions on how to be more in-depth, let me know.
You don't talk about how to hold of popular timing I.E. 7 gate blink +2 allins. I feel to be comprehensive you should descibe how to hold these builds of with something like. "After scouting the twilight and second upgrade you should begin making units (all gas going to roach rest ling) because otherwise if you don't have enough blink stalkers will snowball out of control. And maybe if a tip like " If the fight is at you third make roaches at your third hatch and the rest lings for fast reinforcement"
I do, and give replays. Pretty straightforward, get 60+ drones, get hydras. I say you scout by seeing 4 gas, mass gateways. As for your tip, I don't know how useful that is, but it's not something I've ever done.
It's a very solid guide. I prefer 15 pool to 14 though, that's just preference. For me this means that I will be able to put down my third hatch at 4.30 with a second queen in production and larvae popping soon after the third hatch is down and I'll be able to spend all my money on drones.
I would imagine 15 pool is better than 14 pool, actually.
I think a lot of players will benefit from using macro benchmarks, as you mention. However this section could definitely be expanded upon greatly. Though it is very important, how many drones you have at the 8minute mark is not the only benchmark you should set. When I'm teaching a student macro ZvP, I encourage using multiple benchmarks at the 5 minute, 6 minute, 7 minute, and 8 minute marks. You can count more than just drones, you can check..
For example, seeing that they aren't at the required 60-70 drones by 8minutes sometimes isn't enough information, but if you check to make sure you're at 52 supply by 7min, you can quickly pinpoint where your mistakes lie.
Thanks for the guide, Belial. Keep being confident but do try not to fall into being arrogant!
Sure, using more benchmarks could very much be useful. Personally, I just use the 8:00 minute benchmark. I think what's most important is how many drones you have, some players don't find creep as important, et cetera. For example a lot don't get creep until a 4th queen, made at third, and spread it from third on daybreak. Anyways, sure, what you say is useful.
These macro benchmarks are really helpful, have you outlined additional ones for other match-ups/builds?
I don't think there's as much disparity as fast third, but I have. You can do it for anything, but I think with fast third it's really important, and something that should be in your mind while your actually playing the game.
BTW, I have a question to you Belial: You cover scouting pretty well, it's very nice to see someone explain how to minimally scout everything you need to know. However, you never give any timings... When should I be on the lookout for the gases at the natural? On most maps, I can probably place the overlords so I have constant coverage, but even so, when should gases usually go down for the toss?
You just keep an overlord by the natural's gas. If you see Toss take those gases, that's the information right there. Timing of when he takes it, of course, can be useful, but you can usually figure it all out just by seeing that he took gas.
He'll be taking gas when you want to be scouting. So at 7:00, you are going to be asking "what is he doing?" At 10:00, you aren't going to be asking that question, so 4 gas then doesn't matter (he would'vew pushed by this point, obviously, you would have an overseer by then too). So it's just him taking those gas while you are still powering econ.
I also stated that at 7:00-7:30 is when you need to really scout and see what's up, so yea, seeing gas then, and sac'ing overlords, should tell you enough.
If you're looking to still refine and add things to your guide I would look around for vods of the different gas to ling relationships that occur: If you wait for around 45 supply to take all three gas, a +1 4gate will hit a very vulnerable time in your build, so it's worth noting that when delaying gas and therefor roach warren for so long, one should have enough lings to repel the 2 zealot 1 stalker 1 probe poke that is the beginning of the pressure and escort for the proxy pylon of a +1 4gate. Denying that pylon is absolutely key, it buys time for speed and raoch tech to kick in.
I guess my point is there is a relationship of number of lings built when scouting an early plus 1 in relationship to the point in time when you take your gas. And these lings are purely to deny the proxy pylon.
Very, very good guide. Extremely well written and organized. I have seen variations of this build almost exclusively in recent tournaments that I have watched (IPL TAC, Iron Squid, etc). It really helps to explain all the nuances I have seen.
I also read the Comprehensive Guide to Beating 6 Pool. I will be stalking you from now on because all your posts are awesome.
for as much as i seen you post, you appear to come off pretty douchy... very well written guide however. Been reworking my zvp and i agree with pretty much 100% of what iv seen. I may have some stylistic differences but in the fundamentals you got them. But still try to be a little less agro in general lol. You make good points but your soo angry sounding alot of the time
Just think your a bit pessimistic about the benchmarking, im diamond and generally get 70-75 by 8:00 without specifically trying, really don't get how people could get below 60 without serious supply blocks.
What I've been seeing as of late is some Pro's making a round of lings around 50 ish supply, which allows them to drone for longer as they can deny pylons, thus delaying the toss's build. How do you feel about this, and how many would you suggest making?
On March 13 2012 15:26 Iksf wrote: Just think your a bit pessimistic about the benchmarking, im diamond and generally get 70-75 by 8:00 without specifically trying, really don't get how people could get below 60 without serious supply blocks.
it happens pretty easily in a real game, if the toss does some early timing with zealots or something, the lings you get forced to make cut your drone count, you trade the lings for zealots, and end up at a supply deficit.
but with this level of production even a minor supply block (5 sec or so) hurts pretty bad
On March 13 2012 15:26 Iksf wrote: Just think your a bit pessimistic about the benchmarking, im diamond and generally get 70-75 by 8:00 without specifically trying, really don't get how people could get below 60 without serious supply blocks.
Without trying? Can you post a replay please, I'm vary curious. Here's a replay of me doing a very greedy and precise build by myself. At 8:00 I have: 1 larva, 65 drones, 2 lings, 9 overlords, 3 queens (Total: 78/78) 2 gas roach warren basically done +1 attack starting
This is under ideal conditions where the the hatcheries do not get blocked and I only make 2 lings early. I'd like to see other players post their builds so I can compare.
There seems to be a lot of negativity going on here, this is a really good guide that outlines extremely standard play at the highest levels of effectiveness. Good job Belial.
On March 13 2012 16:18 KhAmun wrote: There seems to be a lot of negativity going on here, this is a really good guide that outlines extremely standard play at the highest levels of effectiveness. Good job Belial.
Ya, this is a really nice guide. I've been doing pretty much this for a while, aside from the delayed lair. I like to get mine early so I can get roach speed and fast overseer scout but I'll consider delaying it since it may be costing me too many drones.
Great guide. Don't normally like your haughty know-it-all attitude, but this is a good contribution.
Regarding the +1 4 gate timing or even 4 gate timing(like MLG Winter Arena- DRG vs Naniwa G1), it is not likely that you will be able to hold your 3rd base with only lings. DRG place down a Roach Warren at 6:55, when the Naniwa started placing forward pylons aggressively. He still lost his third, but the good thing here was that he could morph another hatchery immediately, with the roach to hold off more pressure.
Another point to note here was that Naniwa was basically telling DRG straight up that he was gonna 4 gate pressure with the timing of the pylons, and we see DRG's reaction to it. My personal reasoning of why roach warren is the best answer - 1) Roaches can be very very cost effective against zealot, and on creep, equal to stalkers. 2) Roaches are very larva efficient, even with losing your 3rd hatch you can still produce enough to hold off a strong pressure like what Naniwa did. (Very low probe count). Additionally extra larvae = more drones when you hold off the pressure..
Edit : Hit the Post button before i finished my post.
On creep spread. I don't think it's necessary to build another queen to spread creep earlier on. As you say, DRG is one of the players with best macro and what he does is.
1st queen --> inject main hatch --> move to natural --> inject natural --> creep tumour
The reasoning behind it is probably that at that stage you can't spend all your larvae anyway. And an additional queen will cost 150 mineral + opportunity cost if you get it that early.
This is a very good guide, I was looking for something like this. I've seen others, too, though, and they basically disagree with your evo timing saying 7min is far too late and also the roach warren should go down at 7minutes at the latest (stephano rarely throws that down any later because you can never know when those few pesky zealots are coming, and zerglings really don't cut it being larva heavy and ineffective.).
Also what about spines? And, most importantly: you have three bases... so early spines to defend pressure seem to be inapproprite in any case because if you spine your third, he will go nat and vica versa, and you can't spine up both without being very seriously behind in the early stages.
This is a very good guide, I was looking for something like this. I've seen others, too, though, and they basically disagree with your evo timing saying 7min is far too late and also the roach warren should go down at 7minutes at the latest (stephano rarely throws that down any later because you can never know when those few pesky zealots are coming, and zerglings really don't cut it being larva heavy and ineffective.).
Also what about spines? And, most importantly: you have three bases... so early spines to defend pressure seem to be inapproprite in any case because if you spine your third, he will go nat and vica versa, and you can't spine up both without being very seriously behind in the early stages.
Wrong, you are already way ahead in workers. Wasting 3-4 workers to become safe from a zealot rush is an amazing trade, and you need no more since you have queens as well. It's also pretty easy to scout a zealot timing coming and you don't need to build your spines early. If you see commital to zealot+1 attack, put down 2 spines at nat and third, then just keep droning, your opponent just gave you extra time to outrun his economy completely.
If you wait for around 45 supply to take all three gas, a +1 4gate will hit a very vulnerable time in your build, so it's worth noting that when delaying gas and therefor roach warren for so long, one should have enough lings to repel the 2 zealot 1 stalker 1 probe poke that is the beginning of the pressure and escort for the proxy pylon of a +1 4gate. Denying that pylon is absolutely key, it buys time for speed and raoch tech to kick in.
I made it pretty clear it's up to you when you want to take gas. I just stated that the 'standard' gas is 2 gas taken when third finishes, so if you want to, you can definitely go 1 gas much earlier, the 2nd gas much later, or take 3 gas later like DRG, or just 2 gas a bit earlier, or a bit earlier.
You won't have speed out in time to deal with 1 zealot or 1 stalker pressure though, regardless, and if you were to get gas before third, that would just be too much. Getting 2 gas when third finishes around 6:00 should get you speed in time for 8:00 4 gate pressure though, or soon enough that slowlings or a spine or queen will buy that extra 10 more seconds for you.
If you want to get gas earlier, it's up to you, but there are serious ramifications for doing so, and there's a reason pros wait as long as when third is up for 2 gas, or even later than that. I mean, delaying gas as much as possible is kind of the idea, and you really don't need gas for a long time anyways (even if it's 4 gate pressure you are talking about, it's still relatively late).
And, this is a basic guide. You can tailor your third to what you scout. So if you see no gas taken, maybe you can get a single gas quicker to get speed out or fast roaches to deal with pressure and then just not take the 2nd gas for a super long time. For example, when I see doublegate before core, I throw down gas asap, like right when I plant third, if not sooner, so I can have speed against +1 double gate zealots (which you will crush with speedlings).
for as much as i seen you post, you appear to come off pretty douchy.
Just because I write good guides, does not mean I'm not a douchebag.
However, am I really a douchebag? Or do I just have a lot of confidence in what I say, even if I may be wrong? I believe every choice you make in this game should have a reason, and that there is a 'solid' way to play. Now, obviously, we aren't all robots, but this is a forum, and basically everything here is theorycraft, so I think assuming 'standard' play is the correct thing to do when discussing the game (or, explain why substandard choices may be okay, like making a ton of speedlings is a gamble in ZvZ, but maybe it works against the natural intuition of zerg players so it works better in practice than theory, but you should state that).
I state what I believe with full confidence, and from what I've observed. If I'm wrong, please, prove me wrong. I'd rather be a loudmouth saying incorrect shit, be proven wrong, and literally improve my game by 10% in 10 seconds, than be quiet and just keep doing the wrong shit over and over. That's how you improve as a player.
I mean, if I really thought everything I said was 100% correct, I just wouldn't post at all here. I post here so some of the times, someone calls me out, and then I learn from it. I know this may sound a bit douchey, but I don't learn too much from reading stuff here, but I learn a lot from being called out, or from having assumptions that I'm not sure of, and going back to test them because someone on these forums challenged something I thought was just common sense and turns out to be wrong. Obviously, I learn from people here, but most people who are higher level than me (I'm mid-masters, so I'm talking high-masters+) never ever post on TL because they would rather spend their time playing and view TL like we view the starcraft2.com forums, so most of what I learn here is just being called out, or hashing unsure things out because me and someone I'm arguing with aren't 100% sure.
That's just where I'm coming from.
But still try to be a little less agro in general lol.
I don't know what to say about that. I think 6 pooling is stupid. That's just my honest opinion on the matter, for example. I'm not going to change my honest opinion on the matter just because someone might take offense to it and would rather read "I feel that 6 pool is sub-optimal play, good sir". I mean, do you think "I don't agree with what he's doing" or "that guy was a fucking idiot" in your head?
But I am kind of an asshole. My girlfriend tells me that all the time.
You make good points but your soo angry sounding alot of the time
I'm actually not an angry guy at all. I never get offended, in real life, in work, and in this game, nothing really pisses me off. I just don't take things personally at all. But I have a conviction and passion about what I believe in. When I have a point of view, I'm pretty extreme about it. Why bother believing something if you aren't passionate about it? Like, you can't be christian and not be all about christ right? Well, I think certain things about the game, because I feel like I have strong evidence and reasoning behind it. And it's got me this far (mid-masters, and I don't play much at all too, I work 70+ hours a week). I must be doing something right.
Just think your a bit pessimistic about the benchmarking, im diamond and generally get 70-75 by 8:00 without specifically trying, really don't get how people could get below 60 without serious supply blocks.
Good job! Believe me, you are doing much better than most people in diamond. Now work on those other issues you have
What I've been seeing as of late is some Pro's making a round of lings around 50 ish supply, which allows them to drone for longer as they can deny pylons, thus delaying the toss's build. How do you feel about this, and how many would you suggest making?
That's interesting. I don't know. Provide a replay/vod where zerg did this and it benefited him.
Damn Zerg is unforgiving on the macro side. Thanks for the info, I will use it to improve mine, which definitely needs a lot of work.
I know right. How easy is it to just every 17 seconds, hit a button twice (for 2 bases, assuming an expand) and then every 29 seconds hit another button twice (chrono/mule).
While T/P is definitely delayed if they make a pylon or depot a bit early, in their build, they can still constantly make workers. Zerg, on the other hand, ahve to hit that button literally every second because of how larva spawns so irregularly, especially with larva inject. With T/P there is a constant rate of production that increases linearly, so it's easier to forecast when to throw down depo/pylons. For Zerg, it's exponential, so it's much harder to foretell when to make overlords, and it's literally anytime you bank more than 50 minerals, it means you aren't producing workers when you should be.
So much harder, in my opinion. I didn't realize how hard it was until I played with some GMs and they pointed out what a scrub I was because I never made workers. I was like... uhh.. I'm mid-masters and floating 90 average APM at the moment, what are you talking about? Sure enough, at like 30+ supply, I would go like 5-10 seconds where I just don't make workers, like constantly. It was weird when I first realized it.
There seems to be a lot of negativity going on here, this is a really good guide that outlines extremely standard play at the highest levels of effectiveness. Good job Belial.
Thanks! I expected a lot more criticism to come in regards to the scouting section, specifically the vagueness surrounding 3 gas.
Great guide. Don't normally like your haughty know-it-all attitude, but this is a good contribution.
My know it all attitude is not necessarily knowing it all. It's just how I see the game, and I have very strong convictions about how I see the game, because I feel every choice must have a reason behind it in this game. For Zerg, throwing down an evo chamber before 50 supply, for example, for no reason, is completely stupid and puts you behind, as harmless as it may seem. Of course, there may be reasons to put such an evo down, but have reasons, don't just put it down just to put it down. There's just a good an argument to not put it down, if you aren't sure what to do.
So I don't presume to know everything, not at all. I just have beliefs about everything. I have an opinion about everything in the game, from the zerg perspective (i claim total ignorance to the most boring match-up of all, TvP).
And I'm very much willing to be proven wrong. I've been proven wrong, and called out, plenty of times on this forum. Even my recent "Upgrades on Mutalisks" thread, I was called out, and proven wrong, and I happily admitted so.
And that's why I post here. I say my shit, I lay it out, and hopefully, someone can prove me wrong on something, and then I learn from it and become a better player. If I really thought I knew it all, I wouldn't bother posting here, because I 'know it all'. Get it?
I do understand my attitude may seem like I actually think I'm full of myself. But believe me, I'm not. I'm sure this may rub low masters or diamonds the wrong way, but as a mid-masters player, I think I'm a total fucking scrub. Like, I know I am. The difference between me and a high masters, is like diamond and bronze. Not even a well known high masters like tt1 could troll the fuck out of me with pure zealot all day, any 'random' top 8 masters could just make pure overseers and beat me 10 out of 10 games. I am completely aware of how shitty I am. Playing just a few games with high masters will convince you that right away if you aren't somewhat aware of it already.
Like I play with some high masters and GMs. They will offrace on races they never play, play me in a best of 5, tell me what build they are doing, and beat me. That's how different mid masters to high masters is. Hell, I even played ZvZs with some people, they told me they were going to 6 pool, I hold it off well ahead of them, I expand, and I still lose the game to them. While they are offracing, too.
So believe me when i say, I do not think I know everything. I just lay my shit out in the open, so people can see it, and call me out when I'm wrong, and then I can improve. I mean, how awesome is it that you can X league, be wrong on something, be called out on it, and then win like 1 game extra a day simply because you learn something you did was wrong. Pretty cool stuff.
This is a very good guide, I was looking for something like this. I've seen others, too, though, and they basically disagree with your evo timing saying 7min is far too late and also the roach warren should go down at 7minutes at the latest (stephano rarely throws that down any later because you can never know when those few pesky zealots are coming, and zerglings really don't cut it being larva heavy and ineffective.).
Also what about spines? And, most importantly: you have three bases... so early spines to defend pressure seem to be inapproprite in any case because if you spine your third, he will go nat and vica versa, and you can't spine up both without being very seriously behind in the early stages.
I just said what was standard and what I've observed most pros do. you can obviously change it if your play gets to a higher level, you scout something that prompts you to do so. This is a basic guide, after all.
However, I've never had spores come too late against any type of SG/DT play when going 7:00 evo (and I usually throw it a bit later than that, actually).
As for fast roach warren, I feel you should at least scout something to prompt a faster warren. 7:00 as standard seems too low.
Spines can be great, especially to buy time before your 'late' roach warren pops. I think spines are definitely amazing. But I thinhk they are best used to deal with warp gate pressure, not gateway pressure.
As for what you said about toss going around - toss units are slow. The spine isn't necessarily meant to defend, it's meant to buy time. So toss goes to third, sees a spine, has to back off, and then head all the way around to nat. By that time, you should have units, or a spine at nat coming. You are just buying time with spines.
For example, blink all-ins or sentry/immortal will get to a point where the army is large enough to smash through spines. But, getting 2 spines when the push arrives, is a great way to buy time for those hydras to pop out, to which you can push Toss back, hopefully all the way back home.
Just testing a practice game vs Very Easy AI (but with fake scouting, going 11 Overpool with 6 Lings AND screwing up a drone rallypoint when nat was up).
My problem with ZvP is scouting and understanding what I scout so I'll read this great guide of yours a couple of times and I should be fine .
Also I actually like your attitude, I don't mind when people have confidence in what they say instead of trying to be 'politically correct' all the time... hope you understand what I mean.
Just testing a practice game vs Very Easy AI (but with fake scouting, going 11 Overpool with 6 Lings AND screwing up a drone rallypoint when nat was up).
I'm at 69 at 8:00 with a warren and evo up.
Plat Zerg here. So... dunno.
Did you not read The Opening section at all? Why in the world would you go 11 overpool? It's a really, really shitty build. Also, 6 lings is wayyyyyyyyyyy too much. You should never, ever, ever make 6 lings. Like, ever, in ZvP. Except maybe against proxy doublegate, that's it.
If your warren is actually up at 8:00, maybe you are making that warren up a bit too early.
Also, were you ever supply blocked? Did you ever make an overlord, and when it finished, your supply was at or below where the cap was before the overlord popped (ie you made it too early)? Did you ever bank more than 100 gas before the 8:00 mark?
Try again with this advice, and if you are still having trouble, post the replay. But 69 for plat is definitely fine. Pros often only hit 65 in tournament games when the pressure is on.
But really, 11 overpool is the shittiest build ever. Don't ever do that build in ZvP. Really. Read The Opening section of the guide as to why. And making 6 lings is too much. You only make 4, and you are only making 4 because Protoss can throw down a proxy pylon and 2 lings+1drone won't kill it fast enough, you need 3 lings (however if you get that hatch off in front of the probe, then you can just make only 2 lings because there's no pylon you need to deal with).
My problem with ZvP is scouting and understanding what I scout so I'll read this great guide of yours a couple of times and I should be fine .
Also I actually like your attitude, I don't mind when people have confidence in what they say instead of trying to be 'politically correct' all the time... hope you understand what I mean.
Belial, 11 overpool is not shit. I might have to show you again why.
Also for those saying they get 70 drones at the 8 minute mark, you won't be able to considering that you will lose overlords while saccing and you will be forced to make half a round of lings to kill scouting zealots.
^ I don't recall you ever showing me why in ZvP it is not shit. If you are talking about getting lings in, isn't that 10 pool, not 11 pool? If you are talking about 11 pool for aggression though, that's a little bit different though, although yes, that would mean I might be a bit wrong in my criticism.
Also for those saying they get 70 drones at the 8 minute mark, you won't be able to considering that you will lose overlords while saccing and you will be forced to make half a round of lings to kill scouting zealots.
Even while losing overlords and making lings to deal with zealots, you can still get to 70+. On Dual Sight Idra vs Oz, Idra got to 70+ even after dealing with a cannon rush at his third.
Obviously, that's idra though, and obviously, many of us can't hit 70+ in a real game, nor ever hope to consistently hit that, but that doesn't change the fact that it's a benchmark we should be striving for. If you can get past 60 supply, you should be good. If you are in the lower 50's in a real game, that's definitely a problem.
On March 13 2012 19:50 Belial88 wrote: ^ I don't recall you ever showing me why in ZvP it is not shit. If you are talking about getting lings in, isn't that 10 pool, not 11 pool? If you are talking about 11 pool for aggression though, that's a little bit different though, although yes, that would mean I might be a bit wrong in my criticism.
Also for those saying they get 70 drones at the 8 minute mark, you won't be able to considering that you will lose overlords while saccing and you will be forced to make half a round of lings to kill scouting zealots.
Even while losing overlords and making lings to deal with zealots, you can still get to 70+. On Dual Sight Idra vs Oz, Idra got to 70+ even after dealing with a cannon rush at his third.
Obviously, that's idra though, and obviously, many of us can't hit 70+ in a real game, nor ever hope to consistently hit that, but that doesn't change the fact that it's a benchmark we should be striving for. If you can get past 60 supply, you should be good. If you are in the lower 50's in a real game, that's definitely a problem.
Not aggression. You screw with the protoss BO with this build. 11pool forces cannon before nexus and forces a cannon before gateway as well as a second gateway + cancel. Beyond that, you also force the building probe to stay near his cannons. Everything adds up.
A lot of pros argue its still not as good, but its definitely not shit.
70 drones should only be allowed when the protoss lets you. Cannon rush being failed actually makes it easier to drone. Almost all protosses send 1 or 2 zealots to pressure to force you to make lings. You will not make 70, not in the current metagame zealot timings.
Well, you can, but you are probably skimping on creepspread or spines or gas.
That's how protosses are still winning 2-base all-ins against zerg, the uncertainty in timings of zealot/stalker harass forcing zerg to macro safer. Otherwise, no one should be able to win a 2-base all-in against anyone beyond high masters because a risktaking zerg will always crush 2-base allins.
Either the Protoss is very greedy, then you will do serious damage and delay mining with those lings (running by the cannon if it's up).
This is a huge gamble, and Toss will always scout you before your pool pops (that's just standard at high level play to make sure no 6 pool and if can get away with nexus first possibly, although not sure about platinum). Once they arrive in your base and realize your pool was planted before 2:00 (ie earlier than 14), they will sure as hell not be 'greedy'. Maybe in platinum Toss players can't tell if it's 10 pool or 14 pool, but I would imagine fairly early on and in lower levels, toss players can tell if it's 14 pool or 10 pool, and will react accordingly. Even if you go 11 pool, they can still get away with being fairly greedy.
And, the worst thing is, is that you are behind about 5-7 drones by going 11 pool with 6 lings. So you need to kill at least 4 probes with your 6 lings. There is no way in hell you are ever going to kill more than maybe 2 workers, if any, against a competent protoss, with only 6 lings, especially when they get a cannon up to shoot the lings so it's only 4-5 lings, which suddenly can't take on a zealot that has probes near by to help out if you engage.
So I really don't think 11 pool is good in that regard. You are just gambling that the Toss is a fucking moron and can't tell you put down the pool before 2:00 by seeing it more than half done when they arrive.
Or, the protoss isn't greedy, which gives you even more time in the long run. This is basically win/win situation.
How is that a win, that you made a bunch of units early on that accomplished nothing? it's not like making 6 lings at the start loses you 3 drones - it's more like losing 10 drones compared to going just 14 pool and making only drones. I mean, there's a reason why zergs don't make units at the start of the game if it's such a 'win/win'. You know, the whole, if you don't make pure drones at the start of the game, you are going to be so fucking far behind chrono/mule that you will just die in econ game.
Just take your natural, and compare 14 pool to 11 pool. You should notice a huge difference in economy.
I really disagree with your points.
Jombo, please tell me what the merits of 11 pool are.
Not aggression. You screw with the protoss BO with this build. 11pool forces cannon before nexus and forces a cannon before gateway as well as a second gateway + cancel. Beyond that, you also force the building probe to stay near his cannons. Everything adds up.
A lot of pros argue its still not as good, but its definitely not shit.
Okay, so it's not shit, but it still sounds like a gamble that Toss can see ahead of time and, if they react correctly, come out ahead. Toss will see it coming, and there's a definite thing Toss can do to end up ahead at the end of it all.
70 drones should only be allowed when the protoss lets you. Cannon rush being failed actually makes it easier to drone. Almost all protosses send 1 or 2 zealots to pressure to force you to make lings. You will not make 70, not in the current metagame zealot timings.
I mean in that game Oz did a cannon rush + gateway attack at the third, not just a standard cannon rush at the natural that failed miserably.
I see DRG regularly hit 70+ drones in tournament play. If you are saying that I, personally, won't hit 70 against a competent toss (and competent of course means high masters, which I'm not, so by definition, of course not, but against a mid-masters toss, I could). But it's definitely possible.
That's how protosses are still winning 2-base all-ins against zerg, the uncertainty in timings of zealot/stalker harass forcing zerg to macro safer. Otherwise, no one should be able to win a 2-base all-in against anyone beyond high masters because a risktaking zerg will always crush 2-base allins.
DRG gets to 75 drones because he's taking risks against FE protosses. Mainly from the fact that protosses know that DRG will be able to parry any type of small pressure with simply better play that they wont do it, plus Godly scout timings. If MC 2-base all-in'ed a DRG grabbing 75 drones at 8:00, MC will most likely clinch the win.
^ It's still possible to still hit 75 supply (not saying drones there) while dealing with such harass, like idra dealing with oz's pressure. I never said 75 drones, and 75 supply anyways means more like 67 drones at most, although really at 8:00 that means like 60 drones actually mining at that moment at most.
On March 13 2012 20:18 Belial88 wrote: ^ It's still possible to still hit 75 supply (not saying drones there) while dealing with such harass, like idra dealing with oz's pressure. I never said 75 drones, and 75 supply anyways means more like 67 drones at most, although really at 8:00 that means like 60 drones actually mining at that moment at most.
A failed pressure means that timings are ruined. It makes droning completely free. Show me the game, and I'll analyze it. I use 75 drones because we are specifically talking DRG, of which I'm taking him as the specific example. I still think that letting zerg hit 70 drones at 8 minute mark uncontested means that protoss did something WRONG (or he took a greedy third/tech). In which case, it wasn't that you CAN hit 70 drones, its that the protoss has let you.
With that logic, I can hit 200 drones at the 13 minute mark.
EDIT:
I'm going to tl;dr my argument. 70 drones at 8:00 is not a benchmark, it is a hopeful. If zergs start believing that they have to hit 70 drones at 8:00 to be optimal, you're going to start seeing a lot of zergs die to random zealot or zealot/stalker pressures.
Pretty solid. I would offer that you should in fact get an upgrade if he's going for a WG pressure build to insure you have a followup. Most people just go all-in with 6 gate, but there are protoss (especially on the NA server) who 6 gate while taking a double gas behind it to put together a pretty strong timing on the third with a lot of sentry energy and then transitions (usually to 3rd and blink with a slower colossus). You can hold off a really aggressive and well controlled 6 gate while getting an upgrade; 4 roaches definitely is better than 100 gas tied up in something that won't be helpful durring this attack, but I think it's important that you have that upgrade after the attack.
I never said 70 drones, I said 70 supply. Also, 70 supply at 8:00 minute means you have like somewhere in the upper 50's of drones actually mining, and the rest of the supply taken up by morphing drones, as well as queens and random lings.
The game I was talking about was Idra vs Oz on Dual Sight, but if you watch any of the DRG vs Genius finals he hit 70+ in all but one game (i think the one he didn't was daybreak).
Pretty solid. I would offer that you should in fact get an upgrade if he's going for a WG pressure build to insure you have a followup.
If you want to, sure, of course. I would even recommend it. I was just saying DONT get it if toss is doing a hardcore gateway all-in. Otherwise, do whatever you think is necessary.
Most people just go all-in with 6 gate, but there are protoss (especially on the NA server) who 6 gate while taking a double gas behind it to put together a pretty strong timing on the third with a lot of sentry energy and then transitions (usually to 3rd and blink with a slower colossus).
If someone is doing a sub-optimal bad play, and you react as if they were playing optimally, you will only end up ahead. That's like saying making roach/ling against a 4 gate is bad because Toss could theoretically just be going 4 gate expand.
No.
Toss goes 6 gate, you see no gas, so he won't ahve lots of sentries. Geysers take a while to warp in, and build up gas. So you get 1 gas, that's 30 seconds, then about another minute to basically get 1 sentry out. So it's like 2 minutes to get 2 sentries from 1 gas. So getting 2 gas late, as in past the 8:00 mark, means Toss won't have sentries for a long fucking time. He would've done better going 3 gas normal 5 gate sentry expand to third or something.
So. You see 6 gate no gas. You make a ton of roach/ling. Toss can't possibly take his third, because he won't have enough sentries, and you will have a shitton of units. You will just fucking deny his third forever. He will have to wait forever. In the meantime, you are teching hard and fast to speed roaches, burrow, hydras, mutas, whatever, while taking a fourth base.
And really, by taht point in the game, you've won. You can't really do any 'timing' with 6 gate besides when warp gate first finishes. If a 6 gate timing isn't going to work (assuming proper reaction from zerg) at the standard 8:30 mark, how is it possibly going to work at the 10:00 mark? Makes no sense. By then you'll have a tonof hydras.
I don't really understand what you are saying. If toss is doing a sub-optimal build, you just end up ahead by responding as if he was playing optimally.
If Toss is doing a no gas 6 gate that's really bad and late, you don't need an upgrade. So what you do is see no gas, so you know what's up, you don't get the upgrade. You mass roach/ling, make no lair. Then, you are at about 90-110 supply. You realize, hey, my army way fucking bigger now, I'd stomp him if he moves out, because my econ has kicked in, and he has no tech that can take advantage of my hatch tech. So then you tech up, and get upgrades, lair, speed, et cetera.
For the same reason, you can't just 4 gate expand. You just die. Like you can't do it. For the same reasons, you can't just 6 gate no gas FFE and just play out.
On March 13 2012 20:18 Belial88 wrote: ^ It's still possible to still hit 75 supply (not saying drones there) while dealing with such harass, like idra dealing with oz's pressure. I never said 75 drones, and 75 supply anyways means more like 67 drones at most, although really at 8:00 that means like 60 drones actually mining at that moment at most.
A failed pressure means that timings are ruined. It makes droning completely free. Show me the game, and I'll analyze it. I use 75 drones because we are specifically talking DRG, of which I'm taking him as the specific example. I still think that letting zerg hit 70 drones at 8 minute mark uncontested means that protoss did something WRONG (or he took a greedy third/tech). In which case, it wasn't that you CAN hit 70 drones, its that the protoss has let you.
With that logic, I can hit 200 drones at the 13 minute mark.
EDIT:
I'm going to tl;dr my argument. 70 drones at 8:00 is not a benchmark, it is a hopeful. If zergs start believing that they have to hit 70 drones at 8:00 to be optimal, you're going to start seeing a lot of zergs die to random zealot or zealot/stalker pressures.
First of all, we're talking supply, not drones. Having 75 drones by 8min mark is probably impossible. Second of all, why is it not a benchmark just because pressure can put you in a situation where it isn't adviceable? I would think it pretty easy to open a replay, go to 8minutes, check your supply and try to defend why supply is lower than 65 when it's very possible to get to 75 supply in mostly drones if you macro well unless you're somehow forced not to.
The benchmark is just saying "It's possible to macro this well, even when dealing with some pressure, do your best to get there", if you have good reasons for being at lower supply, there's nothing more to it. Of course zergs would think it optimal to hit 75 supply by the 8 minute mark, in mostly drones... all zergs know this, the whole race is basically centered around getting as many workers as quickly as possible and then win off the economic advantage. Just because it's optimal doesn't mean zergs actually go for it if they have a reason not to.
Yea I'm thinking 75 in a real game is just perfect play. You should at least strive for 60+. By gauging that, you can tell if you lost that game not because colossi are OP, but because you were only at 50 at the 8:00 minute mark, or, that colossi really are OP, because you were at 70 at the 8:00 mark. It can rule out what you are doing wrong or right, that's all. If you are hitting 80+, maybe that means you are getting gas way too late (made obvious by, you know, no lair or speed until forever though). That's all.
Ok but then how am I able to hit 69 supply anyways?
The difference isn't very large between 14/15pool and 11overpool in economy. Probably have to look at creep spread or gas count.
The difference is actually pretty big, but it isn't noticeable until the 8 minute mark and later.
11 overpool with 18 hatch is slightly ahead of 14pool 16 hatch in the early game due to the fact that you get your queens out earlier. But 14 pool/16 hatch get's its third placed quicker and thereby gets two extra injects worth of drones before 11 overpool at the third.
14p16h: If you place the third at 4:30 it finishes at 6:10, the first inject pops at 6:50, the second at 7:30 and the 3rd at 8:10. 11p18h: If you place the third at 5:50 it finishes at 7:30, the first inject pops at 8:10.
In addition to the injects, you get 5 more larvae spawned from the hatchery placed 1m20s earlier. That's a total of 4+4+5 = 13 larvae if you hit your injects.
On 2 base maps, like shattered or tal'darim, I think 11 overpool might be better.
Thank you Belial, this is an excellent guide. Guess my friends will now have a hard time beating me PvZ!
I may have overlooked something, but could you possibly elaborate a bit on queen production and usage? I get my first queen directly at 16 supply. Then I inject two times, while producing the second queen in the main after I have taken the third base. I then place one creep tumor in the main and move that queen to the nat, while I produce a third queen at the third. In case I suspect some air play, I get two additional queens at my 2nd and 3rd at around 7:30. It would be helpful for me if you could comment on that.
Yesterday I watched a bit of Stephano's stream and his supply around 8 minutes.
Then I realized that I need to follow Belial's advise to get enough supply for that time mark. I need to build more stuff instead of randomly morph tech structures or having larvae idling around. While the guide is quite long with much content, this is the single thing I want to achieve.
^ It's up to you when to get queens, but I would think you want to start injecting ASAP. You can get some queens a bit before a hatch pops to maybe get creep going. What you say you do seems pretty sound to me.
Stephano makes 6 lings (last 4 lings are a little later as a response to scouting zealot), has 3 queens, 1 creep tumor, 6:00 double gas, 7:00 evo+roach warren and is at 64 supply at 8:00. This is AFTER making a huge mistake of being supply blocked and way overmaking overlords. Also he went for lair with first 100 gas, then ling speed, and has barely enough for +1 or for roaches depending what he needs at 8:00. This is all the most standard of standard things so it would be a good benchmark.
Ofcourse imagine if the didnt supply block himself for ~45 seconds then make 5 overlords in advance.
I recommend people try playing vs the very easy AI and practice their ability to drone. Once they reach a goal then they should try their droning habits vs protoss opponents.
Im also debating starting a thread about "Racing to 200/200" as a way to teach zergs to learn how to drone perfectly including build orders, optimized rally points/drone stacking, queen timings, ect. To be more relevant to actual gameplay, I want to add certain perameters like "6:00 double gas, 7:00 evo+roach". Would anyone be interested in that?
I've typically gone 12p19h in ZvP, but I'm trying to change my ways to 14p16h. If you don't get the hatch down, how do you correctly follow it up? I currently do overlord, 4 lings, then double extractor trick to get the queen out. How should it be done, as that feels kind of 'funky'/
^ 14 pool, 17 overlord, get 4 lings out and queen when pool pops. I set 2 drones out at 16, one to block ramp, one to follow probe. If I get lucky, I'll put down a 16 hatch, but that rarely happens.
Yes, you are delayed from 16 hatch to 21 hatch, but toss is just as delayed by making that pylon. It's not a big deal, a pylon is useless, but 4 lings aren't. So you are both spending 100 minerals, but you actually get something out of it.
I think the idea back in the day was that if Zerg goes pool first, toss should not put down a pylon. This was back when people went hatch first a lot. But then I think a lot of Toss went for the pylon because it forces Zerg to make a choice of either go for the probe and delay their hatch by a lot, or let the probe go free. This is solved by just putting 3 lings on the pylon (add a drone there, or not) and 1 ling to chase the probe.
Excellent guide with tons of useful timings, scouting information and macro benchmarks.
Kudos for putting it together, very thorough
Glad you like it, but aren't you at a level where you already know all of this?
I'm assuming all of the diamond+ knew all of this. Maybe the benchmarking section might have been useful? I don't know. Let me know what you learned when you post guys. And I can't be all-knowing, surely there is stuff out there that I could add that I am not aware of.
For silver-league zergs like me it's not important that a guide is verbose. Actually I am often overwhelmed by details. I think the macro advise is so useful it could be used for a stand-alone beginner's guide: Learn how to macro up (and defend) to be prepared for the 8 minute mark.
On March 13 2012 21:53 Host- wrote: I've typically gone 12p19h in ZvP, but I'm trying to change my ways to 14p16h. If you don't get the hatch down, how do you correctly follow it up? I currently do overlord, 4 lings, then double extractor trick to get the queen out. How should it be done, as that feels kind of 'funky'/
Personally, I go overlord immediately on 16 if I see a probe since I never count on getting that hatch down, I find one should always play with the idea that your opponent is playing well... blocking a hatchery is really easy, I'm not going to bank on my opponent failing. I still send a drone, more to force the pylon, obviously taking the hatch if I get the chance (overlord is just 100 minerals, it doesn't take long to get back to 300 if you get a shot at it), but I quite quickly decide if I'm not going to get the hatch and build my queen. Overlord shouldn't be far away, so I immediately get the lings.
I'm pretty sure that if you do a measurement in economy, the difference isn't all that big between 16 hatch and 20 hatch, you're just switching the order of the hatch and the queen/lings, so I think that you could even go as far as to blindly build the queens and lings first, just to be safe.
Excellent guide with tons of useful timings, scouting information and macro benchmarks.
Kudos for putting it together, very thorough
Glad you like it, but aren't you at a level where you already know all of this?
I'm assuming all of the diamond+ knew all of this. Maybe the benchmarking section might have been useful? I don't know. Let me know what you learned when you post guys. And I can't be all-knowing, surely there is stuff out there that I could add that I am not aware of.
Belial88 thank you for creating this Guide! Allways searched for guide with scout timings + how to react to specific things!
As is said you before, i vote Belial for Blueposts :D. Allways #1 posts with essential explanations. Cheers
Thanks, but I don't think that will ever happen. Most blues hate me. I *am* a lower level person, so I say the wrong thing every now and then with, you know, my standard conviction, and they take it as me spreading misinformation maliciously, and when I explain myself, they take it that I'm explaining "The Correct Answer" as opposed to "Belial's Answer". I've been proven wrong many times here, and have been very open about when I've been wrong and changed (i used to say you should always hatch first in zvp because you can stop cannon rush, i used to say 11 pool was best zvp build for ladder due to patrol vs ramp block and timing vs gateway expand, i used to say hydras were worthless in every match up, et cetera).
You'll notice a lack of blues ever posting in any of my guide threads. I think the ups know I'm a force for good and I try to behave, and that my mind is in the right place, but also don't like me. They seem to only post near me to say im a condescending dick.
Which is understandable, I get frustrated when I see lower level players say wrong things too sometimes. But any higher level player who talks to me, usually gets where I'm coming from and doesn't have a problem, and I think I'm humble enough that I'll listen and admit I'm wrong when someone explains what's up to me. Some of them don't bother though. They just think that any time I make a point, or disagree on something, I'm 'arguing'.
On March 13 2012 15:26 Iksf wrote: Just think your a bit pessimistic about the benchmarking, im diamond and generally get 70-75 by 8:00 without specifically trying, really don't get how people could get below 60 without serious supply blocks.
Without trying? Can you post a replay please, I'm vary curious. Here's a replay of me doing a very greedy and precise build by myself. At 8:00 I have: 1 larva, 65 drones, 2 lings, 9 overlords, 3 queens (Total: 78/78) 2 gas roach warren basically done +1 attack starting
This is under ideal conditions where the the hatcheries do not get blocked and I only make 2 lings early. I'd like to see other players post their builds so I can compare.
Heres one of my more recent ZvP games, not the best but iv not been playing much recently. http://drop.sc/133255
Also not sure if you uploaded wrong replay but at 8:10, adding 10 seconds for the start, was more like 4 or so tbh, u were at 66 food without roach warren complete or +1 attack anywhere in sight. I dunno if its the wrong replay or a bug, the timer doesn't seem to work for me only the replay bar.
73/78 here. I really like the benchmarking. Im gonna have to take a closer look at your scouting cause Im terrible at scouting protoss tech routes in the early and middle game.
On March 13 2012 21:03 Belial88 wrote: I never said 70 drones, I said 70 supply. Also, 70 supply at 8:00 minute means you have like somewhere in the upper 50's of drones actually mining, and the rest of the supply taken up by morphing drones, as well as queens and random lings.
The game I was talking about was Idra vs Oz on Dual Sight, but if you watch any of the DRG vs Genius finals he hit 70+ in all but one game (i think the one he didn't was daybreak).
Pretty solid. I would offer that you should in fact get an upgrade if he's going for a WG pressure build to insure you have a followup.
If you want to, sure, of course. I would even recommend it. I was just saying DONT get it if toss is doing a hardcore gateway all-in. Otherwise, do whatever you think is necessary.
Most people just go all-in with 6 gate, but there are protoss (especially on the NA server) who 6 gate while taking a double gas behind it to put together a pretty strong timing on the third with a lot of sentry energy and then transitions (usually to 3rd and blink with a slower colossus).
If someone is doing a sub-optimal bad play, and you react as if they were playing optimally, you will only end up ahead. That's like saying making roach/ling against a 4 gate is bad because Toss could theoretically just be going 4 gate expand.
No.
Toss goes 6 gate, you see no gas, so he won't ahve lots of sentries. Geysers take a while to warp in, and build up gas. So you get 1 gas, that's 30 seconds, then about another minute to basically get 1 sentry out. So it's like 2 minutes to get 2 sentries from 1 gas. So getting 2 gas late, as in past the 8:00 mark, means Toss won't have sentries for a long fucking time. He would've done better going 3 gas normal 5 gate sentry expand to third or something.
So. You see 6 gate no gas. You make a ton of roach/ling. Toss can't possibly take his third, because he won't have enough sentries, and you will have a shitton of units. You will just fucking deny his third forever. He will have to wait forever. In the meantime, you are teching hard and fast to speed roaches, burrow, hydras, mutas, whatever, while taking a fourth base.
And really, by taht point in the game, you've won. You can't really do any 'timing' with 6 gate besides when warp gate first finishes. If a 6 gate timing isn't going to work (assuming proper reaction from zerg) at the standard 8:30 mark, how is it possibly going to work at the 10:00 mark? Makes no sense. By then you'll have a tonof hydras.
I don't really understand what you are saying. If toss is doing a sub-optimal build, you just end up ahead by responding as if he was playing optimally.
If Toss is doing a no gas 6 gate that's really bad and late, you don't need an upgrade. So what you do is see no gas, so you know what's up, you don't get the upgrade. You mass roach/ling, make no lair. Then, you are at about 90-110 supply. You realize, hey, my army way fucking bigger now, I'd stomp him if he moves out, because my econ has kicked in, and he has no tech that can take advantage of my hatch tech. So then you tech up, and get upgrades, lair, speed, et cetera.
For the same reason, you can't just 4 gate expand. You just die. Like you can't do it. For the same reasons, you can't just 6 gate no gas FFE and just play out.
I don't know why you're harping that it's a supoptimal build; it's a build NonY brought up on SotG after IEM Gaungzhao and has used several times on ladder to great success. The has the regular timing with his orignal sentries where he has a lot of energy and he uses it for one timine where protoss has a lot of FF's and tries to snipe the third. Instead of doing the hardcore just warping in stalkers/zealots to try to kill the guy, he continues probing and takes the double gas as he leaves and doesn't over commit into the third. I know you like to argue, but perhaps you can admit that a low masters zerg doesn't know the game as well as NonY, so you might take his word for it?
Okay, I tried that to concentrate on my macro while playing against a friend of mine. My macro sucks. Maybe not hardcore, but it sucks.
I hit about something between 60 and 65 supply and here are some key points I noticed:
1. Hatch goes down at 16 (no pylon block). Whats the correct macro followup? a) 16 queen 18 overlord 18 lings b) 16 overlord 16 queen 18 lings c) 15 queen (followup?)
2. Which are the overlord steps? I find myself supplyblocked quite often at the early stages. Sometimes I got the feeling Im on point with my overlords but most times I end up having larvae that I cant use ... Should I just try to get overlords slightly earier?
If you wait for around 45 supply to take all three gas, a +1 4gate will hit a very vulnerable time in your build, so it's worth noting that when delaying gas and therefor roach warren for so long, one should have enough lings to repel the 2 zealot 1 stalker 1 probe poke that is the beginning of the pressure and escort for the proxy pylon of a +1 4gate. Denying that pylon is absolutely key, it buys time for speed and raoch tech to kick in.
I made it pretty clear it's up to you when you want to take gas. I just stated that the 'standard' gas is 2 gas taken when third finishes, so if you want to, you can definitely go 1 gas much earlier, the 2nd gas much later, or take 3 gas later like DRG, or just 2 gas a bit earlier, or a bit earlier.
You won't have speed out in time to deal with 1 zealot or 1 stalker pressure though, regardless, and if you were to get gas before third, that would just be too much. Getting 2 gas when third finishes around 6:00 should get you speed in time for 8:00 4 gate pressure though, or soon enough that slowlings or a spine or queen will buy that extra 10 more seconds for you.
If you want to get gas earlier, it's up to you, but there are serious ramifications for doing so, and there's a reason pros wait as long as when third is up for 2 gas, or even later than that. I mean, delaying gas as much as possible is kind of the idea, and you really don't need gas for a long time anyways (even if it's 4 gate pressure you are talking about, it's still relatively late).
And, this is a basic guide. You can tailor your third to what you scout. So if you see no gas taken, maybe you can get a single gas quicker to get speed out or fast roaches to deal with pressure and then just not take the 2nd gas for a super long time. For example, when I see doublegate before core, I throw down gas asap, like right when I plant third, if not sooner, so I can have speed against +1 double gate zealots (which you will crush with speedlings).
I'm not challenging anything you wrote in your guide, I was adding guidelines knowledge and discussion for people who wanted to delay gas for a longer period of time.
I guess my point is there is a relationship of number of lings built when scouting an early plus 1 in relative to the point in time when you take your gas. And these lings are purely to deny the proxy pylon.
The start of +1 4gate pressure usually begins with 2 zealots and 1 stalker escorting a probe, and if you want to delay gas, I'm informing other users you have to know the timing to build enough lings to deny that pylon, or risk taking a lot of damage because roaches aren't quite ready yet. This is the center piece of information I was getting at. I was posting this for other users to discuss and learn from, not challenging your guide, or asking if you agree. I'm confident in my information.
On March 14 2012 04:41 stichlasser wrote: Okay, I tried that to concentrate on my macro while playing against a friend of mine. My macro sucks. Maybe not hardcore, but it sucks.
I hit about something between 60 and 65 supply and here are some key points I noticed:
1. Hatch goes down at 16 (no pylon block). Whats the correct macro followup? a) 16 queen 18 overlord 18 lings b) 16 overlord 16 queen 18 lings c) 15 queen (followup?)
2. Which are the overlord steps? I find myself supplyblocked quite often at the early stages. Sometimes I got the feeling Im on point with my overlords but most times I end up having larvae that I cant use ... Should I just try to get overlords slightly earier?
The way I prefer to follow up a non blocked 16hatch is: 4 lings right away to look for probe. This gives you 17/18. queue up a queen, then make and cancel an extractor (19/18) The rest as far as queen timings are really up to the player, you can decide to cut a queen for a period of time to in order to use all your droning capability right away, or you can get the queens earlier and have 2 creep tumors down to help with pushes and pressures.
So the overlord numbers that are pretty easy to get down as I use them are: 19/18 24/28 32/38 38/46 45/54
Those are the super basic making drones overlord numbers, very conducive to constant and efficient drone production.
On March 14 2012 04:41 stichlasser wrote: Okay, I tried that to concentrate on my macro while playing against a friend of mine. My macro sucks. Maybe not hardcore, but it sucks.
I hit about something between 60 and 65 supply and here are some key points I noticed:
1. Hatch goes down at 16 (no pylon block). Whats the correct macro followup? a) 16 queen 18 overlord 18 lings b) 16 overlord 16 queen 18 lings c) 15 queen (followup?)
2. Which are the overlord steps? I find myself supplyblocked quite often at the early stages. Sometimes I got the feeling Im on point with my overlords but most times I end up having larvae that I cant use ... Should I just try to get overlords slightly earier?
The way I prefer to follow up a non blocked 16hatch is: 4 lings right away to look for probe. This gives you 17/18. queue up a queen, then make and cancel an extractor (19/18) The rest as far as queen timings are really up to the player, you can decide to cut a queen for a period of time to in order to use all your droning capability right away, or you can get the queens earlier and have 2 creep tumors down to help with pushes and pressures.
So the overlord numbers that are pretty easy to get down as I use them are: 19/18 24/28 32/38 38/46 45/54
Those are the super basic making drones overlord numbers, very conducive to constant and efficient drone production.
Thank you sir! I think if I can work out the basics better, i will have less trouble doing proper scouting and reacting =)
On March 13 2012 21:48 TheGreenMachine wrote: About the maximum of 60 to 75 supply at 8:00 thing, i'm watching stephano's most recent vod vs whitera and realizing he hits that goal rather easily.
Stephano makes 6 lings (last 4 lings are a little later as a response to scouting zealot), has 3 queens, 1 creep tumor, 6:00 double gas, 7:00 evo+roach warren and is at 64 supply at 8:00. This is AFTER making a huge mistake of being supply blocked and way overmaking overlords. Also he went for lair with first 100 gas, then ling speed, and has barely enough for +1 or for roaches depending what he needs at 8:00. This is all the most standard of standard things so it would be a good benchmark.
Ofcourse imagine if the didnt supply block himself for ~45 seconds then make 5 overlords in advance.
I recommend people try playing vs the very easy AI and practice their ability to drone. Once they reach a goal then they should try their droning habits vs protoss opponents.
Im also debating starting a thread about "Racing to 200/200" as a way to teach zergs to learn how to drone perfectly including build orders, optimized rally points/drone stacking, queen timings, ect. To be more relevant to actual gameplay, I want to add certain perameters like "6:00 double gas, 7:00 evo+roach". Would anyone be interested in that?
I'd be really interested in a thread like that. There are a lot of protoss players who put on little to no pressure for the first 9-10 min of the game and it would be good to find some optimal 3 base build orders. Feel free to PM me if you want to talk about it some more. I have a build I've been working on and would like to share ideas.
On March 14 2012 04:41 stichlasser wrote: Okay, I tried that to concentrate on my macro while playing against a friend of mine. My macro sucks. Maybe not hardcore, but it sucks.
I hit about something between 60 and 65 supply and here are some key points I noticed:
1. Hatch goes down at 16 (no pylon block). Whats the correct macro followup? a) 16 queen 18 overlord 18 lings b) 16 overlord 16 queen 18 lings c) 15 queen (followup?)
2. Which are the overlord steps? I find myself supplyblocked quite often at the early stages. Sometimes I got the feeling Im on point with my overlords but most times I end up having larvae that I cant use ... Should I just try to get overlords slightly earier?
For your second point, try creating one overlord every time you inject. So if you inject two hatcheries at once, immediately build 2 overlords.
This is far from optimal, but it's easy to remember and focus on. You'll see that you'll quickly run away in supply, so all you have to do then is scale it down, which you do by checking your supply every time you inject and estimating how many overlords you need. This will eventually give you all the tools you need for proper overlord attainment. Later, when you build roaches, one overlord per inject isn't even enough, so it's not as bad of a theory as it might sound. Early on though, you can skip an overlord some injects, which will be nice for your economy.
On March 13 2012 15:26 Iksf wrote: Just think your a bit pessimistic about the benchmarking, im diamond and generally get 70-75 by 8:00 without specifically trying, really don't get how people could get below 60 without serious supply blocks.
Without trying? Can you post a replay please, I'm vary curious. Here's a replay of me doing a very greedy and precise build by myself. At 8:00 I have: 1 larva, 65 drones, 2 lings, 9 overlords, 3 queens (Total: 78/78) 2 gas roach warren basically done +1 attack starting
This is under ideal conditions where the the hatcheries do not get blocked and I only make 2 lings early. I'd like to see other players post their builds so I can compare.
Heres one of my more recent ZvP games, not the best but iv not been playing much recently. http://drop.sc/133255
Also not sure if you uploaded wrong replay but at 8:10, adding 10 seconds for the start, was more like 4 or so tbh, u were at 66 food without roach warren complete or +1 attack anywhere in sight. I dunno if its the wrong replay or a bug, the timer doesn't seem to work for me only the replay bar.
Ah, I see. I was using the SALT version of the map (the save/load one) in case I made a mistake so I wouldn't have to redo it. The extra replay time was because it pauses the game when you save but the overlay game timer showed 8:01.
I think the reason it was so hard for me to get 75 supply was because I make the evo chamber and start my lair early, stephano style.
On March 13 2012 21:03 Belial88 wrote: I never said 70 drones, I said 70 supply. Also, 70 supply at 8:00 minute means you have like somewhere in the upper 50's of drones actually mining, and the rest of the supply taken up by morphing drones, as well as queens and random lings.
The game I was talking about was Idra vs Oz on Dual Sight, but if you watch any of the DRG vs Genius finals he hit 70+ in all but one game (i think the one he didn't was daybreak).
Pretty solid. I would offer that you should in fact get an upgrade if he's going for a WG pressure build to insure you have a followup.
If you want to, sure, of course. I would even recommend it. I was just saying DONT get it if toss is doing a hardcore gateway all-in. Otherwise, do whatever you think is necessary.
Most people just go all-in with 6 gate, but there are protoss (especially on the NA server) who 6 gate while taking a double gas behind it to put together a pretty strong timing on the third with a lot of sentry energy and then transitions (usually to 3rd and blink with a slower colossus).
If someone is doing a sub-optimal bad play, and you react as if they were playing optimally, you will only end up ahead. That's like saying making roach/ling against a 4 gate is bad because Toss could theoretically just be going 4 gate expand.
No.
Toss goes 6 gate, you see no gas, so he won't ahve lots of sentries. Geysers take a while to warp in, and build up gas. So you get 1 gas, that's 30 seconds, then about another minute to basically get 1 sentry out. So it's like 2 minutes to get 2 sentries from 1 gas. So getting 2 gas late, as in past the 8:00 mark, means Toss won't have sentries for a long fucking time. He would've done better going 3 gas normal 5 gate sentry expand to third or something.
So. You see 6 gate no gas. You make a ton of roach/ling. Toss can't possibly take his third, because he won't have enough sentries, and you will have a shitton of units. You will just fucking deny his third forever. He will have to wait forever. In the meantime, you are teching hard and fast to speed roaches, burrow, hydras, mutas, whatever, while taking a fourth base.
And really, by taht point in the game, you've won. You can't really do any 'timing' with 6 gate besides when warp gate first finishes. If a 6 gate timing isn't going to work (assuming proper reaction from zerg) at the standard 8:30 mark, how is it possibly going to work at the 10:00 mark? Makes no sense. By then you'll have a tonof hydras.
I don't really understand what you are saying. If toss is doing a sub-optimal build, you just end up ahead by responding as if he was playing optimally.
If Toss is doing a no gas 6 gate that's really bad and late, you don't need an upgrade. So what you do is see no gas, so you know what's up, you don't get the upgrade. You mass roach/ling, make no lair. Then, you are at about 90-110 supply. You realize, hey, my army way fucking bigger now, I'd stomp him if he moves out, because my econ has kicked in, and he has no tech that can take advantage of my hatch tech. So then you tech up, and get upgrades, lair, speed, et cetera.
For the same reason, you can't just 4 gate expand. You just die. Like you can't do it. For the same reasons, you can't just 6 gate no gas FFE and just play out.
I don't know why you're harping that it's a supoptimal build; it's a build NonY brought up on SotG after IEM Gaungzhao and has used several times on ladder to great success. The has the regular timing with his orignal sentries where he has a lot of energy and he uses it for one timine where protoss has a lot of FF's and tries to snipe the third. Instead of doing the hardcore just warping in stalkers/zealots to try to kill the guy, he continues probing and takes the double gas as he leaves and doesn't over commit into the third. I know you like to argue, but perhaps you can admit that a low masters zerg doesn't know the game as well as NonY, so you might take his word for it?
I mostly do it off 3gate now and it is very strong against zergs who have taken a 3rd going roach/ling. 6gate is still good against some zerg builds but it's pretty terrible against others and you've got no control over what they're doing, nor scouting to cut your losses and change builds at an early enough time.
Another important point is that I play from 2gate or 3gate expand, not FFE. I don't think FFE can pressure a 3rd very safely.
On March 13 2012 21:03 Belial88 wrote: I never said 70 drones, I said 70 supply. Also, 70 supply at 8:00 minute means you have like somewhere in the upper 50's of drones actually mining, and the rest of the supply taken up by morphing drones, as well as queens and random lings.
The game I was talking about was Idra vs Oz on Dual Sight, but if you watch any of the DRG vs Genius finals he hit 70+ in all but one game (i think the one he didn't was daybreak).
Pretty solid. I would offer that you should in fact get an upgrade if he's going for a WG pressure build to insure you have a followup.
If you want to, sure, of course. I would even recommend it. I was just saying DONT get it if toss is doing a hardcore gateway all-in. Otherwise, do whatever you think is necessary.
Most people just go all-in with 6 gate, but there are protoss (especially on the NA server) who 6 gate while taking a double gas behind it to put together a pretty strong timing on the third with a lot of sentry energy and then transitions (usually to 3rd and blink with a slower colossus).
If someone is doing a sub-optimal bad play, and you react as if they were playing optimally, you will only end up ahead. That's like saying making roach/ling against a 4 gate is bad because Toss could theoretically just be going 4 gate expand.
No.
Toss goes 6 gate, you see no gas, so he won't ahve lots of sentries. Geysers take a while to warp in, and build up gas. So you get 1 gas, that's 30 seconds, then about another minute to basically get 1 sentry out. So it's like 2 minutes to get 2 sentries from 1 gas. So getting 2 gas late, as in past the 8:00 mark, means Toss won't have sentries for a long fucking time. He would've done better going 3 gas normal 5 gate sentry expand to third or something.
So. You see 6 gate no gas. You make a ton of roach/ling. Toss can't possibly take his third, because he won't have enough sentries, and you will have a shitton of units. You will just fucking deny his third forever. He will have to wait forever. In the meantime, you are teching hard and fast to speed roaches, burrow, hydras, mutas, whatever, while taking a fourth base.
And really, by taht point in the game, you've won. You can't really do any 'timing' with 6 gate besides when warp gate first finishes. If a 6 gate timing isn't going to work (assuming proper reaction from zerg) at the standard 8:30 mark, how is it possibly going to work at the 10:00 mark? Makes no sense. By then you'll have a tonof hydras.
I don't really understand what you are saying. If toss is doing a sub-optimal build, you just end up ahead by responding as if he was playing optimally.
If Toss is doing a no gas 6 gate that's really bad and late, you don't need an upgrade. So what you do is see no gas, so you know what's up, you don't get the upgrade. You mass roach/ling, make no lair. Then, you are at about 90-110 supply. You realize, hey, my army way fucking bigger now, I'd stomp him if he moves out, because my econ has kicked in, and he has no tech that can take advantage of my hatch tech. So then you tech up, and get upgrades, lair, speed, et cetera.
For the same reason, you can't just 4 gate expand. You just die. Like you can't do it. For the same reasons, you can't just 6 gate no gas FFE and just play out.
I don't know why you're harping that it's a supoptimal build; it's a build NonY brought up on SotG after IEM Gaungzhao and has used several times on ladder to great success. The has the regular timing with his orignal sentries where he has a lot of energy and he uses it for one timine where protoss has a lot of FF's and tries to snipe the third. Instead of doing the hardcore just warping in stalkers/zealots to try to kill the guy, he continues probing and takes the double gas as he leaves and doesn't over commit into the third. I know you like to argue, but perhaps you can admit that a low masters zerg doesn't know the game as well as NonY, so you might take his word for it?
I mostly do it off 3gate now and it is very strong against zergs who have taken a 3rd going roach/ling. 6gate is still good against some zerg builds but it's pretty terrible against others and you've got no control over what they're doing, nor scouting to cut your losses and change builds at an early enough time.
Another important point is that I play from 2gate or 3gate expand, not FFE. I don't think FFE can pressure a 3rd very safely.
Do you forsee the meta game going back to 3gate expand for that reason? Or perhaps more exciting FFE 4gates?
On March 13 2012 21:03 Belial88 wrote: I never said 70 drones, I said 70 supply. Also, 70 supply at 8:00 minute means you have like somewhere in the upper 50's of drones actually mining, and the rest of the supply taken up by morphing drones, as well as queens and random lings.
The game I was talking about was Idra vs Oz on Dual Sight, but if you watch any of the DRG vs Genius finals he hit 70+ in all but one game (i think the one he didn't was daybreak).
Pretty solid. I would offer that you should in fact get an upgrade if he's going for a WG pressure build to insure you have a followup.
If you want to, sure, of course. I would even recommend it. I was just saying DONT get it if toss is doing a hardcore gateway all-in. Otherwise, do whatever you think is necessary.
Most people just go all-in with 6 gate, but there are protoss (especially on the NA server) who 6 gate while taking a double gas behind it to put together a pretty strong timing on the third with a lot of sentry energy and then transitions (usually to 3rd and blink with a slower colossus).
If someone is doing a sub-optimal bad play, and you react as if they were playing optimally, you will only end up ahead. That's like saying making roach/ling against a 4 gate is bad because Toss could theoretically just be going 4 gate expand.
No.
Toss goes 6 gate, you see no gas, so he won't ahve lots of sentries. Geysers take a while to warp in, and build up gas. So you get 1 gas, that's 30 seconds, then about another minute to basically get 1 sentry out. So it's like 2 minutes to get 2 sentries from 1 gas. So getting 2 gas late, as in past the 8:00 mark, means Toss won't have sentries for a long fucking time. He would've done better going 3 gas normal 5 gate sentry expand to third or something.
So. You see 6 gate no gas. You make a ton of roach/ling. Toss can't possibly take his third, because he won't have enough sentries, and you will have a shitton of units. You will just fucking deny his third forever. He will have to wait forever. In the meantime, you are teching hard and fast to speed roaches, burrow, hydras, mutas, whatever, while taking a fourth base.
And really, by taht point in the game, you've won. You can't really do any 'timing' with 6 gate besides when warp gate first finishes. If a 6 gate timing isn't going to work (assuming proper reaction from zerg) at the standard 8:30 mark, how is it possibly going to work at the 10:00 mark? Makes no sense. By then you'll have a tonof hydras.
I don't really understand what you are saying. If toss is doing a sub-optimal build, you just end up ahead by responding as if he was playing optimally.
If Toss is doing a no gas 6 gate that's really bad and late, you don't need an upgrade. So what you do is see no gas, so you know what's up, you don't get the upgrade. You mass roach/ling, make no lair. Then, you are at about 90-110 supply. You realize, hey, my army way fucking bigger now, I'd stomp him if he moves out, because my econ has kicked in, and he has no tech that can take advantage of my hatch tech. So then you tech up, and get upgrades, lair, speed, et cetera.
For the same reason, you can't just 4 gate expand. You just die. Like you can't do it. For the same reasons, you can't just 6 gate no gas FFE and just play out.
I don't know why you're harping that it's a supoptimal build; it's a build NonY brought up on SotG after IEM Gaungzhao and has used several times on ladder to great success. The has the regular timing with his orignal sentries where he has a lot of energy and he uses it for one timine where protoss has a lot of FF's and tries to snipe the third. Instead of doing the hardcore just warping in stalkers/zealots to try to kill the guy, he continues probing and takes the double gas as he leaves and doesn't over commit into the third. I know you like to argue, but perhaps you can admit that a low masters zerg doesn't know the game as well as NonY, so you might take his word for it?
I mostly do it off 3gate now and it is very strong against zergs who have taken a 3rd going roach/ling. 6gate is still good against some zerg builds but it's pretty terrible against others and you've got no control over what they're doing, nor scouting to cut your losses and change builds at an early enough time.
Another important point is that I play from 2gate or 3gate expand, not FFE. I don't think FFE can pressure a 3rd very safely.
Why 2 or 3gate expand?
At a time when almost all zergs are gonig gasless, why wouldn't 1gate expo be a better option? Is it because that early of an expansion lets zerg know that there are no additional gateways and they can still safely fast third?
I mean with 1gate expo speedling pressure seems like the primary drawback, so I'm just curious what the reason for 2-3gate expo is.
On March 13 2012 21:03 Belial88 wrote: I never said 70 drones, I said 70 supply. Also, 70 supply at 8:00 minute means you have like somewhere in the upper 50's of drones actually mining, and the rest of the supply taken up by morphing drones, as well as queens and random lings.
The game I was talking about was Idra vs Oz on Dual Sight, but if you watch any of the DRG vs Genius finals he hit 70+ in all but one game (i think the one he didn't was daybreak).
Pretty solid. I would offer that you should in fact get an upgrade if he's going for a WG pressure build to insure you have a followup.
If you want to, sure, of course. I would even recommend it. I was just saying DONT get it if toss is doing a hardcore gateway all-in. Otherwise, do whatever you think is necessary.
Most people just go all-in with 6 gate, but there are protoss (especially on the NA server) who 6 gate while taking a double gas behind it to put together a pretty strong timing on the third with a lot of sentry energy and then transitions (usually to 3rd and blink with a slower colossus).
If someone is doing a sub-optimal bad play, and you react as if they were playing optimally, you will only end up ahead. That's like saying making roach/ling against a 4 gate is bad because Toss could theoretically just be going 4 gate expand.
No.
Toss goes 6 gate, you see no gas, so he won't ahve lots of sentries. Geysers take a while to warp in, and build up gas. So you get 1 gas, that's 30 seconds, then about another minute to basically get 1 sentry out. So it's like 2 minutes to get 2 sentries from 1 gas. So getting 2 gas late, as in past the 8:00 mark, means Toss won't have sentries for a long fucking time. He would've done better going 3 gas normal 5 gate sentry expand to third or something.
So. You see 6 gate no gas. You make a ton of roach/ling. Toss can't possibly take his third, because he won't have enough sentries, and you will have a shitton of units. You will just fucking deny his third forever. He will have to wait forever. In the meantime, you are teching hard and fast to speed roaches, burrow, hydras, mutas, whatever, while taking a fourth base.
And really, by taht point in the game, you've won. You can't really do any 'timing' with 6 gate besides when warp gate first finishes. If a 6 gate timing isn't going to work (assuming proper reaction from zerg) at the standard 8:30 mark, how is it possibly going to work at the 10:00 mark? Makes no sense. By then you'll have a tonof hydras.
I don't really understand what you are saying. If toss is doing a sub-optimal build, you just end up ahead by responding as if he was playing optimally.
If Toss is doing a no gas 6 gate that's really bad and late, you don't need an upgrade. So what you do is see no gas, so you know what's up, you don't get the upgrade. You mass roach/ling, make no lair. Then, you are at about 90-110 supply. You realize, hey, my army way fucking bigger now, I'd stomp him if he moves out, because my econ has kicked in, and he has no tech that can take advantage of my hatch tech. So then you tech up, and get upgrades, lair, speed, et cetera.
For the same reason, you can't just 4 gate expand. You just die. Like you can't do it. For the same reasons, you can't just 6 gate no gas FFE and just play out.
I don't know why you're harping that it's a supoptimal build; it's a build NonY brought up on SotG after IEM Gaungzhao and has used several times on ladder to great success. The has the regular timing with his orignal sentries where he has a lot of energy and he uses it for one timine where protoss has a lot of FF's and tries to snipe the third. Instead of doing the hardcore just warping in stalkers/zealots to try to kill the guy, he continues probing and takes the double gas as he leaves and doesn't over commit into the third. I know you like to argue, but perhaps you can admit that a low masters zerg doesn't know the game as well as NonY, so you might take his word for it?
I mostly do it off 3gate now and it is very strong against zergs who have taken a 3rd going roach/ling. 6gate is still good against some zerg builds but it's pretty terrible against others and you've got no control over what they're doing, nor scouting to cut your losses and change builds at an early enough time.
Another important point is that I play from 2gate or 3gate expand, not FFE. I don't think FFE can pressure a 3rd very safely.
Why 2 or 3gate expand?
At a time when almost all zergs are gonig gasless, why wouldn't 1gate expo be a better option? Is it because that early of an expansion lets zerg know that there are no additional gateways and they can still safely fast third?
I mean with 1gate expo speedling pressure seems like the primary drawback, so I'm just curious what the reason for 2-3gate expo is.
From personal experience, a lot of 1gate expands make a few less sentries (and often dont take the second gas altogether), which in turn makes your pressure a tiny bit weaker. Also 2gate expand will never ever be broken by any amount of speeldlings (which can come off a delayed gas too), while 1gate has a small timing where it can be hurt (6minutes ish).
Excellent guide with tons of useful timings, scouting information and macro benchmarks.
Kudos for putting it together, very thorough
Glad you like it, but aren't you at a level where you already know all of this?
I'm assuming all of the diamond+ knew all of this. Maybe the benchmarking section might have been useful? I don't know. Let me know what you learned when you post guys. And I can't be all-knowing, surely there is stuff out there that I could add that I am not aware of.
Great guide for high diamond players like myself. Mainly the specifics about what to look for when scouting, and proper responses, as some of the all-ins require a special response beyond "have more stuff" . Definitely interesting to hear that hydras are probably needed vs blink stalker all-in, I am still tryin to hold with macro hatch, roach/ling, and flanking with small groups of lings. It hasn't gone well.
Benchmarks are nice because they provide a clear, verifiable data point as a goal to work toward. This is invaluable when trying to train your macro. Useful, yes, but mostly as a threshold to measure whether you should keep reading the guide or go practice macro first.
I'm also enjoying how much debate there is over evo and roach warren timings. i do it on feel: if I'm suspicious of air/dt tech, put down evo; if I suspect heavy pressure or an attack, I put down roach warren. 30 second difference in placement time really isn't what is under discussion, but everyone keeps trying to give specific times, in spite of the fact that most of the time, you just want them as late as you can get away with.
On March 13 2012 21:03 Belial88 wrote: I never said 70 drones, I said 70 supply. Also, 70 supply at 8:00 minute means you have like somewhere in the upper 50's of drones actually mining, and the rest of the supply taken up by morphing drones, as well as queens and random lings.
The game I was talking about was Idra vs Oz on Dual Sight, but if you watch any of the DRG vs Genius finals he hit 70+ in all but one game (i think the one he didn't was daybreak).
Pretty solid. I would offer that you should in fact get an upgrade if he's going for a WG pressure build to insure you have a followup.
If you want to, sure, of course. I would even recommend it. I was just saying DONT get it if toss is doing a hardcore gateway all-in. Otherwise, do whatever you think is necessary.
Most people just go all-in with 6 gate, but there are protoss (especially on the NA server) who 6 gate while taking a double gas behind it to put together a pretty strong timing on the third with a lot of sentry energy and then transitions (usually to 3rd and blink with a slower colossus).
If someone is doing a sub-optimal bad play, and you react as if they were playing optimally, you will only end up ahead. That's like saying making roach/ling against a 4 gate is bad because Toss could theoretically just be going 4 gate expand.
No.
Toss goes 6 gate, you see no gas, so he won't ahve lots of sentries. Geysers take a while to warp in, and build up gas. So you get 1 gas, that's 30 seconds, then about another minute to basically get 1 sentry out. So it's like 2 minutes to get 2 sentries from 1 gas. So getting 2 gas late, as in past the 8:00 mark, means Toss won't have sentries for a long fucking time. He would've done better going 3 gas normal 5 gate sentry expand to third or something.
So. You see 6 gate no gas. You make a ton of roach/ling. Toss can't possibly take his third, because he won't have enough sentries, and you will have a shitton of units. You will just fucking deny his third forever. He will have to wait forever. In the meantime, you are teching hard and fast to speed roaches, burrow, hydras, mutas, whatever, while taking a fourth base.
And really, by taht point in the game, you've won. You can't really do any 'timing' with 6 gate besides when warp gate first finishes. If a 6 gate timing isn't going to work (assuming proper reaction from zerg) at the standard 8:30 mark, how is it possibly going to work at the 10:00 mark? Makes no sense. By then you'll have a tonof hydras.
I don't really understand what you are saying. If toss is doing a sub-optimal build, you just end up ahead by responding as if he was playing optimally.
If Toss is doing a no gas 6 gate that's really bad and late, you don't need an upgrade. So what you do is see no gas, so you know what's up, you don't get the upgrade. You mass roach/ling, make no lair. Then, you are at about 90-110 supply. You realize, hey, my army way fucking bigger now, I'd stomp him if he moves out, because my econ has kicked in, and he has no tech that can take advantage of my hatch tech. So then you tech up, and get upgrades, lair, speed, et cetera.
For the same reason, you can't just 4 gate expand. You just die. Like you can't do it. For the same reasons, you can't just 6 gate no gas FFE and just play out.
I don't know why you're harping that it's a supoptimal build; it's a build NonY brought up on SotG after IEM Gaungzhao and has used several times on ladder to great success. The has the regular timing with his orignal sentries where he has a lot of energy and he uses it for one timine where protoss has a lot of FF's and tries to snipe the third. Instead of doing the hardcore just warping in stalkers/zealots to try to kill the guy, he continues probing and takes the double gas as he leaves and doesn't over commit into the third. I know you like to argue, but perhaps you can admit that a low masters zerg doesn't know the game as well as NonY, so you might take his word for it?
I mostly do it off 3gate now and it is very strong against zergs who have taken a 3rd going roach/ling. 6gate is still good against some zerg builds but it's pretty terrible against others and you've got no control over what they're doing, nor scouting to cut your losses and change builds at an early enough time.
Another important point is that I play from 2gate or 3gate expand, not FFE. I don't think FFE can pressure a 3rd very safely.
I do agree on this one. The only FFE version I sometimes use is the Kiwikaki double gateway style (Mid-Masters). It is able to snip a fast third, but sometimes gets destroyed by a simple roach defense. 2/3 Gate expand allows you to hit almost the same timings but much more safely and allows you to push for the third in a safer way.
What exactly do you mean by "0 gas into 1 gas" and how is that different from 1 gas? Does 1 gas means it 1 gas is finished right when the nexus is done and 0 gas into 1 gas means he is not mining from any geysers for a while, but then he throws down a third gas?
The easiest way for people to understand that you are saying the "Belial Answer" instead of the "Correct Answer" is to preface your post with the words: "In my opinion." And in my opinion, these two should be interchangable if you are giving someone DIRECT advice (for example: build a starport as soon as you can with your cloaked banshee build). But I'm off topic, and this doesn't really matter too much to me.
Thanks for this Belial. I tried this out, and I prefer my own reactionary timings; 3rd at 40ish supply, and as many queens as possible in the early game. But now I know a better lair timing, when to throw down roaches, and evo timings.
On March 13 2012 21:03 Belial88 wrote: I never said 70 drones, I said 70 supply. Also, 70 supply at 8:00 minute means you have like somewhere in the upper 50's of drones actually mining, and the rest of the supply taken up by morphing drones, as well as queens and random lings.
The game I was talking about was Idra vs Oz on Dual Sight, but if you watch any of the DRG vs Genius finals he hit 70+ in all but one game (i think the one he didn't was daybreak).
Pretty solid. I would offer that you should in fact get an upgrade if he's going for a WG pressure build to insure you have a followup.
If you want to, sure, of course. I would even recommend it. I was just saying DONT get it if toss is doing a hardcore gateway all-in. Otherwise, do whatever you think is necessary.
Most people just go all-in with 6 gate, but there are protoss (especially on the NA server) who 6 gate while taking a double gas behind it to put together a pretty strong timing on the third with a lot of sentry energy and then transitions (usually to 3rd and blink with a slower colossus).
If someone is doing a sub-optimal bad play, and you react as if they were playing optimally, you will only end up ahead. That's like saying making roach/ling against a 4 gate is bad because Toss could theoretically just be going 4 gate expand.
No.
Toss goes 6 gate, you see no gas, so he won't ahve lots of sentries. Geysers take a while to warp in, and build up gas. So you get 1 gas, that's 30 seconds, then about another minute to basically get 1 sentry out. So it's like 2 minutes to get 2 sentries from 1 gas. So getting 2 gas late, as in past the 8:00 mark, means Toss won't have sentries for a long fucking time. He would've done better going 3 gas normal 5 gate sentry expand to third or something.
So. You see 6 gate no gas. You make a ton of roach/ling. Toss can't possibly take his third, because he won't have enough sentries, and you will have a shitton of units. You will just fucking deny his third forever. He will have to wait forever. In the meantime, you are teching hard and fast to speed roaches, burrow, hydras, mutas, whatever, while taking a fourth base.
And really, by taht point in the game, you've won. You can't really do any 'timing' with 6 gate besides when warp gate first finishes. If a 6 gate timing isn't going to work (assuming proper reaction from zerg) at the standard 8:30 mark, how is it possibly going to work at the 10:00 mark? Makes no sense. By then you'll have a tonof hydras.
I don't really understand what you are saying. If toss is doing a sub-optimal build, you just end up ahead by responding as if he was playing optimally.
If Toss is doing a no gas 6 gate that's really bad and late, you don't need an upgrade. So what you do is see no gas, so you know what's up, you don't get the upgrade. You mass roach/ling, make no lair. Then, you are at about 90-110 supply. You realize, hey, my army way fucking bigger now, I'd stomp him if he moves out, because my econ has kicked in, and he has no tech that can take advantage of my hatch tech. So then you tech up, and get upgrades, lair, speed, et cetera.
For the same reason, you can't just 4 gate expand. You just die. Like you can't do it. For the same reasons, you can't just 6 gate no gas FFE and just play out.
I don't know why you're harping that it's a supoptimal build; it's a build NonY brought up on SotG after IEM Gaungzhao and has used several times on ladder to great success. The has the regular timing with his orignal sentries where he has a lot of energy and he uses it for one timine where protoss has a lot of FF's and tries to snipe the third. Instead of doing the hardcore just warping in stalkers/zealots to try to kill the guy, he continues probing and takes the double gas as he leaves and doesn't over commit into the third. I know you like to argue, but perhaps you can admit that a low masters zerg doesn't know the game as well as NonY, so you might take his word for it?
I mostly do it off 3gate now and it is very strong against zergs who have taken a 3rd going roach/ling. 6gate is still good against some zerg builds but it's pretty terrible against others and you've got no control over what they're doing, nor scouting to cut your losses and change builds at an early enough time.
Another important point is that I play from 2gate or 3gate expand, not FFE. I don't think FFE can pressure a 3rd very safely.
Wow pretty cool to get NoNy himself to support my argument - Take that Arisen ^^
belial!
This thread is awesome. I found the history of drone saturation and maynarding very interesting.
Also, the the structure of this guide is very easy to read. Looking forward to your next one!
I think you just beat me on ladder earlier, in a zvz on antiga right? God i hate that map. i think i went lair too fast because i was ahead in drones the whole time but you killed a bunch when I maynarded them for safety reasons - i shouldve just left them at the third.
Anyways, please, please, please EndofLine, you need to message Zatic and tell him that. He thinks my guide looks extremely unorganized, he prefers a different formatting for guides and says people don't want to click a bunch of shit. I told him people don't want to look at a wall of text.
On March 14 2012 08:23 Xizorz wrote: I'm confused as to how you are getting 3 overlords to the toss base by 8 minutes. Are you sending your 1st created overlord over?
I've always kept that over my natural to prevent some pylon nonsense.
Well, belial has included a replay, so you can just see how he's doing it there. It's definitely most common and standard to send your first created overlord to your nat though, I would consider it pretty risky to send it to your opponent, at least until your nat is up.
From personal experience, a lot of 1gate expands make a few less sentries (and often dont take the second gas altogether), which in turn makes your pressure a tiny bit weaker. Also 2gate expand will never ever be broken by any amount of speeldlings (which can come off a delayed gas too), while 1gate has a small timing where it can be hurt (6minutes ish)
1 gate expand will not be hurt by a gasless pool. All there is, is that 1 gate expand has a small timing where it can be hurt, around 6 minutes, by a speedling expand, but if zerg does not go gas first, he will not have speed in time to pressure a 1 gate expand. All Toss has to do is scout and see if Zerg went gas first or pool first, and then go 1/3 gate expand accordingly.
I'm confused as to how you are getting 3 overlords to the toss base by 8 minutes. Are you sending your 1st created overlord over?
I've always kept that over my natural to prevent some pylon nonsense.
So do I.
Beginning Overlord - To farthest point (so say on Shakuras, I will have this overlord go to one of the bases, and then go past it. You really want it so both overlords to sac into main approach from opposite sides of the base, and since this is the first overlord, I make it go farthest) 1st Overlord - Watches Natural until about 16 supply, when I bring drones down, then I send it to gas. This overlord is safe to cross over even the middle of the map, or in front of the Toss base. 2nd Overlord - Send this one on the nearest side of Toss base. 3rd Overlord - Watches over my third. (sometimes 2 and 3 are swapped if map is smaller).
I think TDA cross positions is one of the few places where you may have trouble, but the 3 overlords will at least be near the Toss base, so anything like stargate will be made obvious by them.
Hi Belial,
What exactly do you mean by "0 gas into 1 gas" and how is that different from 1 gas? Does 1 gas means it 1 gas is finished right when the nexus is done and 0 gas into 1 gas means he is not mining from any geysers for a while, but then he throws down a third gas?
There's a small note in that section where I explained this. But in that section, I was talking about natural's gas. Toss has to take his 1st and 2nd gas in main at ~22 supply when going FFE 1 gate cyber, they just have to. The only time Toss doesn't get those 2 gas in main at same time at 22 before the natural nexus completes, is when going FFE 2 gate before cyber, which is obvious (2 gates,no cyber, no gas taken in main super early).
So when I said 0 gas, it means 2 gas only taken in main, and that the natural is empty. When I said 0 gas into 1 gas, it means 0 gas at natural for a long time, and then 1 gas taken around maybe 7:00-7:30. You can replace 0 with 2, 1 with 3, and 2 with 4 total gas, if that makes sense.
There's a note in that section above the gas spoilered parts where I cleared this up.
The easiest way for people to understand that you are saying the "Belial Answer" instead of the "Correct Answer" is to preface your post with the words: "In my opinion." And in my opinion, these two should be interchangable if you are giving someone DIRECT advice (for example: build a starport as soon as you can with your cloaked banshee build). But I'm off topic, and this doesn't really matter too much to me.
Everything I say is obviously my point of view. Like.... it's obvious it's coming out of my mouth, and no where else. My username isn't TL-Godbot. And everything everyone says here is they're own opinion. It's a forum.
Thanks for this Belial. I tried this out, and I prefer my own reactionary timings; 3rd at 40ish supply, and as many queens as possible in the early game. But now I know a better lair timing, when to throw down roaches, and evo timings.
I'm glad you like it, but I'm going to ask why you do what you do.
3rd at 40? That seems a bit late. Now, the issue is that no one has actually done any testing on when is the best time to take a fast third, and I did a bit of testing in evo chamber and it seemed to imply that ~30 is the best time to take it, but really, who knows when the best time to take a third is. Pro players seem to grab it between 25-35, but pro koreans were going extractor trick every single game for a long time, and many of them still go 15 hatch instead of 14 hatch, so just because pros do something doesn't make it right.
But why are you making as many queens as possible at the start? That just kills your econ, queens cost so much! We all learned by now that the Spanishiwa build of 4 queens is horrible, because it just absolutely trashed your econ, as well as map control. Why would you make so many queens?
Mind if you provide your benchmark of what your supply is at 8:00?
On a side note, I used to mass queens when going fast third when I first started doing it in diamond, when I was first learning it. It was a fun way to play it, any bad diamond toss push came and I just had mass queens to push them back, and it was quite effective at low level play. It was fun ^^
Belial I have read just about everything in the guide but I don't see any advice on 4 gate +1 zealot pressure. I run into this a heck of a lot.
Its basically a +1 attack before warpgate then no gas at natural until like 6/7 minutes in. Its pretty obvious to see coming but it has so many transitions afterwards I have trouble figuring out what hes doing.
Ive played vs 4 gate into blink +2 allin, 4 gate into DT, 4 gate into +2 blink immortal allin, 4 gate into 1/1 8 gate, ect ect.
I know i can handle the pressure as it comes but how to scout the transition with a delayed lair (after seeing no gas in his nat).
The easiest way for people to understand that you are saying the "Belial Answer" instead of the "Correct Answer" is to preface your post with the words: "In my opinion." And in my opinion, these two should be interchangable if you are giving someone DIRECT advice (for example: build a starport as soon as you can with your cloaked banshee build). But I'm off topic, and this doesn't really matter too much to me.
Everything I say is obviously my point of view. Like.... it's obvious it's coming out of my mouth, and no where else. My username isn't TL-Godbot. And everything everyone says here is they're own opinion. It's a forum.
Thanks for this Belial. I tried this out, and I prefer my own reactionary timings; 3rd at 40ish supply, and as many queens as possible in the early game. But now I know a better lair timing, when to throw down roaches, and evo timings.
I'm glad you like it, but I'm going to ask why you do what you do.
3rd at 40? That seems a bit late. Now, the issue is that no one has actually done any testing on when is the best time to take a fast third, and I did a bit of testing in evo chamber and it seemed to imply that ~30 is the best time to take it, but really, who knows when the best time to take a third is. Pro players seem to grab it between 25-35, but pro koreans were going extractor trick every single game for a long time, and many of them still go 15 hatch instead of 14 hatch, so just because pros do something doesn't make it right.
But why are you making as many queens as possible at the start? That just kills your econ, queens cost so much! We all learned by now that the Spanishiwa build of 4 queens is horrible, because it just absolutely trashed your econ, as well as map control. Why would you make so many queens?
Mind if you provide your benchmark of what your supply is at 8:00?
On a side note, I used to mass queens when going fast third when I first started doing it in diamond, when I was first learning it. It was a fun way to play it, any bad diamond toss push came and I just had mass queens to push them back, and it was quite effective at low level play. It was fun ^^
The issue is people don't understand that what comes out of a person's mouth is always their opinion.
Either way: I get queens when I have no larva in the early game. That way I don't delay my droning; however I am delaying my gas, and expand. The reason I do this is because I feel that having good creep spread is vital in the early game, and transfuse is vital in the late game. This also means that I'm not delaying my expands' queens, and if I lose queens to anything, I will have more. Lastly, it helps me a lot vs phoenix play (I didn't have a timing for my evo yet).
I'll give you a benchmark in the next day or so; have never used them, so I have to check.
^ As I stated earlier, I get frustrated too when I see lower level people say incorrect stuff, so I can understand why some higher level people don't like me. But the ones that talk to me usually understand I'm not here to do anything wrong, I'm just another low level scrub. It's just the ones that I guess don't care to talk to me that stay stuck in their ways. Some people might be surprised to learn I'm willing to admit I'm wrong, and it doesnt take much to convince me that, or that I'm a nice guy (although I'm still a bit of a troll).
Either way: I get queens when I have no larva in the early game. That way I don't delay my droning; however I am delaying my gas, and expand. The reason I do this is because I feel that having good creep spread is vital in the early game, and transfuse is vital in the late game. This also means that I'm not delaying my expands' queens, and if I lose queens to anything, I will have more. Lastly, it helps me a lot vs phoenix play (I didn't have a timing for my evo yet).
So what happens if Toss plays greedy, like he goes for fast third? It seems like you would be really weak to that. Or what if Toss does a techy push like blink all-in or immortal/sentry? Your drone count would end up too low.
I don't know if creep spread is 'vital' - it's helpful, sure, but unlike, say, ZvT, I think it's definitely ambiguous as to how important creep spread is, and it's not clear cut if creep spread is super important. Like, you might want to just inject, or delay queens for drones, whereas in ZvT, you *must* get creep going.
I don't know about transfuse being vital though. Lategame you can pump mass queens very quickly off of your many bases.
It sounds like you are just setting yourself up to be better against certain styles of play, and worse against others.
On March 13 2012 21:03 Belial88 wrote: I never said 70 drones, I said 70 supply. Also, 70 supply at 8:00 minute means you have like somewhere in the upper 50's of drones actually mining, and the rest of the supply taken up by morphing drones, as well as queens and random lings.
The game I was talking about was Idra vs Oz on Dual Sight, but if you watch any of the DRG vs Genius finals he hit 70+ in all but one game (i think the one he didn't was daybreak).
Pretty solid. I would offer that you should in fact get an upgrade if he's going for a WG pressure build to insure you have a followup.
If you want to, sure, of course. I would even recommend it. I was just saying DONT get it if toss is doing a hardcore gateway all-in. Otherwise, do whatever you think is necessary.
Most people just go all-in with 6 gate, but there are protoss (especially on the NA server) who 6 gate while taking a double gas behind it to put together a pretty strong timing on the third with a lot of sentry energy and then transitions (usually to 3rd and blink with a slower colossus).
If someone is doing a sub-optimal bad play, and you react as if they were playing optimally, you will only end up ahead. That's like saying making roach/ling against a 4 gate is bad because Toss could theoretically just be going 4 gate expand.
No.
Toss goes 6 gate, you see no gas, so he won't ahve lots of sentries. Geysers take a while to warp in, and build up gas. So you get 1 gas, that's 30 seconds, then about another minute to basically get 1 sentry out. So it's like 2 minutes to get 2 sentries from 1 gas. So getting 2 gas late, as in past the 8:00 mark, means Toss won't have sentries for a long fucking time. He would've done better going 3 gas normal 5 gate sentry expand to third or something.
So. You see 6 gate no gas. You make a ton of roach/ling. Toss can't possibly take his third, because he won't have enough sentries, and you will have a shitton of units. You will just fucking deny his third forever. He will have to wait forever. In the meantime, you are teching hard and fast to speed roaches, burrow, hydras, mutas, whatever, while taking a fourth base.
And really, by taht point in the game, you've won. You can't really do any 'timing' with 6 gate besides when warp gate first finishes. If a 6 gate timing isn't going to work (assuming proper reaction from zerg) at the standard 8:30 mark, how is it possibly going to work at the 10:00 mark? Makes no sense. By then you'll have a tonof hydras.
I don't really understand what you are saying. If toss is doing a sub-optimal build, you just end up ahead by responding as if he was playing optimally.
If Toss is doing a no gas 6 gate that's really bad and late, you don't need an upgrade. So what you do is see no gas, so you know what's up, you don't get the upgrade. You mass roach/ling, make no lair. Then, you are at about 90-110 supply. You realize, hey, my army way fucking bigger now, I'd stomp him if he moves out, because my econ has kicked in, and he has no tech that can take advantage of my hatch tech. So then you tech up, and get upgrades, lair, speed, et cetera.
For the same reason, you can't just 4 gate expand. You just die. Like you can't do it. For the same reasons, you can't just 6 gate no gas FFE and just play out.
I don't know why you're harping that it's a supoptimal build; it's a build NonY brought up on SotG after IEM Gaungzhao and has used several times on ladder to great success. The has the regular timing with his orignal sentries where he has a lot of energy and he uses it for one timine where protoss has a lot of FF's and tries to snipe the third. Instead of doing the hardcore just warping in stalkers/zealots to try to kill the guy, he continues probing and takes the double gas as he leaves and doesn't over commit into the third. I know you like to argue, but perhaps you can admit that a low masters zerg doesn't know the game as well as NonY, so you might take his word for it?
I mostly do it off 3gate now and it is very strong against zergs who have taken a 3rd going roach/ling. 6gate is still good against some zerg builds but it's pretty terrible against others and you've got no control over what they're doing, nor scouting to cut your losses and change builds at an early enough time.
Another important point is that I play from 2gate or 3gate expand, not FFE. I don't think FFE can pressure a 3rd very safely.
Wow pretty cool to get NoNy himself to support my argument - Take that Arisen ^^
This thread is awesome. I found the history of drone saturation and maynarding very interesting.
Also, the the structure of this guide is very easy to read. Looking forward to your next one!
I think you just beat me on ladder earlier, in a zvz on antiga right? God i hate that map. i think i went lair too fast because i was ahead in drones the whole time but you killed a bunch when I maynarded them for safety reasons - i shouldve just left them at the third.
Anyways, please, please, please EndofLine, you need to message Zatic and tell him that. He thinks my guide looks extremely unorganized, he prefers a different formatting for guides and says people don't want to click a bunch of shit. I told him people don't want to look at a wall of text.
Yep, that was me - I love antiga ^_^ I thought we were around the same with drones? I will go watch the replay, you may be right. Is Zatic a mod? Either way, I will tell him I think your style of guide makes it easy to find the info you are seeking when re reading the guide.
Excellent guide with tons of useful timings, scouting information and macro benchmarks.
Kudos for putting it together, very thorough
Glad you like it, but aren't you at a level where you already know all of this?
I'm assuming all of the diamond+ knew all of this. Maybe the benchmarking section might have been useful? I don't know. Let me know what you learned when you post guys. And I can't be all-knowing, surely there is stuff out there that I could add that I am not aware of.
Oh I knew the scouting information and some timings. But the macro benchmarks are just really nice to allow people to have something to work to. Myself I use larvae-spawning gap reduction as a macro benchmark, but using supply can be an easy to understand way for people below and on diamond to use as a goal. Since training macro is quite boring most of the time, but it's what most of the game hinges on and its where most of the lower level people are stuck at. That's why I like those benchmarks.
Just because I know the stuff doesn't mean I can't compliment the thorough information you put together .
On March 15 2012 02:32 shadowrunner99 wrote: These benchmarks are really useful. TY Is there a thread that is just full of benchmarks for all the different matchups and build?
This doesn't particularly work for anything other than FFE v 3base zerg, mostly because this is the only situation where early game variants are devoid and unit loss doesn't occur before a specified time. That's why many tosses are transitioning out of 2base blind all-in builds vs zerg because zerg knows that if they can macro to 70 drones at 8 minute mark everytime they will beat toss of similar skill micro everytime if toss 2base all-ins.
EDIT: Unless you do some sort of non-reactive build or non-scouting builds, of course.
On March 15 2012 22:24 Advocado wrote: Is 6-7 gate easier to hold with burrow/tunneling claws or does that come too late?
6-7 gate hits just about when lair finishes, around 8:00-8:30, and tunneling claws take 1:50 to research. But it's not uncommon for the pressure to last for a while, sometimes even until burrow finishes.
So was Maynard pretty much the God of BW during his zenith? Seems like he mustve been fucking dominant if he was the one innovating seemingly simple shit such as transferring workers
On March 15 2012 23:21 FallDownMarigold wrote: So was Maynard pretty much the God of BW during his zenith? Seems like he mustve been fucking dominant if he was the one innovating seemingly simple shit such as transferring workers
He was actually called "the god of starcraft" and he was probably the best american player on KALI for a while. Froz was arguably better in the b.net era though.
Thanks for taking the time to write all that down. I'm just a newb zergling but your information gives me something to focus on while I try to lift myself out of the bronze pit.
I really liked your zvz tips too... really helped me to understand what was going on.
If you had any suggestions on how to hold off a 4gate........
i think there's something wrong with your 8 min supply benchmarks cause i hit around 78 as an average and i get supply blocked like every game so it should be possible to have more than that easily.
Assuming I don't drone scout, which a lot of the pros seem to decide against, what should I do if my lings get to my opponent's base (having been delayed dealing with a pylon block at the natural) and I find out they haven't FFE'd? By that time my third will already be pretty close to being finished and I'm not really sure whether I should cancel it or what. It seems pretty hard to keep it alive against any kind of pressure from protoss off one base without making a bunch of lings pretty quickly.. but maybe it's worth taking the hit to my drone count to keep it alive?
I'm a Platinum Terran (that's my main race) and I guess Gold Zerg, and on my second attempt of doing the fast third build, I hit 66 supply at 8:00 even though I supply blocked myself....are Diamonds really that bad at build execution?
EDIT: I had 73 my third time. I suppose your supply counts factor in some lost units....about how much resources lost are you estimating?
Great Guide, I switched to Z from Master Terran level, and I was always so confused at wtf the P is doing. Read the guide, then first game in the P was going 4 gas with empty base............ So I put my faith in the guide made blind spores and sure enough the Phxs and the void rays were quite sad....... MAUAHAHAHAH knowledge is power.
Why would you want more then 60 drones at the 8 minute mark? That's 16 per base on minerals with 4 fully saturated gas, that's more then enough in my opinion.
On March 16 2012 12:12 Spaceboy wrote: Assuming I don't drone scout, which a lot of the pros seem to decide against, what should I do if my lings get to my opponent's base (having been delayed dealing with a pylon block at the natural) and I find out they haven't FFE'd? By that time my third will already be pretty close to being finished and I'm not really sure whether I should cancel it or what. It seems pretty hard to keep it alive against any kind of pressure from protoss off one base without making a bunch of lings pretty quickly.. but maybe it's worth taking the hit to my drone count to keep it alive?
you should scout him. it doesn't make a difference if it's pros or bronzes, if they don't scout they are effectively gambling. don't think everything pro gamers do is legit play that you should copy.
Is 6-7 gate easier to hold with burrow/tunneling claws or does that come too late?
Yea, it comes way too late, and a strong, 8:30 6 gate will basically autowin if you go standard lair timing, you just won't have enough. If you see no gas at the toss base, you want to start pumping units at about 8:00. If you do this, you should be able to have enough out by about 9:30, to push toss back, and tech to lair. You'll have mass speedroaches before he has anything like blink, robo, etc, and he'll just die.
Hi Belial,
Thanks for taking the time to write all that down. I'm just a newb zergling but your information gives me something to focus on while I try to lift myself out of the bronze pit.
I really liked your zvz tips too... really helped me to understand what was going on.
If you had any suggestions on how to hold off a 4gate........
Cheers!
Glad you like it! This guide is probably more aimed towards Diamond players, with some additional information that I hope Masters can use. At bronze level, I would recommend you just play on 2 base lair, just see the Toss went FFE, and 100% drone up your 2 bases, get lair at 50, and not make any units until 60 supply, and then go mass speedroaches to secure third, taking the third earlier and earlier until you get a feel with what you can get away with. There are bigger issues for bronze, and I don't think bronze macro would allow you to take advantage of fast third more than it would hurt you.
I mean you really have to be pushing 50+ drones by the 8:00 mark, or else it's just hurting more than helping. Otherwise you could just play 2 base and have lair instead.
As for 4 gate, sac an overlord at 5:40 into toss base if toss hasn't expanded yet (remember you sent that original overlord you start with to a base, and if it's not toss base, you redirect it asap). But you should know it's 4 gate before then by no sentries, no 2nd gas taken by toss. You want to cut drones at about 30-35 supply. Personally, when I see gateway expands, I get ling speed asap, and then leave 1 drone in gas in case of pressure, so I can get a roach warren and roaches asap. On maps with chokes, you can rely more on 2-4 spines and lings.
There's a lot more written on the subject by a lot better players though. Search on it.
i think there's something wrong with your 8 min supply benchmarks cause i hit around 78 as an average and i get supply blocked like every game so it should be possible to have more than that easily.
I struggle to hit past 70 supply. I know I'm pretty bad, but if you hit 78 as average and get supply blocked, you must be taking your gas super late or doing something greedy. Provide rep?
I'm a Platinum Terran (that's my main race) and I guess Gold Zerg, and on my second attempt of doing the fast third build, I hit 66 supply at 8:00 even though I supply blocked myself....are Diamonds really that bad at build execution?
EDIT: I had 73 my third time. I suppose your supply counts factor in some lost units....about how much resources lost are you estimating?
Is that against computer? Playing against AI is a lot easier than playing against an opponent whom you have to sac overlords, move overlords against, check for probes, deal with zealot or stalker pressure, et cetera.
It's kind of hard to get those overlord timings right when 2 overlords need to die right the same time you have 3 injects on 3 hatches going on.
Great Guide, I switched to Z from Master Terran level, and I was always so confused at wtf the P is doing. Read the guide, then first game in the P was going 4 gas with empty base............ So I put my faith in the guide made blind spores and sure enough the Phxs and the void rays were quite sad....... MAUAHAHAHAH knowledge is power.
ahahaha, great! Really cool feedback, people say they like my guides but I rarely hear someone say they actually applied it. Very cool!
You did win right? For a long time when I played as a diamond zerg, I had these situations in ZvP where I was way ahead, like after holding double stargate, and lose because I didn't play the match-up right, as Toss just got 3 base and massed colossi. Oh the infestor era.... I'm sure if they re-buffed the infestor, no one would use it because we've found better ways to play now.
Why would you want more then 60 drones at the 8 minute mark? That's 16 per base on minerals with 4 fully saturated gas, that's more then enough in my opinion.
16 drones per 4 bases mining minerals, with gas taken at every base, would mean you would stockpile a shitton of gas, even if you were going pure mutas. You need wayyyy more drones as zerg.
And certain timings, like blink stalker or immortal/sentry, will KILL you if you only have 60 drones (and you only need 3-4 gas too). Toss R Strong.
But I say 60 supply, not drones. I say it multiple times - supply, supply, supply. What's with everyone thinking I'm saying drones? Some high masters challenged me about that and played me, and I told him in game, it's supply, supply, supply, not drone count. At 8:00 you will have maybe 50 actually mining drones. The other 20 supply is lings, queens, and about 15 is held up in morphing eggs.
And it's entirely posssible that zealot pressure or stalker pressure comes, in which case you need to make 10 supply worth of lings. That's not worth to do at all.
KingOctavius and TheV, thanks! I'm glad it's useful, please let me know when you actually apply the information ^^
Assuming I don't drone scout, which a lot of the pros seem to decide against, what should I do if my lings get to my opponent's base (having been delayed dealing with a pylon block at the natural) and I find out they haven't FFE'd? By that time my third will already be pretty close to being finished and I'm not really sure whether I should cancel it or what. It seems pretty hard to keep it alive against any kind of pressure from protoss off one base without making a bunch of lings pretty quickly.. but maybe it's worth taking the hit to my drone count to keep it alive?
If you are having trouble because you don't drone scout, maybe you should start drone scouting? Plenty of pros do drone scout in ZvP. I don't think I've heard anyone explicitly say it was wrong to drone scout either. Some people are just greedy. You don't have to be greedy yet, you aren't at that level of play.
I mean, daybreak, ZvP? Is Toss really going to gateway expand on that map? Zerg just takes fast third and drones to 70 supply even if you go 3 gate on that map!
Your lings should arrive at the Toss base before you should be planting the third down anyways, too. You should be plating around 25-35 supply, not 20.
And 4 gate will obviously autowin if you went fast third blindly. Sounds like you are making a lot of trouble for yourself when tyou could just drone scout.
When I see gateway play (and I drone scout), personally, I react by getting ling speed asap, try to get the hatch down asap (possibly even hatch first), then I remove all but 1 from gas after speed. When I see Toss expand, my plan is to take a third - so I throw down a third ~35-40 supply, and roach warren ~35-45 supply (depending on map size). Then, I mass a shitton of roaches at about 45-55 supply, depending on map size, then go back to adding drones, adding units as necessary as to Toss pushing out. I grab 2nd gas, get lair, and try to be as greedy as possible while making as many roaches as possible (since sentries will be the main problem). Then I get mutas as soon as toss backs off.
You could also just drone the fuck up on 2 bases, get lair, then take third, with roach speed starting. If toss pushes out, you'll have roach speed when he gets to third. Darkforce, a pro zerg much higher level than me, among many others, have said this is what they do. Maybe I play scrubs and is why I can get away with fast third, but I like to be greedy.
A lot of information spit out, but I can't see many of the numbers you say to be completely accurate. 1 supply block is equivalent to losing 10 drones? So if 1 supply block = running 10 drones into the toss army for free lol
^ Yes. If you get supply blocked at, say, 50 supply, you will be about 10 drones short at the 8:00 mark. That's what the numbers are coming from, how that supply block or overlord timing will manifest at the 8:00 mark.
really great guide,hope this ill solve my problem with this match up(iam high dia),because i have a lot of problems with opennings and dealing with early push, iam going to try it right now,thanks a lot
Assuming I don't drone scout, which a lot of the pros seem to decide against, what should I do if my lings get to my opponent's base (having been delayed dealing with a pylon block at the natural) and I find out they haven't FFE'd? By that time my third will already be pretty close to being finished and I'm not really sure whether I should cancel it or what. It seems pretty hard to keep it alive against any kind of pressure from protoss off one base without making a bunch of lings pretty quickly.. but maybe it's worth taking the hit to my drone count to keep it alive?
If you are having trouble because you don't drone scout, maybe you should start drone scouting? Plenty of pros do drone scout in ZvP. I don't think I've heard anyone explicitly say it was wrong to drone scout either. Some people are just greedy. You don't have to be greedy yet, you aren't at that level of play.
I mean, daybreak, ZvP? Is Toss really going to gateway expand on that map? Zerg just takes fast third and drones to 70 supply even if you go 3 gate on that map!
Your lings should arrive at the Toss base before you should be planting the third down anyways, too. You should be plating around 25-35 supply, not 20.
And 4 gate will obviously autowin if you went fast third blindly. Sounds like you are making a lot of trouble for yourself when tyou could just drone scout.
When I see gateway play (and I drone scout), personally, I react by getting ling speed asap, try to get the hatch down asap (possibly even hatch first), then I remove all but 1 from gas after speed. When I see Toss expand, my plan is to take a third - so I throw down a third ~35-40 supply, and roach warren ~35-45 supply (depending on map size). Then, I mass a shitton of roaches at about 45-55 supply, depending on map size, then go back to adding drones, adding units as necessary as to Toss pushing out. I grab 2nd gas, get lair, and try to be as greedy as possible while making as many roaches as possible (since sentries will be the main problem). Then I get mutas as soon as toss backs off.
You could also just drone the fuck up on 2 bases, get lair, then take third, with roach speed starting. If toss pushes out, you'll have roach speed when he gets to third. Darkforce, a pro zerg much higher level than me, among many others, have said this is what they do. Maybe I play scrubs and is why I can get away with fast third, but I like to be greedy.
Thanks very much for the reply. I have actually started drone scouting after this happened to me (as I figure in general I'm too low level for any slight economic compromise to prove significant so I should just play safer),
I was just wondering as I've seen players such as Stephano and Idra not drone scout (though this could be map dependent I guess) and it just struck me that there must be some way of progressing from there that isn't totally crippling or you'd think people would just blindly 4 gate them and get free wins. That said I'm pretty terrible so as you say I may have just screwed up the build (I think what may have happened is that I took the third first due to the pylon block, which I know you advise against anyway) and maybe I was just in a situation they'd never get into.
On another note I have definitely seen Ret use an 11 overpool opening and though I have been dissuaded from doing this by your arguments my guess is that there must be some reasonable rationale for this as I have enough respect for Ret that I don't think he'd be doing something that's objectively terrible.. especially when it comes to getting an edge economically
That said, great guide! (especially the scouting section which is really amazing.. everyone always mentions the need for decent scouting but it's so rare for someone to put down a concrete list of what you need to look for to actually peg someone on a build). Keep up the good work
11 pool is defintely viable, but not for macro games, and its a coinflip. Pros do coinfli builds too. But they only do it once in a while, rarely, and against an oponent they have studied and prepared against. You won't see ret do it consistently on the ladder against random opponents. And you sure it was 11 overpool and not 10 pooll?
^ Actually my bad it's not an overpool, it's actually just a kinda standard 11 pool (with ovi at 9). You can see him do it in this episode of the Artosis Hour. You're right though in that I haven't seen him do it a million times or anything so it could just be a one off!
Evo chamber should be grabbed at 7:00 at earliest if you see gas at the Toss natural, and still aren't exactly sure what Toss is up to. Any earlier than 7:00 is too early.
Also note, that if you see no gas at the Toss natural, indicating a 6 gate +1 super early all-in, you should not get an upgrade. Upgrades take a long time to complete - 160 seconds! Add 35 for evo build time, that's 185 seconds, or 3 minutes 5 seconds, for an upgrade to complete from when you throw down the evo. Since no gas gateway all-ins can hit from 8:00-9:00, that upgrade would be worthless if it won't be done until 10:05 at earliest, and may even contribute to your loss (if not actually cost you the game). So if you go for upgrades, be sure that Toss isn't doing such super fast gateway timing before 10:05.
Roach warren should be taken at 7:30. You can take it earlier if you see no gas at the Toss natural, or scout early warp gate pressure, and make it at 7:00 instead. If Toss makes any sort of early zealot aggression, like double gate, you are better off using lings - don't make a roach warren earlier just to deal with such pressure. However, if you see lots of gas at Toss natural, or a stargate opener, you can take your roach warren later.
Do you have any comments on preference of upgrades? I know Stephano likes to get +1 melee for lategame purposes while I personally like getting +1 range for midgame control. Do you change your upgrade based on certain gas timings taking guesses as to how you feel the lategame will play out based on the midgame? Is double evo at 7:30 mark just suicide most of the time? I have personally been trying Lair into double evo when I see 2 gas being taken and mined with macro hatch. Is the second evo to greedy that early?
It is difficult for me to tell since my macro mechanics are pretty shit and it is hard for me to see what build to "aim" for.
Brown does an 8 gate timing. He had zero gas taken, so right there lucky should have been suspicious of a gateway all-in (like I've said, zero gas 99% means mass gate all-in at 8:30, 1% stupidly fast third behind zero units, which isn't really viable, except maybe on a map like entombed).
Brown pushed out at 8:00 with minimal units, and had his first warp-in at 8:30, and was shooting shit down at 8:50. It was at 8:50 basically that Lucky lost, when too many stalkers were shooting at too few stalkers.
Also to note, Lucky was at 63 of 70, and had started lair. It is strange, but Lucky made an evo chamber, when he saw no gas (which is okay, but he could have cut it). He made his roach warren at 7:00 instead of 7:30+. Some people here have said certain 4 gate pressure can kill you if you make a roach warren as late as 7:30 instead of 7:00 (personally, I've had no trouble holding 2 gate before core, +1 zealot, and 4 gate pressure using just zerglings, but that's just me). Regardless, I don't think it was bad to make a 7:00 roach warren upon seeing no gas.
One thing that's interesting - at 8:00, lucky only had 47 drones. He made half lings, half drones after the 7:00 mark. I'm really not sure why he made so many lings, but he definitely had a very low drone count.
I think applying what I wrote about in the guide, you can apply what happened in this game and use it to better your own play. Lucky could have cut that evo out, he basically autolost for making lair against no-gass Toss, and he should've droned purely up to 8:00 instead of making half lings/half drone (besides making a few reactionary lings to deal with stalker/zealot pressure).
But it's also clear that Lucky definitely doesn't have the same level of macro as players like DRG and Nestea and Idra.
Do you have any comments on preference of upgrades? I know Stephano likes to get +1 melee for lategame purposes while I personally like getting +1 range for midgame control. Do you change your upgrade based on certain gas timings taking guesses as to how you feel the lategame will play out based on the midgame? Is double evo at 7:30 mark just suicide most of the time? I have personally been trying Lair into double evo when I see 2 gas being taken and mined with macro hatch. Is the second evo to greedy that early?
It is difficult for me to tell since my macro mechanics are pretty shit and it is hard for me to see what build to "aim" for.
Thanks.
Well, I don't know if I'm qualified on this. This is a very basic guide, on something that most masters+ already know about, so it's not like I'm saying anything new there. This isn't like a guide on "How to play mutas in ZvP' or "my 12 minute timing attack ZvT" or "Ultras in ZvZ". it's just everything put together that some have done, but no one has done in a way that wasn't complete shit cough skoozy cough.
That said, I have done a lot of testing on upgrades. So here's my thoughts (you can check my Upgrades on Mutalisks thread, as well as multiple comments I've made about it, I think I've made a few other upgrade threads as well).
First off, if I see zero gas at the Toss natural, I don't make any upgrades. If toss is doing a 6/7/8 gate all-in, the upgrade won't help at all (as I stated in the Roach Warren and Evo Chamber section of the OP). Evo is 35 seconds, upgrade is 160, that's 185 seconds, or 3 minutes 5 seconds. 7:00 + 185 seconds, that means the upgrades kicks in at 10:05.
So when you start an upgrade, you need to ask yourself "Will the upgrade finish in time?". Against a 7 gate all-in that will be inside your base at 8:50, no - you will have GGd out by 9:00. Against blink stalker all-in, it's questionable. You won't have it at first, but with a spine, you can maybe buy that 30 seconds, and then have it. Against sentry/immortal? Yea, you should definitely get it, since lings are kind of useless in that situation, but hydras are king. If Toss opens stargate, getting upgrades can always be useful, as you can get on that ground upgrade ahead of Toss. But, you know, it's up to you. I'm actually a big advocate of NOT getting upgrades in most match-ups, and preferring to mass units or tech up to infestors or hive much quicker.
You also have to wonder about critical hit numbers.
So, for example, +1 carapace. I largely think carapace is useless in most cases, especially in ZvP, when Toss is always racing attack upgrades. I will often go 0 carapace in ZvP, as it doesnt often change critical hit numbers (+3 zealots will always rape lings).
But for that first upgrade? If Toss is putting on gateway pressure, I'll get ranged attack. Roaches benefit a lot from attack, actually, against zealots and stalkers. If not, I'll try to go for melee, for the long game plan (broodlords gain most damage from broodlings, not air attack, and lings can appreciate them too, for ling/muta). Of course, carapace can be useful if you aren't sure which way to go, since both roaches and lings can appreciate it. Personally,I go for either melee or ranged, depending on how much pressure toss is going to put, and only if I see toss have gas.
Sometimes I go double evo though, if I know I'm not going to go mutas (like if toss goes immortal/sentry or blink stalker at start). I know I just said upgrades aren't that great, but if I dont go mutas, I tend to max out on just roach/ling, so I have to put the gas somewhere, and I don't tech up to infestor bane rain roach until I hit 200/200 first.
However I DEFINITELY don't go double evo at 7:30. I get the 2nd evo wayyyyy later. I understand what you are talking about, you see toss teching heavily, but toss getting double gas at nat could also be immortal/sentry or blink stalker pressure, both of which where you just want as much shit as possible, hydras, and units - not upgrades necessarily. I'd recommend you stop going double evo so quickly. If you like double evo, just grab them later, when you are safe. Don't open it, definitely not. Toss will always be ahead in weapons upgrade, so I don't think it's a big deal to get carapace in ZvP. Don't be at 0 armour when Toss is at 3 bases, of course, but it's definitely not a priority.
That's just my thoughts on the matter. When you get down to it, as long as toss has gas at natural, you are okay to get one of the three upgrades. Which one, is up to you - ranged if you plan to stay on roaches, melee if you plan to go towards mutas, carapace if you aren't sure which way you are going. I'd prefer going toward either melee or ranged, since carapace is more expensive and I feel is less useful. Zerg isn't toss where you make strong units, you strength is making lots and lots of units, not upgrades or tech necessarily.
Do you have any comments on preference of upgrades? I know Stephano likes to get +1 melee for lategame purposes while I personally like getting +1 range for midgame control. Do you change your upgrade based on certain gas timings taking guesses as to how you feel the lategame will play out based on the midgame? Is double evo at 7:30 mark just suicide most of the time? I have personally been trying Lair into double evo when I see 2 gas being taken and mined with macro hatch. Is the second evo to greedy that early?
It is difficult for me to tell since my macro mechanics are pretty shit and it is hard for me to see what build to "aim" for.
Thanks.
Well, I don't know if I'm qualified on this. This is a very basic guide, on something that most masters+ already know about, so it's not like I'm saying anything new there. This isn't like a guide on "How to play mutas in ZvP' or "my 12 minute timing attack ZvT" or "Ultras in ZvZ". it's just everything put together that some have done, but no one has done in a way that wasn't complete shit cough skoozy cough.
That said, I have done a lot of testing on upgrades. So here's my thoughts (you can check my Upgrades on Mutalisks thread, as well as multiple comments I've made about it, I think I've made a few other upgrade threads as well).
First off, if I see zero gas at the Toss natural, I don't make any upgrades. If toss is doing a 6/7/8 gate all-in, the upgrade won't help at all (as I stated in the Roach Warren and Evo Chamber section of the OP). Evo is 35 seconds, upgrade is 160, that's 185 seconds, or 3 minutes 5 seconds. 7:00 + 185 seconds, that means the upgrades kicks in at 10:05.
So when you start an upgrade, you need to ask yourself "Will the upgrade finish in time?". Against a 7 gate all-in that will be inside your base at 8:50, no - you will have GGd out by 9:00. Against blink stalker all-in, it's questionable. You won't have it at first, but with a spine, you can maybe buy that 30 seconds, and then have it. Against sentry/immortal? Yea, you should definitely get it, since lings are kind of useless in that situation, but hydras are king. If Toss opens stargate, getting upgrades can always be useful, as you can get on that ground upgrade ahead of Toss. But, you know, it's up to you. I'm actually a big advocate of NOT getting upgrades in most match-ups, and preferring to mass units or tech up to infestors or hive much quicker.
You also have to wonder about critical hit numbers.
So, for example, +1 carapace. I largely think carapace is useless in most cases, especially in ZvP, when Toss is always racing attack upgrades. I will often go 0 carapace in ZvP, as it doesnt often change critical hit numbers (+3 zealots will always rape lings).
But for that first upgrade? If Toss is putting on gateway pressure, I'll get ranged attack. Roaches benefit a lot from attack, actually, against zealots and stalkers. If not, I'll try to go for melee, for the long game plan (broodlords gain most damage from broodlings, not air attack, and lings can appreciate them too, for ling/muta). Of course, carapace can be useful if you aren't sure which way to go, since both roaches and lings can appreciate it. Personally,I go for either melee or ranged, depending on how much pressure toss is going to put, and only if I see toss have gas.
Sometimes I go double evo though, if I know I'm not going to go mutas (like if toss goes immortal/sentry or blink stalker at start). I know I just said upgrades aren't that great, but if I dont go mutas, I tend to max out on just roach/ling, so I have to put the gas somewhere, and I don't tech up to infestor bane rain roach until I hit 200/200 first.
However I DEFINITELY don't go double evo at 7:30. I get the 2nd evo wayyyyy later. I understand what you are talking about, you see toss teching heavily, but toss getting double gas at nat could also be immortal/sentry or blink stalker pressure, both of which where you just want as much shit as possible, hydras, and units - not upgrades necessarily. I'd recommend you stop going double evo so quickly. If you like double evo, just grab them later, when you are safe. Don't open it, definitely not. Toss will always be ahead in weapons upgrade, so I don't think it's a big deal to get carapace in ZvP. Don't be at 0 armour when Toss is at 3 bases, of course, but it's definitely not a priority.
That's just my thoughts on the matter. When you get down to it, as long as toss has gas at natural, you are okay to get one of the three upgrades. Which one, is up to you - ranged if you plan to stay on roaches, melee if you plan to go towards mutas, carapace if you aren't sure which way you are going. I'd prefer going toward either melee or ranged, since carapace is more expensive and I feel is less useful. Zerg isn't toss where you make strong units, you strength is making lots and lots of units, not upgrades or tech necessarily.
Thanks for the informative reply. Im glad that you have consolidated so much information that "everyone" knew.
I have felt the same way towards +1 carapace ever since chronoboost was introduced in Sc2. I tried builds where I go for early +1 carapace with early macro hatch and go very quick infestor ling with drops delaying my third. It's similar to a style Sheth uses some times. This is ok on some maps where it would take a ton of forcefields to zone lings but on maps where you can't do it so well. forcefield and +2 zealots just rip it apart. Lings just don't seem to be that good ZvP except for baneling drops eventually
It's a hell of a lot easier to hold a blink stalker all-in with 65 drones instead of 55 drones, if you work on your macro. Chances are, if you have games where you go "wtf Toss just masses a huge fucking ball of stalkers/units/etc and wins", check your supply at 8:00. If you aren't past 60 supply, you've pretty much already lost the game, really. If it's in the 50's, I don't think it's possible for you to win unless the opponent is just as bad.
This confuses me, since in the only replay you linked in the OP the zerg is at 53 supply at 8:00.
^ DRG vs Genius games, he's at 70+ supply in all but daybreak. Nestea was 65 vs Naniwa on shakuras (and as fast as he took his gas, that was quite impressive).
70+ in a real game is really damn impressive. ~65-70 is pretty standard (everyone seems to get supply blocked at 70/70). Below 60 is 'bad'.
This confuses me, since in the only replay you linked in the OP the zerg is at 53 supply at 8:00.
i havent read the whole thread, but i have experimented with a 60-70 drone count roach/baneling army ZvT. max asap on roach/ling/bling, the longer the game goes on the more of your lings become blings. the terran will usually push a few times before you max, bling/roach, with ling reinforcements, seems to clean up these smaller attacks really well. to finish them off use your last 20 supply on mutas and attack. in the game i just played i was 200/200, he was ~120/200 after 2 failed attacks at about 16 min.
Yea I agree I think the roaches really inflate the food count if you make them at that point, you get lower with pure drones.
Best iv managed after trying a few times is 79 supply with 4 lings 3 queens and no roaches. Get the same with a later triple gas build as the standard earlier double gas build.
If anyone has any ideas how to improve further please suggest, im only diamond. But my SQ these games was like 118 or so, so i doubt theres too much room for improvement.
14 pool is standard, but that doesnt make it right or better. I would think 15 pool is way better, but I havent tested it. I mean, 14 hatch is supposed to be better than 15 hatch, but the same pros who still extractor trick every game also 15 hatch evety time.
On March 22 2012 03:36 Iksf wrote: Yea I agree I think the roaches really inflate the food count if you make them at that point, you get lower with pure drones.
Best iv managed after trying a few times is 79 supply with 4 lings 3 queens and no roaches. Get the same with a later triple gas build as the standard earlier double gas build.
If anyone has any ideas how to improve further please suggest, im only diamond. But my SQ these games was like 118 or so, so i doubt theres too much room for improvement.
Really well done! But there is still room for improvement, you missed a some injects.
Also, I think you need to add some scouting lings and creep spread between nat and third for it to be safe in a real game scenario.
Great guide. Any comments on the location of the 3rd base? For example is it safe enough to take the gold as a third on Metalopolis? How about Tal Darim where the third is blocked by rocks? Do you need to produce more units to take them down? Sorry for newbie questions.
On March 22 2012 07:25 Belial88 wrote: 14 pool is standard, but that doesnt make it right or better. I would think 15 pool is way better, but I havent tested it. I mean, 14 hatch is supposed to be better than 15 hatch, but the same pros who still extractor trick every game also 15 hatch evety time.
Great guide. Any comments on the location of the 3rd base? For example is it safe enough to take the gold as a third on Metalopolis? How about Tal Darim where the third is blocked by rocks? Do you need to produce more units to take them down? Sorry for newbie questions.
In ZvP you always want to take it farther away from Toss.
I really don't like playing on map where you need to break rocks to take your third, like TDA. It just means you kill your econ trying to get that third, making any possibility of you holding a strong timing impossible - even if you survive, Toss will have built up his probes or whatever while forcing you to be way lower on drones/tech than normal, and being able to normalize much easier, because on most maps you would have more drones, and thus be able to tech more, and then counterattack with mass roaches because you will be so far ahead in econ and eventually tech. And on TDA specifically, toss can take a super easy third, so really, he can just all-in, if it fails, take his third, and it's okay, because he won't be punished for all-inning on that map.
If you take a faraway third on TDA, you will straight up lose to a blink stalker or imortal/sentry all-in. Your rallying roaches, or hydras, will just be way too slow, and you will die. Your army will be cut in half basically, because your third is so far away and his push will occur at your natural, between the 3 hatches, or just at the third and your main/nat hatch production will just take forever to rally there.
You can't go fast third on TDA, is what I'm saying basically. You can't go fast third on maps where the third has rocks or is super far away.
As for Metalopolis - I actually think this map is completely broken, in favor of zerg, for ZvP. This is the one map I think is truly broken for ZvP in favor of Zerg, against Toss.
Not only can you take a third on Metal, you SHOULD! It's close enough to your main/nat that rallying units is not a problem, and even if it was, it doesn't matter, because the income boost is HUGE. Whereas on Shattered, the gold has rocks that take forever and a lot to break down, on Metal, you can take it. Toss cannot do ANY gateway pressure before the 8:00 mark, so yea, you are 100% safe to take gold first on metal against toss FFE. The standard zealot/stalker can be pressure of course, but you can easily make a spine or some lings for that on Metal (or any map) for that matter.
There is absolutely no reason not to take the gold as your FIRST base on metal, against FFE ZvP. It's so completely broken, and nothing Toss can do to punish it.
At least antiga has bullshit close spawn to make up for gold.
So yea. Veto Shattered and TDA for being bullshit maps that FORCE you to only play 2 base against Toss, which limits your options. So toss can just blind counter your build. The problem with this, is that fast third is the standard response to FFE, so when you can't even play standard, the map is broken, because you are forced to do builds that require aggression or you fall behind, ie you have to be all-innish. Now TDA is REALLY good for mutas, so 2 base muta is still okay on that map, even if they blind counter it, due to how good mutas are on that map, but I still don't like being forced into a single build on that map, that ISNT the standard build.
Man... I just went back through this thread and the macro benchmarks section is so awesome for comparing my own play.
I'm personally low masters and I rarely break 65 food in any game that I've analyzed! The fact that DRG routinely breaks 75 food is so impressive to me, just sick macro from that guy!
I just did it a few times in yabot and the most I ever got was 72, but that's with no multitasking (scouting, taking towers, losing lings to protoss unit etc etc)
Good guide. Agree except on 11 pool statement. Because pylon blocks are so common, an 11 pool will let you get your nat earlier frequently. You can even go 11 pool-over in order to get 2 early lings. Expanding to 3rd first when blocked is a risk and beacuse of the maynarding distance, you'll loose some mining time.
Benchmarks of 11 pool-over (not overpool !) with a pair of early lings and 4'00 extractor
3'00 2 lings at nat 3'10 put nat (if not blocked) 3'18 first inject 4'00 extractor 4'30 2cnd queen 5'07 26 drones mining ^^ ~5'20 take 3rd 5'30 start speed (or lair) 7'30 macro hatch (migth be to early though, but i like to over-larvae)
8'00 70/70 with 4 queens 3 bases + macro hatch underway (even with somewhat sloppy play)
The long-term impact of the pool timings are overrated as in SC2 earlier pool => earlier queen => earlier spawn evens out things a lot. As long you are capable to spend your larvae, the build is ok. As you can see an 11-pool-over gets you lotso drones early (as long you take gas >= 4 minutes).
^ I don't agree. If you go 14 pool, you can get your hatch in reasonable time by going 14 pool, 2 drone to prevent pylon ramp blocks/cannon rushes send out at ~16 supply, then take the nat (if toss doesn't put the pylon down, even if you go 11 pool, if they put pylon down it delays the hatch), then make 4 lings, you will be at 17/18, then make overlord. You can extractor trick to get queen, or do that in another order (queen then overlord then 2xlings with 17 overlord and extractor trick, or not.
I get your supply benchmarking, but that doesn't exactly tell you if a build is better or not exactly. You need to go by resources mined and drones created. I'm not sure it's possible to really do with fast third to 9 minute without a pro or a computer to figure it out, since people almost always screw up macro, having idle larva, et cetera.
The problem with earlier pool to earlier queen is that, yea spawn larva evens out drone production, but it doesn't even out expansions. By going 14 pool, you are purposefully banking money over constant larva usage, so you can plant that hatch down, while 11 pool, you will not bank money, and so need to idle larva or make some sort of cut.
I suppose with pylon blocks being so popular, it might be better to go earlier pool? I don't know. I know that I tried out 15 pool for a while, and it was just too late on the lings and the pylon fucked me over for too long that I felt that 14 pool was way better. Later pool is just soo much more econ, but that pylon block sure fucks things up. I guess if you could guarantee no pylon block, then later pool, and if they do block, you want earlier pool maybe? that might be interesting to try out, like 14pool/21 hatch vs 11 pool/15hatch.
Could you give an overview of how your supply evolves (awkward 2 gate proxy zealot or similir shenanigans not withstanding) in order to have +60 supply at the 8.00? I'm talking drones, zerglings, queens, spawn larva and overlords.
That would be great.
I specified it in the guide, but to put it all in one place:
14 pool 16-21 hatch, 4 lings Queens as necessary 25-30 Third 2 x gas ~ when third is done 7:00 evo 7:30 roach warren, 3 x overlord sacs 8:00 60+ supply Making overlords is just being super careful. Like 34/36 instead of 31-33. Even at the 46 supply you dont need to start 2 overlords at once. You go to like 50/54. Then after that you need to be very careful, because of injects, but start an overlord when the 1st one is nearing completion. At about 60 you are doing like 1 overlord and starting the next very soon afterwards.
Do you prioritize getting queens or spending larvae on drones? Do you make a second queen in main and transfer our build at the hatchery you want the queen at? What about creep tumors? What are the basic food relative overlord timings?
^ Those kinds of details are preference. I'm really not someone you should be asking on this lol. Just watch a pro rep (seems like they go 1 queen, then queen at nat, then queen at nat to walk to third after third is started). As for creep tumors, it's debatable how important that is. Some people start the 3rd queen earlier (like right when third is started) so she can plant a tumor at nat or third depending on which is harder to defend (like daybreak, it's popular to creep out third first thing). Some say a 4th creep queen is very important to holding immortal/sentry and blink all-ins (something I've starting to incorporate more in my play recently, and liking a lot).
I just wrote what is kind of standard. I'm definitely not a good player. This is just sort of the basics.
Basic relative overlord timings, are like just a few food short, like 1-3 for up to about 44. Then one at 50/54, then make 1 and start next overlord soon after and another overlord very soon after that, as you will race from 50-80 supply very quickly due to 3 injects timing together and the income to push it through.
On April 07 2012 23:10 Belial88 wrote: ^ I don't agree. If you go 14 pool, you can get your hatch in reasonable time by going 14 pool, 2 drone to prevent pylon ramp blocks/cannon rushes send out at ~16 supply, then take the nat (if toss doesn't put the pylon down, even if you go 11 pool, if they put pylon down it delays the hatch), then make 4 lings, you will be at 17/18, then make overlord. You can extractor trick to get queen, or do that in another order (queen then overlord then 2xlings with 17 overlord and extractor trick, or not.
Hm, you noticed i am talking of 11 pool-ovie which has a 1'22 pool timing compared to 1'36 of a regular 11 pool (idles larvae but super early lings let you skip the drone scout, so it evens out somewhat) ?
The early lings serve the following: a) cheese defense in case b) 2 early scouts to look for hidden pylons c) take down the blocking pylon much quicker (you can get your nat at 3'30 even if it has been pylon blocked) d) no need for drone patrol at ramp e) after that you got 2 early scouts, i skip the scouting drone (have to make up for idling larvae before pool)
A regular 11 pool has similar advantages, however the lings are somewhat late. If you get pylon blocked (standard on EU) and you patrol 2 drones at your ramp (+ a scouting drone), you will be way behind the 11 pool regarding minerals mined *and* expansion timing.
On April 07 2012 23:10 Belial88 wrote: I get your supply benchmarking, but that doesn't exactly tell you if a build is better or not exactly. You need to go by resources mined and drones created. I'm not sure it's possible to really do with fast third to 9 minute without a pro or a computer to figure it out, since people almost always screw up macro, having idle larva, et cetera.
Well, there is software out there which is capable to compute optimal timings .. take a computed build and tweak it somewhat, you don't need to be a 'pro'
The most important factor for me (because i am bad and get thrown off by multitasking easily) is the number of drones a build creates early on. Obviously more drones = more minerals mined on the long run. I agree that an 11 pool has no 'room' (without idling larvae) to create additional buildings up to the 5 minute mark. That's because it creates more larvae. However a 5'20-5'30 3rd is sufficient imho.
On April 07 2012 23:10 Belial88 wrote: The problem with earlier pool to earlier queen is that, yea spawn larva evens out drone production, but it doesn't even out expansions. By going 14 pool, you are purposefully banking money over constant larva usage, so you can plant that hatch down, while 11 pool, you will not bank money, and so need to idle larva or make some sort of cut.
Well, banking money is suboptimal as it basically tells you your build creates too few larvae ;-) anyway i get your point, however in case of a three hatch build you pretty much like to drone up quickly and as stated above you are still capable to get those 2 hatches up (slightly later: Nat 3'05, 3rd 5'30).
On April 07 2012 23:10 Belial88 wrote: I suppose with pylon blocks being so popular, it might be better to go earlier pool? I don't know. I know that I tried out 15 pool for a while, and it was just too late on the lings and the pylon fucked me over for too long that I felt that 14 pool was way better. Later pool is just soo much more econ, but that pylon block sure fucks things up. I guess if you could guarantee no pylon block, then later pool, and if they do block, you want earlier pool maybe? that might be interesting to try out, like 14pool/21 hatch vs 11 pool/15hatch.
Well i switched from 14 pool to 11 because in reality (pylon block) this opening is just more reliable and save. Ofc i am an amateur laddering, so you won't see those safe openings in pro tourney play that often (except when pro's play underdogs and suspect cheese, Nestea frequently goes 11/12 pool against subpar players). On some maps (e.g. tal darim) you have a good chance to get the hatch before a blocking pylon can be placed, however i dislike gambling and I dislike being thrown off my build order early on .
If you're not doing 14 pool, 11 pool is quite silly. Do 12, at the earliest. If you do 11 Overpool, the earliest pool you can plant is 1:30. With a 12 pool, you get it at 1:39, and 12 pool is the earliest pool you can have without wasting a single larva. Since you're talking about larvae efficiency, I feel like that's more important.
That said, 12 pool is completely inferior to 14 pool because you still do not get the lings out in time to block the pylons, it's much simpler to just put a drone on hold position at your ramp. If he attacks the drone, bring an additional down. If he plants any pylons, bring a handful down, but since he can't block the ramp you can surround easily.
Then ignoring that, in the 'ideal' situation of building a hatch ASAP, a 12 pool gets one out at (earliest, never idling drones) 3:20. The 14 pool gets one out 2:40, DRASTICALLY sooner. That's half the build time for the hatch. Not only that, but they both have 15 drones at the start of building. 11 Pool just makes it worse.
So, park a drone at the bottom of your ramp if you're worried about pylon blocks imo.
Outstanding guide. I have one comment though from the Protoss perspective.
Its more of a mind game than anything else, but I have one issue with the guide. Toss can and will try to deny overlord scouting. Often times, I will wait for an ovie to fly over my pylon, then build gateways behind it. Sacrificing those 30 seconds for surprise is worth it 100%. Its just important to note that as a Zerg you must be very careful and always maintain perfect map control. You'd be surprised how easy it is for Protoss to simply cancel that tech and 8 gate you, or just hide those gateways right after the ovie flew over.
On April 08 2012 01:45 mtszyk wrote: If you're not doing 14 pool, 11 pool is quite silly. Do 12, at the earliest. If you do 11 Overpool, the earliest pool you can plant is 1:30. With a 12 pool, you get it at 1:39, and 12 pool is the earliest pool you can have without wasting a single larva. Since you're talking about larvae efficiency, I feel like that's more important.
That said, 12 pool is completely inferior to 14 pool because you still do not get the lings out in time to block the pylons, it's much simpler to just put a drone on hold position at your ramp. If he attacks the drone, bring an additional down. If he plants any pylons, bring a handful down, but since he can't block the ramp you can surround easily.
Then ignoring that, in the 'ideal' situation of building a hatch ASAP, a 12 pool gets one out at (earliest, never idling drones) 3:20. The 14 pool gets one out 2:40, DRASTICALLY sooner. That's half the build time for the hatch. Not only that, but they both have 15 drones at the start of building. 11 Pool just makes it worse.
So, park a drone at the bottom of your ramp if you're worried about pylon blocks imo.
You are missing the point. Its about a pylon blocking the expansion not the ramp. It takes forever to get this down with drones only and can delay your significantly. The 11 pool before ovie idles larvae ~10 ingame seconds, but gets pool (+queen +lings) 14 ingame seconds earlier. What i am arguing is, that this is worth the slight larvae idling, since the early lings replace the drone scout, replace the ramp drone and take a blocking pylon down quickly. The lost larvae is recovered by a >10 seconds earlier inject (10 seconds earlier queen = 1 larvae). In theory you are behind income wise, in a real game (need drone scout, need drone ramp block, delayed expansion) you are most of the time better off with an early pool. Additionally if you delay gas the 11 underpool (pool before ovie) lets you drone up faster than any other build (30 drones at ~5'10 possible)
On April 07 2012 23:36 Belial88 wrote: ^ Those kinds of details are preference. I'm really not someone you should be asking on this lol. Just watch a pro rep (seems like they go 1 queen, then queen at nat, then queen at nat to walk to third after third is started). As for creep tumors, it's debatable how important that is. Some people start the 3rd queen earlier (like right when third is started) so she can plant a tumor at nat or third depending on which is harder to defend (like daybreak, it's popular to creep out third first thing). Some say a 4th creep queen is very important to holding immortal/sentry and blink all-ins (something I've starting to incorporate more in my play recently, and liking a lot).
I just wrote what is kind of standard. I'm definitely not a good player. This is just sort of the basics.
Basic relative overlord timings, are like just a few food short, like 1-3 for up to about 44. Then one at 50/54, then make 1 and start next overlord soon after and another overlord very soon after that, as you will race from 50-80 supply very quickly due to 3 injects timing together and the income to push it through.
Yeah, I've noticed this same thing with regards to the 50 food break. You hit 3 injects really fast and you need 2-3 overlords fast to make all your larva into drones right away.
Also, I was talking to a friend of mine who's top 20 GM (nxmhendralisk) and he was saying the best order to get your queens/hatcheries is: pool hatch queen queen hatch queen, where the first queen injects the main hatch, plants a tumor at the main, and then goes down to the natural, injects there, and then walks to the third. The second queen is built at the main hatchery and should pop in the main in time to inject there. The third queen is built at the nat, and it should finish in time to inject there.
I thought it was pretty strange in the sense that the first queen does this kind of exodus from the main to the nat to the third, but the first time I did it this way it felt really smooth, since you get pretty small inject gaps plus a creep tumor running.
So, why not simply build it at the third? I do it every time, and for the most part it doesn't hurt me. There have been a small number of games where my enemy capitalizes on it, or I make an error, or they are simply better than I am, but simply taking the third first seems to work great for me. When I do, I also delay a second queen and just take the natural sooner. Some maps make this slightly more difficult than others, but I've done it with success (and failure) on maps like Tal'Darim.
As far as economy goes, you can't possibly argue that 11 pool of any variation is superior to 14 pool. Your other point has merit, and I think it's a matter of style, but 14 pool is just more economy guaranteed.
I'm getting the same number of drones at 5'10 with both builds (plus or minus 1-2), but a third far sooner with 14 pool. The only assumption I really made is that you're not cutting drones to get the hatch out sooner. Additionally, the 11 pool has more larvae initially, but the 14 pool gets the second hatch out much sooner, so if you need to suddenly defend against something with a swell of zerglings it is superior, until around 5:10 (11 pool expansion finishes around here, again assuming you didn't cut drones) where it balances, then it becomes superior again ~30-40 seconds later when the third finishes.
On April 08 2012 03:06 HyperionDreamer wrote: Also, I was talking to a friend of mine who's top 20 GM (nxmhendralisk) and he was saying the best order to get your queens/hatcheries is: pool hatch queen queen hatch queen, where the first queen injects the main hatch, plants a tumor at the main, and then goes down to the natural, injects there, and then walks to the third. The second queen is built at the main hatchery and should pop in the main in time to inject there. The third queen is built at the nat, and it should finish in time to inject there.
I thought it was pretty strange in the sense that the first queen does this kind of exodus from the main to the nat to the third, but the first time I did it this way it felt really smooth, since you get pretty small inject gaps plus a creep tumor running.
Neat little way to structure the injects, I'll be trying that out, thanks.
On April 08 2012 03:11 mtszyk wrote: So, why not simply build it at the third? I do it every time, and for the most part it doesn't hurt me. There have been a small number of games where my enemy capitalizes on it, or I make an error, or they are simply better than I am, but simply taking the third first seems to work great for me. When I do, I also delay a second queen and just take the natural sooner. Some maps make this slightly more difficult than others, but I've done it with success (and failure) on maps like Tal'Darim.
As far as economy goes, you can't possibly argue that 11 pool of any variation is superior to 14 pool. Your other point has merit, and I think it's a matter of style, but 14 pool is just more economy guaranteed.
I'm getting the same number of drones at 5'10 with both builds (plus or minus 1-2), but a third far sooner with 14 pool. The only assumption I really made is that you're not cutting drones to get the hatch out sooner. Additionally, the 11 pool has more larvae initially, but the 14 pool gets the second hatch out much sooner, so if you need to suddenly defend against something with a swell of zerglings it is superior, until around 5:10 (11 pool expansion finishes around here, again assuming you didn't cut drones) where it balances, then it becomes superior again ~30-40 seconds later when the third finishes.
See the problem is that you should be vetoing TDA instead of playing it against P since the map forces you to go 2 base muta and toss just blind counters it since the only other thing to do is like a 2 hatch nydus allin.
If you delay the queens at all then you're sacrificing inject time, which is the most important macro mechanic for Zerg to make more drones. I don't know why you would want to delay queens at all with this build when the goal of it is to get as many drones as you can by 8 minutes before the toss allin hits.
On April 08 2012 03:11 mtszyk wrote: So, why not simply build it at the third? I do it every time, and for the most part it doesn't hurt me. There have been a small number of games where my enemy capitalizes on it, or I make an error, or they are simply better than I am, but simply taking the third first seems to work great for me. When I do, I also delay a second queen and just take the natural sooner. Some maps make this slightly more difficult than others, but I've done it with success (and failure) on maps like Tal'Darim.
As far as economy goes, you can't possibly argue that 11 pool of any variation is superior to 14 pool. Your other point has merit, and I think it's a matter of style, but 14 pool is just more economy guaranteed.
I'm getting the same number of drones at 5'10 with both builds (plus or minus 1-2), but a third far sooner with 14 pool. The only assumption I really made is that you're not cutting drones to get the hatch out sooner. Additionally, the 11 pool has more larvae initially, but the 14 pool gets the second hatch out much sooner, so if you need to suddenly defend against something with a swell of zerglings it is superior, until around 5:10 (11 pool expansion finishes around here, again assuming you didn't cut drones) where it balances, then it becomes superior again ~30-40 seconds later when the third finishes.
Being behind in minerals mined is distinct from being behind in drone count. Ofc the 11 pool is behind minerals-mined wise, but it is not behind (but slightly better) in number of drones. While being behind temporary in minerals-mined will stay constant (or even out due to faster droning), being behind in # drones will snowball your "behindness" (bad).
What i am arguing is the fact, that a real world 14 pool is not able to take the theoretical advantage as you are losing mining time drone scouting, ramp blocking with drones and attacking a blocking pylon with drones. If you expand first at the 3rd, you take a risk and you lose mining time because your drones need to travel a longer distance (again anihilating your theoretical mining advantage). I bet if you do some statistics, you'll lose games more often because your opponent capitalizes on your far first expansion than lose games because you are (sometimes) 100 minerals behind at the 6 minute mark.
Also note a build generating fewer larvae "feels" more economic, because you always have enough money to spend your larvae.
Hm, you noticed i am talking of 11 pool-ovie which has a 1'22 pool timing compared to 1'36 of a regular 11 pool (idles larvae but super early lings let you skip the drone scout, so it evens out somewhat) ?
The early lings serve the following: a) cheese defense in case b) 2 early scouts to look for hidden pylons c) take down the blocking pylon much quicker (you can get your nat at 3'30 even if it has been pylon blocked) d) no need for drone patrol at ramp e) after that you got 2 early scouts, i skip the scouting drone (have to make up for idling larvae before pool)
A regular 11 pool has similar advantages, however the lings are somewhat late. If you get pylon blocked (standard on EU) and you patrol 2 drones at your ramp (+ a scouting drone), you will be way behind the 11 pool regarding minerals mined *and* expansion timing.
No. I already talked about all of this quite clearly in the OP.
First off, 14 pool can hold off any sort of proxy gate cheese just fine. Why would you go 11 pool when you can just go 14 pool just as safely? Like, with that philosophy, why don't you just go defensive 6 pool? Because it's goofy, you don't need to, and it's better to be as greedy as safely possible.
Secondly, you are wrong. If you do not react the exact same way, when going 11 pool, as you would going 14 pool - ie, pulling drones against cannon rushes, patrolling a drone at the ramp until pool pops, you WILL lose to a cannon rush. 11 pool will NOT get lings in time to defend a cannon rush. All they do is make the timings for cannon rushes smaller, and ONLY by a few seconds. It doesn't not prevent them. If you do not patrol a drone at the ramp, 11 pool or 14 pool or hatch first, you WILL lose.
So if 11 pool doesn't get lings out in time, why bother? Just go 14 pool...
Also, making 4 lings much earlier, and sac'ing ~6 drones in econ just to get the hatch a few seconds earlier, is not worthwhile. You aren't making up the econ lost by going 11 pool by planting the hatch a little earlier. Feel free to test it out yourself, or check out other threads where it has been scientifically proven with the Second Look at Zerg Openers and End to Zerg opening debate threads.
However a 5'20-5'30 3rd is sufficient imho.
What's that in supply? That doesn't sound quite right though... it sounds like your third is way too late. If toss does zealot pressure or zealot/stalker/stalker it sounds like you'd just autolose because you won't be able to mine at your third.
Well, banking money is suboptimal as it basically tells you your build creates too few larvae ;-) anyway i get your point, however in case of a three hatch build you pretty much like to drone up quickly and as stated above you are still capable to get those 2 hatches up (slightly later: Nat 3'05, 3rd 5'30).
No... you are constantly using up your larva, and banking money, at which you spend it at an exact moment, for a reason. You want to bank money. Banking 300 minerals for a hatch is good. Banking 300 minerals because of poor macro, is not good. Different.
Well i switched from 14 pool to 11 because in reality (pylon block) this opening is just more reliable and save. Ofc i am an amateur laddering, so you won't see those safe openings in pro tourney play that often (except when pro's play underdogs and suspect cheese, Nestea frequently goes 11/12 pool against subpar players). On some maps (e.g. tal darim) you have a good chance to get the hatch before a blocking pylon can be placed, however i dislike gambling and I dislike being thrown off my build order early on .
Trust me. I went 11 pool for a year for the exact same reason. Then someone, who went very normal 13 forge for a FFE, put 3 pylons at the ramp at the standard 17 supply. 6 lings popped, and ran into a just-finishing cannon behind a wall off.
It'll happen to you one day. Someone will ramp block you, because you are too arrogant with your build, and you'll realize what a waste 11 pool is because it doesn't get lings out in time to stop a ramp block.
Nestea experimented with 11 pool, as an aggressive build. Not as a build for safety. He did it as a timing against the possibility of toss going nexus first. If toss goes nexus first, 11 pool can be awesome. If they don't go nexus first, well, you are a bit behind, but it's a gamble.
By the way, on 4 player maps like TDA, you should not be able to go hatch first against probe blocks. Toss will always send 2 probes out to scout on 4 player maps, unless they are stupid, or trying to be greedy.
Often times, I will wait for an ovie to fly over my pylon, then build gateways behind it.
That's why you keep track of where the pylons are placed, and you fly in at 7:00-7:30. If you are going to wait until after my overlord flies over at 7:30 dies to a single sentry or stalker, to place your gateways, I end up wayyyy ahead than you do from that. There are other clues that will say what's going on to, like spinning forge, chrono on WG or gateways, sentries vs stalkers, gas count...
It would make no sense to put down your gateways at 8:00. That's just sooo damn late. You are going to only have one gateway and cybercore as your entirety of tech for 8 minutes?
No...
Also, a 9:30 8 gate is a joke. I would be completely fine with you cancelling tech like that. Good luck doing anything with tech/gates out that late.
Secondly, Zerg should never assume they know everything from a warping building. You need to see it finish. A 7:30 overlord will always see shit finish. A stargate opening will have a void ray out by 7:50. So good luck cancelling.
What i am arguing is the fact, that a real world 14 pool is not able to take the theoretical advantage as you are losing mining time drone scouting, ramp blocking with drones and attacking a blocking pylon with drones.
And I am arguing you are wrong. You NEED to ramp block with drones and scout and attack blocking pylons with drones when going 11 pool, exactly the same as 14 pool.
I did 11 pool for a year because I thought it had these advantages. Turns out, I was wrong. You can still be cannon rushed, you can still be ramp blocked, your expo can still be pylon blocked, and you still need to drone scout (or not, some people don't drone scout in ZvP, but if you go 11 or 14 pool doesn't change the merits, or disadvantages, of drone scouting).
On April 07 2012 23:36 Belial88 wrote: ^ Those kinds of details are preference. I'm really not someone you should be asking on this lol. Just watch a pro rep (seems like they go 1 queen, then queen at nat, then queen at nat to walk to third after third is started). As for creep tumors, it's debatable how important that is. Some people start the 3rd queen earlier (like right when third is started) so she can plant a tumor at nat or third depending on which is harder to defend (like daybreak, it's popular to creep out third first thing). Some say a 4th creep queen is very important to holding immortal/sentry and blink all-ins (something I've starting to incorporate more in my play recently, and liking a lot).
I just wrote what is kind of standard. I'm definitely not a good player. This is just sort of the basics.
Basic relative overlord timings, are like just a few food short, like 1-3 for up to about 44. Then one at 50/54, then make 1 and start next overlord soon after and another overlord very soon after that, as you will race from 50-80 supply very quickly due to 3 injects timing together and the income to push it through.
Yeah, I've noticed this same thing with regards to the 50 food break. You hit 3 injects really fast and you need 2-3 overlords fast to make all your larva into drones right away.
Also, I was talking to a friend of mine who's top 20 GM (nxmhendralisk) and he was saying the best order to get your queens/hatcheries is: pool hatch queen queen hatch queen, where the first queen injects the main hatch, plants a tumor at the main, and then goes down to the natural, injects there, and then walks to the third. The second queen is built at the main hatchery and should pop in the main in time to inject there. The third queen is built at the nat, and it should finish in time to inject there.
I thought it was pretty strange in the sense that the first queen does this kind of exodus from the main to the nat to the third, but the first time I did it this way it felt really smooth, since you get pretty small inject gaps plus a creep tumor running.
I messed around with my own timings and what I came up with is: 14 pool 16 hatch then queen 2 lings (extractor trick more if you need) overlord then 3rd asap When queen pops finish spending larvae on drones and make another queen in main When natural completes, make a queen there asap. The second queen in main starts a tumor to connect main/natural and goes to natural The queen in natural starts a tumor and goes to third I like to have lots of creep early (hence the two running tumors) to make defending air with queens possible, as well as making slowlings feasible until my late speed kicks in.
This still easily allows for 60+ drones at the 8 min mark along with a roach warren, evo chamber, and enough gas for lair and speed if you aren't getting attacked or a nice 3 hatch round of roaches if you are.
On a side note, I've found that getting blocked at 70/70 (on basically just drones and queens) and doing 4-5 overlords at once works pretty well, since that's basically at the trigger timing to go into full on roach production if you need to. And if you don't need to you're going to do lair + speed + an upgrade + macro hatch at this point anyway, so you can let resources stack up for 20-30 seconds without falling behind.
I've added a bit about 4 gate +1 timing - sorry guys, I was wrong about saying you don't need roaches against it.
I've also added about how to completely play the mid and late game in zvp. I plan to evolve this into a completely comprehensive guide on ZvP.
later on, I plan to add 1 base all-ins like 4 gate, 4 gate robo, 1 base dt/sg, gateway expands and how to play against that. I think that's all that's really missing from this being a completely everything guide, because ZvP is so much about ffe these days, and the rest of the game is pretty straightforward.
I would appreciate if a mod could change the title to:
[G] Belial's Comprehensive Guide to Everything ZvP! alternatives if that doesn't fit: + Show Spoiler +
[G] Belial's Comprehensive Guide to ZvP! [G] Belial's Comprehensive Guide to ZvP [G]Belial's Comprehensive Guide to ZvP! [G] Belial's Guide is so much better than that shitty Swooozy guide that caused him to write this guide in the first place haha banned bitch ZvP!
amazing guide Belial, your made my zvp very very good, now 70% winrate in this matchup, your doing an exccelent work for the community of macro oriented zerg players. Again, thanks alot !
On April 27 2012 13:47 KhAmun wrote: You should add in the 3hatch ling/bane allin tang did a guide on the allin section. My personal thoughts on tang aside, it's an effective build.
Well, that all-in is based on a guide/strat that Tang came up with. The all-in section is about general all ins that have been popular in the metagame at some time, and 'proven' to be effective (or rather, proven to be poular, which may or may not indicate effectiveness - I mean no insult). What he wrote, is a very specific guide he came up, that's his. I would include it if it was very popular in pro play, but I haven't seen that. Basically, what I'm trying to say is the all in section is basic coverage of popular all ins, and I don't want to 'rip off' a guide/all in that someone came up with, that they put a lot of effort into and is uniquely their own. I'm trying to keep this guide pretty devoid of personal styles, but rather as a solid baseline that everyone does follow in zvp.
I have seen dimaga practice it, but I'm not sure if he actually likes it or uses it. I haven't done it either, so ill try it out, but he's always been pretty open that his strats aren't meant for top tier play. I aim for this guide to be something that everyone agrees with and follows, all the way up.
On April 27 2012 13:47 KhAmun wrote: You should add in the 3hatch ling/bane allin tang did a guide on the allin section. My personal thoughts on tang aside, it's an effective build.
Well, that all-in is based on a guide/strat that Tang came up with. The all-in section is about general all ins that have been popular in the metagame at some time, and 'proven' to be effective (or rather, proven to be poular, which may or may not indicate effectiveness - I mean no insult). What he wrote, is a very specific guide he came up, that's his. I would include it if it was very popular in pro play, but I haven't seen that. Basically, what I'm trying to say is the all in section is basic coverage of popular all ins, and I don't want to 'rip off' a guide/all in that someone came up with, that they put a lot of effort into and is uniquely their own. I'm trying to keep this guide pretty devoid of personal styles, but rather as a solid baseline that everyone does follow in zvp.
I have seen dimaga practice it, but I'm not sure if he actually likes it or uses it. I haven't done it either, so ill try it out, but he's always been pretty open that his strats aren't meant for top tier play. I aim for this guide to be something that everyone agrees with and follows, all the way up.
But it isn't uniquely his own, he copied the build from zenio, who you can find replays executing said build in tang's guide, and he obv plays at the highest levels. I've executed the build a few times, and have definitely had success with it. If you're worried about ripping a build from someone else's guide, you can simply provide a link, as to not steal credit. I'm just suggesting you add the build because it's not only viable, but it goes along with your theme of attempting to making an extremely encompassing guide.
Ah, okay. I try not to watch pro games too much (try to play more with the limited time I have). Ill study up on it and add it once I try it out and learn what toss can do that makes the build work well or not.Ill be sure to give credit where its due.
On April 27 2012 14:35 Belial88 wrote: Ah, okay. I try not to watch pro games too much (try to play more with the limited time I have). Ill study up on it and add it once I try it out and learn what toss can do that makes the build work well or not.Ill be sure to give credit where its due.
When protoss goes zealot-zealot-stalker it's pretty good, but like any allin, it's a coinflip. If they were planning sentry heavy play, and their first gas units from gateway are sentry, you're definitely in some trouble.
Wow...I just played a game with one goal in mind: as high as supply as possible by 8:00 mark I made a humongous mistake forgetting ling speed AND roach speed...still crushed him with slow lings and roaches because i simply had way more stuff. Great guide, thank you. Improving mechanics one benchmark at a time
Btw, I had 68 supply @ 8:00 high gold player, former high diamond from seasons 1 & 2...after a long break i got a lot of catching up to do to get back up there!
GThanks, and nice! Good job! I struggle to hit 68 supply in mid masters (ugh, beat by a gold? You can never overestimate how shitty your macro really is).
A lot of zergs will forego liing speed for a long time - its already standard to get ling speed after lair. As long as Toss isn't gasless, you really don't need speed until 9:00 I think (if no speed, will need roaches unless toss does no timing exactly when wg finishes).
Glad the guide helped though! You should really incorporate benchmarking in all of your play. I currently try to meet 45 supply quicker and quicker vs terran expands ( as long as T expanded and didn't open 2+ rax expand, its when speed has already been started and you'd throw down macro hatch/third), 45 suppply in zvz (aggression changes this, but as long a z expanded and isn't massing lings asap and you get 2 queens, speed, 1 spine, 4-6 banes)...
On April 27 2012 16:27 Belial88 wrote: GThanks, and nice! Good job! I struggle to hit 68 supply in mid masters (ugh, beat by a gold? You can never overestimate how shitty your macro really is).
A lot of zergs will forego liing speed for a long time - its already standard to get ling speed after lair. As long as Toss isn't gasless, you really don't need speed until 9:00 I think (if no speed, will need roaches unless toss does no timing exactly when wg finishes).
Glad the guide helped though! You should really incorporate benchmarking in all of your play. I currently try to meet 45 supply quicker and quicker vs terran expands ( as long as T expanded and didn't open 2+ rax expand, its when speed has already been started and you'd throw down macro hatch/third), 45 suppply in zvz (aggression changes this, but as long a z expanded and isn't massing lings asap and you get 2 queens, speed, 1 spine, 4-6 banes)...
I wouldn't give me too much credit, the protoss applied absolutely ZERO pressure and I was able to focus 100% of my macro on attaining my high food count (aside from creep spread). I imagine in mid masters the protoss dont just let you macro uncontested every game lol
There isn't really any pressure before 8:00, maybe zealot but you see them a mile away and lings take up supply just the same, and well not everyone makes a stalker. I think I put too much emphasis in scouting. Even in build order tester its hard to get past 70. Its the whole. Having to make 4 lings, delayed hatch due to pylon, drone scouting and 2 workers to fend off scouting worker (god damn, it really cuts your drone count by 1-3 by needing a drone to ppatrol the ramp), is what sucks.
Guys, if you have read my guide already, be awae I've doubled the content, so check it out again. I have a changelog up top saying what I added!
Really good and detailed guide I can't wait for your next one. I have been using this guide firstly in YABOT to get the basic mechanics and feel down and now in the Plat league I pretty win as soon as I hit 200/200 as the other player just doesnt build enought units to defend so I don't get chance to tech to broodlords. Hopefully as I move up the ladder people will be able to defend it .
I have lost a couple of times to early voidray plays but it's more due to me not making enough queens or bad scouting I think.
^ Can you be more specific? Even if you are totally blind and just aren't thinking, a single stargate will only reveal a single stargate before 8:00 and a single queen with a 2nd queen brought from nat/main or spotted with a watch-tower ling or overlord and making a single spore or a single queen that pops halfway through the single queen vs VR fight, should be enough. I mean, you see Toss took gas, you should make that evo chamber at the standard 7:15. It's not a bad idea to throw down a ton of 3-10 spores if it's the difference between keeping the hatch or not, and you cancel them when they start to pop up (you throw down so many because he kills a few as they morph, but can't kill all of them, hence, why good macro is so important to eliminating a lot of stupid plays).
Double stargate can really screw you over of course, but I've held it plenty of times when blind by just throwing down 10+ spores at the base in question and derping over my queens to buy time, as I prepare defenses elsewhere, queue up more queens, and add a spore or two at the other bases or where I think he may go next (only need to spore up main or nat, don't need to do both - rerooting spines is a very strong tactic, after the buff a single void can't kill a single spore before it finishes rooting). Just if you see 4 gas taken, and are still in the dark at 9:30, chances are that you should be putting down at least 1 spore per base, maybe an extra queen.
Expand/Macro play is obvious as Toss tries to set up for this third (just like with 3g expand, at 5:00 they start to put a pylon on low ground, have their zealot/3xsentries at the bottom of the ramp, etc) around 9:00, gateway all-ins are obvious because he is making a bunch of units and your overseer sees he has a bunch of army units. So 4 gas, 9:30 you are still unsure, obviously need spores.
This is more specific for lower levels, but some toss open up with a 10 gate, and I guess you could elaborate on how to handle that for them. Great guide once again, even in GM it still helps alot!
Edit: also do you know what the best response to a +1 zealot 4gate with vrays is? It's the only build i feel really vulnerable too. (My biggest flaw in this is probably not getting the 6:30 roach warren to deny the zealots killing spores) and finally, zealot stalker stalker, best method to deal with this?
I did talk about 1010 gate. Any other kind of doublegate is simply much easier to deal with.
I talk about star 4 gate (don't know if I mention it by name).
As for 4 gate +1 into stargate, that's simply 2 different things. React accordingly to the +1 4 gate with 630 warren, and if toss adds a stargate afterwards (seeing it with overseer is good, but if you delay lair because upon seeing no gas because you are worried its 6 gate that's perfectly fine, you should realize you are safe to lair once you realize its obviously 4gate +1 hitting you, and then seeing initial void with overlord or when it moves across map or when first vr hits your base and well 1 vr won't kill you as you react with spores or bringing over queens as you make more AA), make queeens to 'counter' void rays.
Once you notice he isn't adding on phoenix (or realizing he will probably go with void ray + warpp gate aggression instead of voidray + phoneixes due to his 4 gate opening and not posturing to take a third), you make queens as additional AA instead of spores (no phoneixes, so make queens instead of spores) and make roaches and lings. I state in the guide you should always make roach/ling, even against stargate play, due to possibility of warp gate aggression (you can skimp/skipp on roach/ling once you rule out warp gate aggression, because he either made some phoenxies, went double stargate, or is going for his third).
A 4 gate +1, and a stargae follow up are 2 separate things, and 2 different opportunities to get ahead, or fall behind, against. Deal with each stage accordingly.
I also talk about how a common follow up to single stargate as the oppening, is 3 more gates. As long as you made an appropriate amount of queens to 'counter' void rays, and then made roaches when your 715 warren finishes (and lings as necessary), you should be fine. If you are sure to never be greedy unless you know exactly what toss is doing, you should always hold this aggression as you made queens for the void rays, spores if any phoenix came out (but there won't be any if he's going with hard wg aggression, meaning you should respond to a lack of phoenixes by making roach/ling to deal with the possibility of wg aggression), and roaches when warren pops, you should be fine.
Think of it this way: when toss is denied scouting if you took a third or not, they HAVE to make a 2nd cannon (at least). If you aren't able to 100% deduce what toss is doing at 730 (or his follow up to initial opening decision, like single star or +1 4 gate, which should be revealed by 7 to 730 overlord sac and gas scout), you NEED to make roach ling when 715 warren pops (or whenevr you took it, based upon how many gas Toss took).
As for zealot stalker stalker, you just have to make lings. If you feel toss is going to pressure you, at any time, its never a bad idea to make a spine. Every game is unique, just because I don't say something explicit in the guide about dealing with certainn pressures, doesn't mean you can't go 'oh shit, he's pressuring, I should make defences'. Spines are great, no matter how the game is going, make one if you see pressure coming.
I don't know if I can really give an answer more specific than that. If anyone else cares to answer more indepth, you are more than welcome to.
This guide isn't my strategy or say what I prefer to do. This guide is just about standard zvp. I hope that everything in this guide is standard, and what everyone generally does (obviously, say, roach/hydra is a strat, even though I don't give specifics, I outline the way its played and its goals given that at its core, its an aggression based style of play; I also explicitly state when I am stating something I feel is less standard and more my own, strong opinion, like fast third vs gateway expand or 1414 vs gateway expand or 14 pool being better than 15 pool)
.. Wait you mean 10 gate in a gate/core opener? Never heard of that. Unless I'm missing something, I imagine that's just a suboptimal opening that puts you further ahead. My initial reaction to hearing that, is still go the same 14 pool and exppand when you can and whatever - you'll easily overpower 1 zealot with 14 pool lings, and even if he kills 1 drone, you should be ahead because he cut what, 2 or 3 probes by going 10 gate instead of 13?.. although I'm also thinking a 10 gate inbase for gate/core opening won't even get a zealot to you base in time to do damage.
I don't know. To me it just sounds like a really bad opening that you don't need to worry about because as long as you macro well in the first 3 minutes up to 20 supply, going 10gatecore instead of 12 or 13 sounds suboptimal and nothing to worry about.
Again, maybe I just don't know what I'm talking about. I've never heard of 10 gate core as an opener instead of 12 or 13 gate. Or does it like try to get a super fast stalker? I don't really see a low level player making use of such an opening.. maybe a high level player could? I don't know...
10 gate used to be somewhat popular a very long time ago. There was a build where the protoss would just chrono out three zealots and go try to destroy the natural hatchery. It hit at a very inconvenient time for zerg to defend. I've seen Nony use it at least once, and I've seen idrA almost fall victim to it (his hatch survived with a few hp I think, and after the game he said he would have gged out if it had gone down - perhaps someone else can remember who that was against, possibly MC?). Compared to a 13 gate it sacrifices a little econ for either faster tech or early zealot pressure.
Interesting. I will just have to claim ignorance on that one then. I guess it should be pretty obvious that when his gateway is clearly finished when your drone scout arrives when normally the gateway is still warping in, but I imagine just making reactive lings or a spine to walk down to the natural would be enough? Won't you see the zealot walking towards you?
On April 28 2012 18:21 Belial88 wrote: Interesting. I will just have to claim ignorance on that one then. I guess it should be pretty obvious that when his gateway is clearly finished when your drone scout arrives when normally the gateway is still warping in, but I imagine just making reactive lings or a spine to walk down to the natural would be enough? Won't you see the zealot walking towards you?
Well i choose not to drone scout, but they basically chrono 3+ zealots out and try to snipe the hatch, occasionally they throw down a cannon aswell for kicks which essentially dooms the hatch, bare in mind this is high master so their micro is typically pretty good..
I decided not to say in the guide "you better fucking drone scout", but I really strongly believe it's better. I guess since not every pro does it, I can't put it in the guide (since the point of this guide is that it's all 'standard' or 'general practice' and very little of my own personal preferences besides that I obviously prefer macro play and 'safe' play rather than chancing it).
Got any replay/vod examples? Right now I'm currently digging up vod/replays to expand that section of the guide, I think I've covered literally everything in zvp so far.
I'm aware that gateway into reactive forge is autowin against hatch first, but I haven't really seen or heard about 10 gate, or gate/forge vs pool first.
I have a question and idk if i missed this, you say the counter to 2 base deathball play is either mutas/macroing better + massing roaches. But how do I know if he's going for a 2 base deathball play or is going 3 base, when is the timing i should be looking for his third base? And how should I react to two base deathball play? should I await his attack passivley or should I pressure his front, forcing him to make units so he can't focus on massing colossus while he's turtling?
I was poking around in your zvp thread and didn't see any mention of checking sentry energy in the section about scouting 1-base toss; I thought I'd chip with something there that I find really helpful.
Basically, energy regen rate maps super nicely to gateway production (25 energy [1/8 energy bar] per 45 seconds, gateway units build every 37-42 seconds), so when you poke the ramp you can more or less "back-date" his unit order by checking sentry energy. ie, sentries with energy 1/8 of a bar apart came one after the next, sentries with fresh + 1/8 of a bar came one unit ago, "gaps" of 1/8-bar between sentry energies (ie 3 sentries at 5/8, 4/8, 2/8) show gaps in sentry production, etc. Looking at it likes this gives you a really good lock on *when* he spent gas and makes any fishy trends kinda leap out ("1 stalker? when? oh just now? brb next production cycle").
Also obviously 2+ sentries with same energy is a quick and dirty way to confirm additional gates.
Just figured this would be good info to have where everyone will read it
-jumai
So what's the practical application of this?
Besdies 2+ sentries with same energy - something that sounds very cool, but then, toss will only have 1 gate running until 5:30? And if he has more than one gateway making sentries, isn't that obvious with more units too early? And he wouldn't have sentries if he threw down extra gates.
Like every 1 base cybercore build is 1 gate cyber, then additional gateways made, or not, at 5:00+ (4 gate makes them what, at 5:00 right?). So you count sentries more than energy. What you say sounds neat, but I don't know what it could tell you. 1 gate expand is at 430 nexus, so you know its 1 gate just by that, while 3gate is 530 nexus. Stargate/dt/twilight on 1 base is made obvious by lack of sentries, and if he did make a sentry, well, its only coming out of 1 gate.
Only time toss would have gateways made that go online before maybe 5:00, would be like 2 gate before core or something? I don't know. Maybe 4 gate? But that pushes when wg is done, until then, its just 1 gateway, always.
Hmm... kind of a hard question to answer directly, it's kind of more an aid for making the correct interpretaion when you get a unit count, than something to be scouting for independantly. Maybe I just like a lot of reassurance, but I always felt more confident making my droning decisions after considering his unit order once I suspected gateway aggression [past tense because I haven't seen an n-gate expand in a good while]. The most recommendable thing about the technique is probably that it's "free" information from your normal scouting.
Something to note is you use it to observe production trends (and shifts in them); basically getting a hint about what the next unit is likely to be, and not overlooking suspicious gaps in gas spending... which I guess translates best into early red flags and confirming your own suspicions. Example: scout 1 zealot, 1 stalker, 2 sentry. Personally I feel a lot more sure about things if I know when that stalker got made. Zealot/stalker/sentry/sentry looks quite passive to me, but zealot/sentry/sentry/stalker makes me wonder if he's making another stalker and gearing up for something.
Subtlety or paranoia? You decide :p
Subtlety ^^
Your welcome to bring it upp in my thread though, I think this is going somewhere.
On April 29 2012 04:56 Eee wrote: I have a question and idk if i missed this, you say the counter to 2 base deathball play is either mutas/macroing better + massing roaches. But how do I know if he's going for a 2 base deathball play or is going 3 base, when is the timing i should be looking for his third base? And how should I react to two base deathball play?
A protoss generally wants to take his third before the 11-minute mark. But you should be saturated on 3 bases by ~8-9 minutes anyway. It doesn't matter if he's going for a 3rd or for an all-in, churning out roaches counters both 2 base all-ins and attempts to take a 3rd, that's the beauty of it.
On April 29 2012 04:56 Eee wrote: should I await his attack passivley or should I pressure his front, forcing him to make units so he can't focus on massing colossus while he's turtling?
Pressuring a 2 base protoss with sentries and cannons is not worth it, you won't be able to do any damage. Use your units to deny the 3rd as long as you can instead, force him to trade units in the open between the natural and the third.
On April 28 2012 19:51 Belial88 wrote: ^ Mind explaining why you don't drone scout?
I decided not to say in the guide "you better fucking drone scout", but I really strongly believe it's better. I guess since not every pro does it, I can't put it in the guide (since the point of this guide is that it's all 'standard' or 'general practice' and very little of my own personal preferences besides that I obviously prefer macro play and 'safe' play rather than chancing it).
Got any replay/vod examples? Right now I'm currently digging up vod/replays to expand that section of the guide, I think I've covered literally everything in zvp so far.
I'm aware that gateway into reactive forge is autowin against hatch first, but I haven't really seen or heard about 10 gate, or gate/forge vs pool first.
Not drone scouting is just personal preferance, and I'm sure one of these days im going to have to start as I reckon Protoss will get cheeky and throw like 1 gas in main and 1 in nat and proceed to 6gate.
The gateway into forge is incredibly gimmicky obviously, I'll try dig up some replays for you.
On April 29 2012 04:56 Eee wrote: I have a question and idk if i missed this, you say the counter to 2 base deathball play is either mutas/macroing better + massing roaches. But how do I know if he's going for a 2 base deathball play or is going 3 base, when is the timing i should be looking for his third base? And how should I react to two base deathball play?
A protoss generally wants to take his third before the 11-minute mark. But you should be saturated on 3 bases by ~8-9 minutes anyway. It doesn't matter if he's going for a 3rd or for an all-in, churning out roaches counters both 2 base all-ins and attempts to take a 3rd, that's the beauty of it.
On April 29 2012 04:56 Eee wrote: should I await his attack passivley or should I pressure his front, forcing him to make units so he can't focus on massing colossus while he's turtling?
Pressuring a 2 base protoss with sentries and cannons is not worth it, you won't be able to do any damage. Use your units to deny the 3rd as long as you can instead, force him to trade units in the open between the natural and the third.
So if he hasn't taken a third by 11 minutes, should I be maxing out on roaches then/countering whatever he's going for or should I try to works towards that broodlord/infestor composition as fast as possible?
On April 29 2012 04:56 Eee wrote: I have a question and idk if i missed this, you say the counter to 2 base deathball play is either mutas/macroing better + massing roaches. But how do I know if he's going for a 2 base deathball play or is going 3 base, when is the timing i should be looking for his third base? And how should I react to two base deathball play?
A protoss generally wants to take his third before the 11-minute mark. But you should be saturated on 3 bases by ~8-9 minutes anyway. It doesn't matter if he's going for a 3rd or for an all-in, churning out roaches counters both 2 base all-ins and attempts to take a 3rd, that's the beauty of it.
On April 29 2012 04:56 Eee wrote: should I await his attack passivley or should I pressure his front, forcing him to make units so he can't focus on massing colossus while he's turtling?
Pressuring a 2 base protoss with sentries and cannons is not worth it, you won't be able to do any damage. Use your units to deny the 3rd as long as you can instead, force him to trade units in the open between the natural and the third.
So if he hasn't taken a third by 11 minutes, should I be maxing out on roaches then/countering whatever he's going for or should I try to works towards that broodlord/infestor composition as fast as possible?
If he's on 2 bases definitely don't go hive. Make an overseer when your lair finishes, and figure out his plan. If he's going 2base non-collossus allin, make it a goal to max out on roach/ling before he hits. If it's a collosus allin, go up to 6 gas, saturate your third, go up to spire, and try to max out on roach/corruptor/ling. If he's taking a third base, progress with your normal game plan, whatever that may be.
Host - A toss can't get cheeky and put gas at his nat. In order to get warp gate or any kind of tech in reasonable time, you have to take doublegas in main as toss ffe. Toss takes doublegas in main anywhere from 16 to 23, which is well before nexus in nat is done. If toss doesn't get 2xgas in main, that's a surefire indication he's going 2gate zealot pressure (kiwikaki style 2gate before core, gate/core/gate with very late tech, and a whole slew of aggressive, obvious builds made obvious by him having 2 gates at wallin instead of gate/core, that pros have pretty much abandoned - in my opnion, because they are bad builds).
Drone scouting is so you know he's going 2gate instead of gate/core when your overlord sees no pylon at natural, and forcing him to put all his tech into natural to avoid showing to drone, which you can easily spot with overlord later on. But I think you might be able to hod 1010 gate with 15p/16h, so I don't know. I always manage to block toss from going nexus first after scouting I went 14 pool, and it tells me I can go for a fast hatch without having to leave a drone patrolling the ramp.
Eeee - voirdoir gave you a great response, but I think your understanding of the mach up is what is limited here (read zvp game flow section).
If toss is going stargate/colossi on 2 base, he won't be able to secure his third against 10 zerlings until he reallys gets 2+ colossi. You aren't necessarily trying to all-in with max roaches (although that is a great reaction to someone who is going deathball, on its own) - what you are trying to do, is force toss to be honest before taking his third, That's why colossi on 2 base, or deathball play is so shitty.
By making just some roaches, you can deny toss from getting his third until 3+ colissi, which takes forever, and is past 13 minute. From there, you can really do anything, as toss will be too far behind to win. Nowadays, deathball play doesn't work because zergs know how to keep a lead. You can fast track to hive, or go mutas. Id really recommend anytime toss goes fast colossi, go for mutas. They are just so awesome against them, especially if you have a lead, like having delayed their third for forever.
He can't have many phoenix if he goes to robo immediately after stargate on 2 base. Just add some corruptors.
If toss is going 2 base colossus, you need to identify the timing Khamub. Naniwa came up with an extremely strong colossi all in in zvp that will kill you if you go for mutas (or that's how I understood it). If toss is goiing 2 base colossus, just mass roach/ling. Once its about 12 minute, and your econ has really kicked in hard and your tech is up, you will crush any colossus based army, and then you deny his third and just go into mutas (or whatever you want). I don't think corruptors are ever used to counter colossi anymore :O
you seem to be ignoring gateway first openings. these often require you to get gas sooner and play a little bit safer against the protoss. perhaps you should change this title to a guide against FFE, rather than a vs. Protoss guide in general?
Dude, there's a whole section about gateway first openings. How do you not see it?
This guide, up to about 2 days ago, was only a fast third vs FFE guide. Then I edited it, and more than tripled the content, and turned it into a complete zvp guide. There's a big ass section on gateway first and 1 base openings from Toss.
I said exactly what you are saying, and much more. Please, if the layout of my guide is confusing, let me know, but I think it's pretty clear where I talk about gateway expands...
Later. I don't know how 'comprehensive' it can be though, I really dislike making roaches anytime before 3 bases in ZvT (only exception is maybe against mech... but even then I prefer not to), and I think infestor play is inferior to muta play. I know that not everyone thinks this, including pros like Idra, so it's totally my personal opinion.
I really just feel that infestor play in ZvT is only viable because certain maps are just horrible for zerg, ie mutas. Maps like Antiga and Entombed, imo, favor terran so hard - lack of airspace, counterattack routes, super easy third, that mutas almost become unviable. But as a result, infestors, which I largely see as not really viable (or maybe 'shitty' is a better term to say than viable), becomes viable/not-as-shitty, because factors that make mutas suck, such as lack of airspace to make drops harder to pull off, super close together bases that make defending drops using ground units instead of air units much easier, easy third for fast hive play while most T's just go expand instead of pressure...
So I don't know how I feel about infestor play in ZvT yet. I feel anyone, of any race, can take advantage of someone turtling by simply being greedy themselves, which I dont think benefits zerg. but I feel certain maps force zerg to go infestor, because muta sucks so much on it and infestors become okay. They are still terran favored maps, but I think it's the maps that are making zergs go infestor, not because they are actually a strong strategy on their own.
So if i made a guide, it would be about ling/bane/muta into bl/infestor. Maybe I could have a guest-writer talk about infestor play or roach into third openers (something I've experimented with, but I wouldnt claim to know all about, and I prefer 2 base muta to it). I could still write a very comprehensive guide, but not an all-everything guide.
That said, I could write an awesome guide about ZvT going ling/bane/muta, and just about everything in ZvT except infestor play (scouting, 1 base play, T strats, etc).
The timings for stopping 7 gate all-ins after FFE are a bit late to be honest. You need to be making roaches about 10-20 seconds earlier then the guide states to stop a proper 7 gate all-in really. I also think checking gas at the natural is important to scout but not the only thing, checking the forge and cybercore is more important I think. Quick +1 on the forge puts toss into gateway pressure, blink +2 or possibly a fast third. No quick +1 puts toss into quick robo or stargate basically. Chrono on warpgate can often tell the difference between blink and +1 attack builds.
^ Can you provide a replay or VOD, or give a little more than "i disagree"? Making units from 8:00 seems to be more than enough.
You can't really check the forge on most maps, due to the ramp at the natural. I mention forge scouting, and it's pretty obvious what a spinning forge means. I don't need to write it in the guide to explain it to people, and I"m writing the guide based on the assumption that it's a standard map, and giving the core stuff to rely on.
Like the drone scouting section, for example. I explain it, but I give lots of information, assuming that most people don't drone scout (for whatever reason).
You don't need to see the forge to deduce what's going on, and just because it isn't spinning, doesn't mean it isn't going to be a gate all-in. I also don't think it's worth losing a ling or 2 to check for, and you can use your overlord from natural to check it as well.
I don't know if you can actually tell if it's 2 upgrades or one, either. You would have to constantly watch it to realize that it stopped upgrading, then started again, and again, it could be any sort of gate pressure if forge did 2 upgrades. I think you'd also be a little late in preparation if you figured out "oh shit i should stop droning" when you see the 2nd upgrade start.
No quick +1 could be anything. Doesnt mean quick robo or stargate.
So you are basically advocating making 6+ lings at the start of the game, 2 for watch towers, 2 checking for the probe, and 2+ to see check chronos and forge? It is just too much to lose.
Sorry, I just don't really agree with you. I think you are bringing up some good stuff though, I can see this discussion going somewhere. I just think you can deduce enough with gas scout and overlord sac, and intuition, and using a ling to scout the forge/core on maps with ramps is relying too much on something that i think is too costly and not solid enough to rely upon. I think most people can figure out for themselves, without me saying in the guide, that checking the forge/core/gate if you can on maps without a ramp, or with your overlord, is good info.
4 gate doesn't get the other 2 gas, that's the thing. You see they just don't have gas at their natural, so you need to suspect both 4 gate +1 and 6/7/8 gate. You can't really reliably figure out which of the two he's going for, so just throw down a 6:30 warren and pump out roaches. It'll be obvious which one he's doing when he either attacks with 4 zealots at 8:00, or not, and you can respond accordingly and be way ahead, because both variations put toss really far behind if he doesnt do a ton of damage.
They don't get the other 2 gas, that's the thing. If they get the other 2 gas, it's because they are doing something else. Which becomes essentially a new stage of the game. It's like saying "You should add a section about how to deal with a stargate made after a proxy 2 gate". Deal with the proxy 2 gate, or in this case, the 4 gate+1 first, and then deal with the stargate/2 gas later. Each is a different and unique opportunity to either get ahead, or fall behind.
But yea, generally toss gets the other 2 gas after 4 gate +1. That's no secret. But you should have lair done by then, and you can see for yourself what is up.
On April 29 2012 19:39 Belial88 wrote: 4 gate doesn't get the other 2 gas, that's the thing. You see they just don't have gas at their natural, so you need to suspect both 4 gate +1 and 6/7/8 gate. You can't really reliably figure out which of the two he's going for, so just throw down a 6:30 warren and pump out roaches. It'll be obvious which one he's doing when he either attacks with 4 zealots at 8:00, or not, and you can respond accordingly and be way ahead, because both variations put toss really far behind if he doesnt do a ton of damage.
They don't get the other 2 gas, that's the thing. If they get the other 2 gas, it's because they are doing something else. Which becomes essentially a new stage of the game. It's like saying "You should add a section about how to deal with a stargate made after a proxy 2 gate". Deal with the proxy 2 gate, or in this case, the 4 gate+1 first, and then deal with the stargate/2 gas later. Each is a different and unique opportunity to either get ahead, or fall behind.
But yea, generally toss gets the other 2 gas after 4 gate +1. That's no secret. But you should have lair done by then, and you can see for yourself what is up.
Thats not what I mean, what I was trying to say is the timing when a 4gate +1 does take the other 2 gasses, in order to go down their tech path.
Edit: As this gas timing could indicate whether its a gateway all in or just a 4gate +1.
Edit again =p: You say that I should have lair by that point in theory, but don't you say to not get lair if they don't have atleast 3 gas
Maybe the timings a bit arbitrary and irrelevant but it could be something to add.
They don't take their other gases, that's the thing. And I already stated that eventually, 2 base Toss will grab all of their gas, because they can support it. What I say about taking gas, is about Toss taking gas before 8:00. You should be sac'ing your overlord around 7:30, wondering whether you can drone up and tech up or have to make units. After about 8:00, you are already either going to go with units or drones, your lair is almost done so you can get an overseer to scout... it becomes less important trying to read what toss is doing, unless you still don't know. But you should know enough by 8:00.
Toss can do anything after their 4 gate +1. Use your overseer to find out. No, lair won't be done by 8:00, but after you've held their 4 gate +1, lair should be close to done, and you can use an overseer to see what decision they made
You see... like... you see no gas at toss natural, so you make a 6:30 warren. You aren't sure if it's 4 gate+1 or 6/7/8 gate. Then, 4 zealots hit you at 8:00. Now, it's obvious what's going on. As a reaction, you maybe add more roaches if you didnt start them already, but you drone up and get some tech, because you know that Toss only has 4 gateways, and no tech.
What is Toss going to do? Who knows, who cares. It doesn't matter. You know enough to make a couple roaches to hold the 4 gate aggression, and that as long as you got those roaches on time, you can drone up hard, and tech up hard to lair. Then, when lair is done, get an overseer, and see what's up, and what you should do now that you have 75 drones on 3 bases with lair tech and the requisite upgrades (speed, an upgrade, burrow or overlord speed if thats your thing, etc). Then, you decide if you want to go spire, or infestor, or whatever.
The timings are arbitrary, and I think irrelelvant. I don't think it's something to add, quite frankly. You know 100% that toss is 4 gating once the 4 gate hits at 8:00, and you know 100% it's either 4 gate or 6/7/8 gate when toss has no gas with that overlord that is by his nat and it's ~7:00 and you thinking "hm, wtf is going on, I know i can drone up for sure to 8:00, but I'm not sure if I need what to do after that point".
That's all you really need to know... that he went 4 gate... and you got the roaches to defend adequately... and then you just drone up hard and tech hard with your advantage and use the overseer to figure out what toss is going to do next... if you can figure out what he's doing because you counted his probes in the natural, great. more power to you.
i feel like this isn't that hard to understand... but i feel like you don't understand. then i make a huge post. who cares what toss is doing, you held his 4gate +1, drone up and tech up hard. it doesnt matter what he's doing, the correct response to holding 4 gate +1 is droning and teching super hard until lair finishes. Then you send overseer to figure out what's up.
I'm having a hard time getting reps/vods right now. It's extremely time consuming to try to watch every zvp in gsl, most of them aren't good textbook examples.
I list the reps I could use, in the changelog. If anyone recalls a certain game that went in that way, please msg me.
I added a rep of me going fast third vs 3 gate expand, against some sucker who did it against me on ladder.
He was probably bad, but man the metagame has really evolved... I remember I used to insist that 3 gate expand was just such a better build than FFE. I think I said that because I saw 2 pro games, that really stuck out, where zerg tried to go fast third vs 3 gate expand, and got raped. It may have been a combination of the map (XNC, rocks blocking away the third...) and just people not knowing how to play back then i guess.
Completely changed my ZvP for the better, thanks! I don't know whether this question has been asked before but how would you suggest transitioning from muta-ling aggression to hive tech? (Preferably brood-lords) I can't really see a timing window to make drones (for the spine crawlers) and keeping a healthy ling/muta count.
Great Belial! thanks a bunch man, tried it tonight and its fixed my whole outlook on the MU
Thanks! What specifically did you find helpful? Have you applied anything in the guide, that you didn't do before?
I'd be interested in some constructive criticism, as well as feedback. I'm kind of looking for the following:
1. Do you agree with everything? This guide is supposed to be a guide on basics, I tried to leave out personal opinion as much as possible. I think that everything in this guide is standard, and I explicitly state when it's my personal opinions coming in (eg: I think some pros like 2 base lair vs gateway expands, but I stated I think 3 base w/speed is better, don't make infestors vs 2 base toss, but I still explain when they are useful). People should not be directing questions towards me, just to anyone who can answer them. I hope that everyone agrees with everything stated here, it's other things that are personal opinion (eg I like going banelings and staying on lair tech and mass muta instead of going to hive, I like to end the game with roach/infestor instead of going hive, I like roach/corruptor so I can transition quickly into broodlords without infestor support, etc).
2. Do you like the format? Anything that could make it better? Zatic said he prefers an open style guide instead of spoilers, but I really think spoilering is the way to go. I wish I could put in more pictures, I just don't know what (even that inject pic is kind of... general)
3. What has helped you the most? Anything in the guide that you've applied?
What do you do against an 8 gate +1 all-in that is at your 3rd by 8:30 with 8 stalkers 7 zealots and warpgates almost back off cooldown? I just got crushed from what (I believe) was Golden offracing as protoss. I know I didnt play perfectly, and I blame that on the fact that I havent played much in a long while... but it seems to counter this build really damned well (that and hes kinda good). http://drop.sc/171466
Your APM is an average of 20 too. While APM isn't really that important, I can already see you missing some critical things. You don't correct your first overlord to go towards the opponent's base as soon as you find out where he spawned, and that he didn't spawn at the location your overlord went. In fact you send zero overlords towards his base at all. I have no idea how you are supposed to figure anything out by this, but you are definitely not going to know in time that he's doing an all-in. I really can't see you winning if you don't have an overlord by his natural. That's all you needed to know - to see zero gas at his nat, but you didn't do that. If you aren't prepared for such a heavy all-in, there is no way for you to defend it in time.
You have to really cut tech and econ to defend an 8 gate, and just having an overlord over his natural's gas would have told you he was doing something that requires a 6:30 roach warren with immediate roach/ling production when the warren pops. It would not be natural to make these decisions unless you saw no gas at the toss natural. You make no attempt to scout. You need to seriously cut workers and drone production to stop a 7 rax all-in or a 4 gate right? Well, you aren't scouting in any way that would lead you to make the necessary cuts to survive.
You get supply blocked at 36.
You make your 28 overlord wayyyyyyyyy too early. This is a really basic macro problem. You avoided a supply block, so this would be acceptable in platinum, diamond, but you are in (low) masters now. You really shouldn't be making such a gaping macro mistake as to make your 28 overlord completely at the wrong time. This is really, really basic. It doesn't matter how good or bad or new you are, every zerg should be making the 28 overlord at 25-27, ideally 26 (unless your natural was taken super late or you were planning to make a building at 27-28, like a roach warren, evo, spine, or gas). I can't stress enough how your macro is already suffering, I'd say this cuts you down by 2 drones then and there.
At 23 you are banking money because you aren't spending larva either. I don't know what this is about, but you really shouldn't be making such basic mistakes. You aren't making probes. This is a problem. You actually let that larva lay around for about 15 seconds, and the 2nd larva for 5. What is going on here?
This may be an issue of your 20 apm. I don't know how you are in masters with 20 apm. Sorry. APM really isn't that big an issue, but bronzes have more than 20 apm. Do you have some sort of handicap, or were you preoccupied? That's fine, just spam that sdsdsdsdsd at the start against FFE. You need to get those drones out.
Late on 2nd inject. Not too big a deal, but it's little... irritating, I suppose.
You just always have idle larva. From about 24 supply onwards the rest of the game, you have idle larva. This is something that is easier for T/P to analyze (no worker being produced, nothing being produced at rax, missing wg) than it is for Z (idle larva counter is fairly new), but you need to start watching your reps and watching how long you have idle larva for. Just watch your reps, with the units tab open, and see if you ever have larva. You should never have larva.
You get supply blocked at 36/36. Not the worst, but you made that overlord at 34. I'd recommend making it at 32 or 33. Not the biggest deal, but since you are in masters, this is important.
You make a lair without scouting if Toss was doing some sort of gasless all-in... like 8 gate.
That's basically why you lost. Not to mention no ling speed started yet, no roach warren started yet. I don't see you winning.
That's really all there is to it. Some basic macro problems that are really a big part of why you lost, but you not scouting and making blind decisions is bad. Making a 7:00 roach warren, standard. 2 gas at 6:00, standard. Taking a third, 3 queens, standard.
Making a lair? You can't do that until you make sure the opponent isn't gasless at his natural. In such cases, you have to start ling speed and roach production before lair. Really, I'd say going lair with your first 100 gas is an autoloss to some gasless toss builds - you can survive +1 4 gate pressure maybe if you go lair with the first 100 gas as long as you get roaches asap from a 6:30 warren and probably threw a spine up, but you can't survive a 6/7/8 gate all-in.
Just scout. Not better, just scout. At all. Clean up your macro too, it's questionable you would have held even if you did know exactly what was going on, and responded correctly. (61 is probably the lowest supply you can be without actually automatically losing because you are so far behind...).
On another note - your opponent executed his build quite well. At exactly 8:00 he was warping in. If he was good, he would have been on your creep and pressuring with his zealot/zealot/stalkers at 8:00, but oh well. He did a build, he didn't execute it like a pro, but he executed it cleanly and competently, and his macro was solid. He basically just played better than you, as bad as that sounds in a game where absolutely nothing happened up until 8:30 and zero scouting occurred on both sides. But the favor in those sorts of games goes to the guy doing an all-in against someone who is being extremely greedy.
You can't just drone up against a 4 gate and never scout, and make decisions that must be made only when ruling out a 4 gate. Similiarly, you can't just drone up hard against a 6/7/8 gate all-in and do things like make lair and not get ling speed without first ruling out a 6/7/8 gate all-in.
By the end of the game, your APM was in the 30's while his was 100. He definitely wasn't pro, the average APM of low masters is about 80ish, sometimes up to 100.
Everything I say on how to handle such an all-in is in my guide. I even have a pro VOD example on how to hold such an all-in. All the information you need is quite clear in the guide.
And no, a 8 gate +1 does not counter a fast third at all. What it does counter, is someone who just doesn't scout at all and blindly plays greedy. A 4 gate does not counter someone who takes their natural, but it does counter someone who doesn't start preparing defenses by, at the latest, 35 supply (really, I'd say 30 supply is when you need to start cutting against a 4gate, especially on maps that are wide open).
I really appreciate the huge reply, but I was just looking for a generic response for future games. I already said I know I made mistakes as I havent played in weeks, so telling me to correct those werent really needed. I mean, in the guide it says to cut drone production at 8 minutes given what I scouted (an 8 gate all in, though the guide doesnt mention +1 or stalkers in such a build) and I did just that, and he had units basically all over my base by then and I didnt think that even if I corrected mistakes that an 8 minute cut-off would be suitable. Scouting something like that perhaps I should be starting units at 7:40 in the future? Would there be enough roaches and lings to hold it off?
On an aside, dont look at my APM... dont even bother. I dont spam at all (I have issues with my wrists, so I prefer not to spam) so my APM in the early stages of the game is quite low. In a 15-20 minute game the average goes up to 100 (which means its higher than that if you exclude the early game low APM).
Also, he may not be the korean Golden, but hes definitely up there considering he was grandmaster in past seasons and has replays winning against dimaga.
Well I can't say anything better than what's spelled out in the guide on how to handle it. But I'll reply to your points
1. I slightly edited the guide because of +1 4gate timing pressure at 8:00. Where against 6/7/8 gate all-ins you can get away with a 7:15 roach warren and then just making units after 8:00, with +1 4gate, you really need to make that roach warren at 6:30 and make those roaches at 7:30 or you'll take heavy losses (a spine can buy time if you made your roach warren later, but any other response that doesn't clearly leave you behind, like 'yea i survived but i lost a bunch of drones or my third or my queen and units', is suboptimal, like massing slowlings from the 6:45 mark or making a ton of spines at 7:00 and hoping he doesn't walk around them or to another base).
You really just need to see if he took gas. If he didn't, you need to make that 6:30 roach warren and make roaches when it pops. If he doesn't hit you with 4 zealots and obviously only 4 warp gates online, not 6/7/8, at 8:00, then you can maybe take your lair or more drones instead of more units, but if it's not at 8:00 but rather at 8:30, you will have to stick with 100% unit production.
If it was a perfect world and you got an overlord in, and saw he made 6/7/8 gates, not 4, yea, then make that 7:15 roach warren and drone up to 8:00, then make units. If it's not a perfect world, oh well, he's so all-in, the fact you went 45 seconds earlier roach warren will hardly be an issue because you hold.
I'm sorry that the guide didn't reflect this earlier on, and that I wrote off +1 4gate pressure, but you can see a lot of people were concerned about 4 gate +1 and that you really need roaches against it, and I did eventually edit it, as well as admit I was wrong. There are VOD examples of both builds that gasless toss can go, 6/7/8 gate and 4 gate +1.
2. Scouting things like a spinning forge or a stalkers are just hints, but there are 2 problems with doing so: 1)It does not definitively tell you anything (he could get a stalkers or +1 for any opening, and he can not get an upgrade or stalker and go for 6/7/8 gate all-in still, such as no-upgrade 7 gate all-in 2. You can't reliably scout these things. Toss can hide the stalker inside his base right until he pushes out, which would then be too late for you to react, and wasting lings to die to run up a ramp to see if his forge is spinning, is not worth the cost.
All you had to do to definitively scout what was going on, was see he had no gas at his natural. That's all you had to know, to know enough to go "Oh, I should get ling speed instead of lair with my first 100 gas, and I should get a 6:30 roach warren and start pumping units when the roach warren pops, and if I hold, I end up way ahead of Toss that I can almost always close the game if I respond perfectly, and after reading Belial's Comprehensive Guide to Everything ZvP!, I know what the perfect responses are!"
3. Yes, you did cut drone production, but I wouldn't say that was good decision making on your part. You were 100% blind to what toss was doing, for all you know he could have been going double stargate, in which case making roaches at 8:00 would be a bad choice. Furthermore, you really got punished because you went lair and skipped ling speed, and you really didn't make enough units. I don't think you started 100% unit production at 8:00 though, but whatever.
4. You didn't scout anything. By the time you sent an overlord in, or saw his stalker had +1, you had well already started lair and neglected ling speed. It was too late. You realize the +1 is timed to finish when he's actually attacking right?
You also never saw his stalker, i don't know why you say that. The first protoss unit you saw besides a probe, was 10+ stalkers and zealots in your third. You didn't even know he was breaking the rocks at your third.
5. I told you what to do. Send your first overlords over to Toss' base. You should re-read the section I have on scouting toss (it's been updated as well). I could really say the #1 reason you lost the game was because you never re-corrected your initial overlord to go to Toss' natural's gas. I highly doubt you would have held even if you knew exactly what was coming, because of your macro, I really do, but since you are probably aware of your macro deficiencies, and because that guy was obviously a much better player (i dont care if you weren't warmed up, he had triple the APM of you, and in general, evenly skilled ZvP, Z tends to average about 20-30 more average APM than the Toss throughout the game), I'll just say you really had no chance when you had no clue what Toss was doing, and 100% what you needed to know would have been revealed by having that initial overlord, over his natural's gas.
I mean, do you just 100% lost to double stargate and DTs? You really never made any effort to scout until it was too late. You sent an overlord toward his base (and you sent that overlord from your main, it wasn't even an overlord that was prepared next to his base!) AFTER you decided to not get ling speed, not get an earlier roach warren, and had started lair.
let me break it down like this: Right before you get 100 gas, you need to scout for: 1. Does toss have gas? = Do I need to get ling speed first, or can I get lair first? 2. Sac an overlord and see his base is full of gateways or is empty = make some roaches when warren pops, to making only drones against double stargate.
6. I will look at your APM. It's not that important, but considering you are masters, yea, it's kind of relevant. If you had a very low APM of 50 or 60, I would not have mentioned it. But your opponent had 99 average APM throughout the game, and you had 30.
I mean let's look at the very, very obvious consequences of this: 1. You had idle larva all the time. No drone production, low supply. 2. You didn't scout at all, you didn't even bother to redirect your initial overlord to his base when you realized where he spawned 3. You lost your scouting worker to a probe or something as equally silly... 4. You didn't realize he was breaking the rocks to your third 5. You didn't have overlords around his base, to be ready to sacrifice at the 7:00 mark, instead of sending an overlord from your main at 8:00 mark. 6. No ling in front of his base to see him push out 7. Just zero scouting at all.
It's not about spamming. I don't spam at all either, and my average APM according to sc2gears is about 151 (that's not the sc2 APM, i think sc2 apm is like 80-90), and EAPM is 101, so that's very little spamming (ranked against everyone I've played on ladder, only 2 people have a higher APM than me while having an equal or higher EAPM, the overwhelming majority only have higher APM because their EAPM, useful actions, is total crap and they spam a lot, or .. well that's it. I guess I'm a bit faster than most people, and I don't spam). But that Toss didn't spam at all, and his apm was over 90 (sc2 apm, although that's higher than my sc2 apm except in tense zvt games).
You had 30 apm, and for the majority of the game, it was 20. I don't think you realize how low that is. That is what bronze-gold players have. I get you don't spam, if you spammed it would be higher. But I don't think it's really possible to execute basic macro without at least 50 apm, 60 with zerg (day9's words, not mine). 60 APM is not considered high at all.
Whatever. It's fine. We can't fix your APM overnight, and spamming won't solve the problem. Let me just put it this way:
PUT A GODDAMN OVERLORD BY A TOSS NATURAL'S GAS. Ideally, the first one you start with, as soon as you find out where Toss spawned. Your lack of APM is what caused this, your lack of any thought to scout is what cemented it.
Also, he may not be the korean Golden, but hes definitely up there considering he was grandmaster in past seasons and has replays winning against dimaga.
cool. If you want to rationalize the loss by saying he's a ladder hero, instead of focusing that you could have very easily held if you just had an overlord by his natural's gas, that's fine.
edit: please don't take it that i'm harsh here. I'm really not trying to be mean, although I'm sure I've probably pissed you off by now. Please understand that what may seem like aggression, is really just my attempt to stress how easy some of these problems could be to solve for you. You did okay, really if you had an overlord by his natural, realized he was doing some sort of 8-8:30 gateway aggression that would very much put him super far behind if he doesn't do damage, and thus skip lair in favor of ling speed and mass roaches, you would have just won the game.
Also, he may not be the korean Golden, but hes definitely up there considering he was grandmaster in past seasons and has replays winning against dimaga.
cool. If you want to rationalize the loss by saying he's a ladder hero, instead of focusing that you could have very easily held if you just had an overlord by his natural's gas, that's fine.
I am not trying to "rationalize" the loss. He is 58-2 this season and he has been grandmaster before and was zerg in previous seasons and beats dimaga with zerg. If you think he is a ladder hero, then he is one hell of a ladder hero.
Little bits of your response like this (and the apm thing, which you hugely overstate the importance of and hugely misunderstand for the first ~6 minutes of a macro game, like I say I average 100 in games that are 15 minutes or longer... which means my average when I have stuff going on is 130+) are why I do not like your response.
EDIT: SC2gears has his apm at 200 btw, or 277 BW apm depending on which setting you want to use
,You didn't lose because the guy was amazing. You lost because you made zero effort to scout. If an overlord was by his naturals gas, which I even state is the number 1 priority to do when going fast third (only behind taking a third in terms of importance), you would havve gotten ling speed instead of lair, and a 630 roach warren, and you would have been prepared.
I'm not saying your macro wasn't good. I'm saying your macro was horrible. Whatever, we all screw up, but you need to understand that your complete fail at your 3rd and 4th overlords is completely unacceptable. I don't know why you write these off, these are mistakes you should never see in diamond. Its a big deal. You basically got supply blocked with your 3rd overlord. That's insane.
I question that you would have held at all if you knew wat was up, given your macro. There were lots of smaller mistakes in your macro, but Id say fix that 3rd and 4th overlord, put an overlord by his naturals gas, and you would have easily won that game. Even against a ladder hero of that calibre. These aren't hard things to do. Anyone can make these fixes overnight.
That's great your APM is higher in other games. You already stated that you were having an off game. And look, we all do. I'm just tellling you what I saw. If your APM was where it normally is, according to what you are telling me, then you should win easily. I just don't think with 20apm, you can pull off a fast third build by constantly making workers and overlords and setting an overlord by his main and nat. And obviously, as the rep shows, your low APM was a part of your loss - you constantly had idle larva, you constantly made overlords at the wrong time, you never scouted, you didn't even move the initial overlord you start with when it went to the wrong base.
The low APM was why your supply was so ridiculously low at the 8:00 mark. You just never made drones.
I don't know what you don't like... you can watch your rep, see you averaged 20 apm and know you were off that game, and just go 'doh' and move on. I don't think the loss should be as confusing as it is was to you. You played against someone with 3x the apm of yours, and your apm was at 20! 50 is low apm, 20 is just odd. But okay, you were having an off game. Armed with new knowledge, you will beat any ex-GM with even 20 apm on an off day who tries to do a 7 gate allin next time.
Its just odd that you are being so defensive about a game where you know you wrere completely off your game, and your oponent was clearly above your level. I think most people wouldn't think twice of the game, I wouldn't think twice about losing to naniwa doing a 4gate.
My guide, as well as my responses, tell you clearly what you should have done, and what your mistakes were. Even before the major edits, the guide tells you what you should have done. It tells you that going lair, not getting ling speed, and not seeing no gas at his natural, is an autoloss - there's even a pro vod example. Its like you didn't read my guide at all. And then you say the build counters fast third.
Please. Put more effort into it, if you read my guide you would know that you lost because you never made any effort to scout, you didn't put an overlord by his gas, and you made a lair, and you skipped ling speed. I even have a whole section saying to check your supply at 8 minute, and yours was quite low, low enough that it would be hard to win no matter what he did.
Really loved the guide and the previous answers especially have been useful to me, thanks :-)
On May 05 2012 17:08 Belial88 wrote: Its just odd that you are being so defensive about a game where you know you wrere completely off your game, and your oponent was clearly above your level. I think most people wouldn't think twice of the game, I wouldn't think twice about losing to naniwa doing a 4gate.
My guide, as well as my responses, tell you clearly what you should have done, and what your mistakes were. Even before the major edits, the guide tells you what you should have done. It tells you that going lair, not getting ling speed, and not seeing no gas at his natural, is an autoloss - there's even a pro vod example. Its like you didn't read my guide at all. And then you say the build counters fast third.
Please. Put more effort into it, if you read my guide you would know that you lost because you never made any effort to scout, you didn't put an overlord by his gas, and you made a lair, and you skipped ling speed. I even have a whole section saying to check your supply at 8 minute, and yours was quite low, low enough that it would be hard to win no matter what he did.
Everyone appreciates the time you put into writing those replies but you could be a little bit more considerate... I mean they know how bad they are or can be, but rubbing their face in it is exactly how you get people to become defensive.
This has been incredibly helpful for me. I'm still not at the stage where I can remember every part of how to respond correctly to scouting something and transitioning out of it smoothly but it's gone from an annoying matchup that I had a good win ratio in just because of being able to spend resources better, to my most comfortable one. Definitely one of the best guides I've read and I really appreciate the time and effort you spend in responding to people's queries.
Send initial overlord to where his natural gas is. so important! 3 bases, duh. 2x gas 6:00 (this is right about when third finishes, in the 30's of supply) 7:00/~45-50 what kind of gas do you see? Try to sac an overlord, poke in. Scout basically. 7:00-7:30/~50+ roach warren, evo - If no gas, a spine at third, speed then roaches, and lots of units until you know what's going on. Otherwise, lair then speed Make a handful of roaches when warren pops Drone up to 9:00, 2 more gas, macro hatch If Toss has 4 gas, and you don't see any gateway units, and it's oddly quiet, put a spore at each base and maybe some extra queens. If it was single stargate, don't overreact (3 spores at once, oh god that's so bad when it's obvious his void ray is all away across the map killing an overlord and it's the only stargate unit out). Just get single queen out where you think he may attack first, add a spore at third or main when a 2nd stargate unit is revealed, your okay. Just add incrementally, is the key. 9:00, unless he was doing DT/SG, or you know exactly what is going on, make a lot of roaches and lings until you can get an overseer to run through his base and see what's up. Don't grab more than 4 gas, or get any tech, until toss is posturing to expand (in most games it seems, toss will all-in and you have to fight to hold, and dont tech until you kill his whole army or he's trying to take his third). Only exception I'd say to this, is double stargate or colossi on 2 base. Get roach speed and an upgrade asap. Things like burrow, burrow move, overlord speed, are useful too.
Just play some games against AI, and figure out what 7:00, 8:00, 9:00 mean to you. It's hard to understand time, but it's easy to understand "oh, I should check the clock when third pops, because I know around then something important happens".
Belial, I have not once argued that my play wasnt flawed. I havent even argued about what I should have done... why? Because I know these things already. I state in my original post that I screwed up a fair amount. I am good enough to know my own mistakes. It is straight up rude that you think you know this other stuff though, I dare say even condescending and arrogant.
I am being defensive because you are actually talking shit about me and my play, not giving advice. Then you keep bringing up irrelevant points. If somebody has 50x my apm in a 6 minute game where there was little to no micro or actions necessary, I didnt lose because my APM was low. I lost because he outplayed me. High APM in the beginning stages of a macro game is NOT NECESSARY. Why do you think you see people completely take their hands off the keyboard even 2 or 3 minutes into the game to stretch their wrists or any number of things? Because APM that early is purely fluff/warmup.
I also think its hilarious as all hell that you now state "Armed with new knowledge, you will beat any ex-GM with even 20 apm on an off day who tries to do a 7 gate allin next time." Really? Initially I must have equal or greater APM to my opponent but now APM doesnt matter!
I appreciate the advice (and keep in mind that I am heeding the actual advice given), but what you are saying is just rude and in the case of APM and my opponents level is just wrong.
Well, you stated in your first post that you thought that 6/7/8 gate all-in counters fast third play, so what I took from that was that on a strategic level, there is no way to hold it. Quite the contrary, 6/7/8 gates are seen as somewhat outdated these days, and considered quite basic and easy to hold (i think when losira when fast third against anypro or hongun or whatever all-in cheese toss it was on dual sight small map, and held his 7 gate all-in, it validated for everyone that if you can hold this most extreme of all-ins in such a convincing and straightforward way, you can hold anything, and so it became standard, and I'm sure every zerg in the world was practicing how to hold it as every toss learned to execute it).
So I pointed out 2 problems in your play - not just the strategic, but your macro as well, as that is probably a big reason why you thought it was such a bo counter.
I don't know what other stuff you think is rude or presumptuous. I'm just pointing out the flaws in that game. I don't think I've impressed upon you how basic and major a problem it is that you didn't redirect your original overlord to go towards his base, and you make your 28 overlord ridiculously early and your 36 overlord majorly supply blocked (essentially, you got supply blocked every single time after your 18 overlord). I also pointed that I thought 20 APM was a big factor in this, you largely made drones and put your buildings down on time, and you weren't THAT far behind at 8:00, so you kept up in those areas while faulting in scouting and overlord production. Did you have an off game, or you normally play with 200+ apm, cool, whatever, then I'm sure you've held this all-in easily many times before.
I mean, in my guide, I state very clearly - send overlord to natural gas, when to make your first 5 overlords, how to hold a 6/7/8 gate all-in, and a vod example. It was clear you didn't read my guide at all, because this are very simple issues that you could easily identify and have fixed if you read the guide.
I'm not trying to talk shit about you, you are obviously a competent player to be in masters and to face such an opponent. I never said any ad hom toward you, I'm sorry you feel that way. I really can't stress enough though, how macro was such a huge factor in your loss, that you had 20 apm and that may explain a lot, and that you made some very basic mistakes.
I mean, maybe you could have said "obviously i was off in this game, so I get why I lost, but I was wondering what to do when I come across this build" or something. I don't mean to bash on you because you actually provided a replay, but in terms of "why did I lose this game?", i think it should have been very clear.
I really do think you lost because of your APM. I don't know what to tell you, 20 APM average for a 10 minute game is where you see bronze-gold sit at. Day9 has said that it takes about 50 apm to pull off basic building, and that this is very low. I really think you made such major mistakes like not sending your initial overlord toward his base when you figured out where he spawned, losing your scouting drone to a probe, not sending a single overlord towards his base, not putting a ling in front of his base, and ALWAYS having idle larva, is a result of your 20 APM in that game. I don't think APM is important, but at masters, I think you really need at least 60 APM to get by.
I mean you never made drones. You just had idle larva all the time, missed injects, overlords were always off. That's fine, everyone does that except high masters, but you literally always had idle larva.
Yea, he outplayed you. But I think the problem with your attitude here, is that you are so intent on focusing on him. But that's not the issue. If YOU had fixed some very minor issues, you could have won.
I mean, if the guy went 5 gate robo, or 5 gate blink, into third, or immortal/sentry all-in, or some type of build where skill can come through, that's fine.
But he literally did a very plain build order, very much similar to a 4 gate, and he won the game instantly. At 8:30 he was in your base, and you just lost right then and there. There was no contest at all, any diamond could have won just by executing the same build, even if his attack came 30 seconds later.
You would have beat this very, very high ranked player, if you just: 1. put an overlord over his natural's gas 2. Not get supply blocked at 28 and 36.
That's it! Very straightforward. But you blindly made a lair, when making a lair before speed is an autoloss to 6/7/8 gate. You never watched his gas, which means you autolose to 6/7/8 gate because you really have to prepare against it by taking very extraordinary measures (no lair, earlier roach warren, no droning after 8).
The game would have been very clear cut if these mistakes didn't occur. You would have gotten ling speed, and would have made enough units to hold his initial 8:30 hit, no matter how good he is.
Maybe if it was be more of an issue if it was a 3 minute back and forth. But it wasnt. He followed a build order. At 8:30 he was at your base. You instantly lose your third because you blindly made a lair and skippped ling speed.
Please. It's not about your opponent at all in this game. If it was a macro game, sure. But he basically did a 4 gate. Now, yea, we would all lose to naniwa's 4 gate. But none of us would lose to someone who was favored on ladder, doing a 4 gate, if we scouted it, by just seeing he doesnt take his gas - or at least, not immediately.
I also think its hilarious as all hell that you now state "Armed with new knowledge, you will beat any ex-GM with even 20 apm on an off day who tries to do a 7 gate allin next time." Really? Initially I must have equal or greater APM to my opponent but now APM doesnt matter!
You stated that normally, your APM is 130+, and that your basic macro mistakes were things that *never* happen to you,and you were just off. Trusting what you say here, I think with 130+ APM you will be more than capable of holding any 6/7/8 gate, even a GM level one, when you have an overlord over his natural seeing he didn't take gas, and pumping roaches and lings after a 6:30 roach warren (or even 7 minute) while you skipped lair in favor of ling speed. If anything, it'll be a close game where you lose after a very intensive battle.
You certaintly won't just flop the second you see his units the first time.
I appreciate the advice (and keep in mind that I am heeding the actual advice given), but what you are saying is just rude and in the case of APM and my opponents level is just wrong.
I don't know if you are heeding them. You had some very real problems in this game, and I feel you aren't taking them seriously at all. You are just writing them off by saying "oh, he was such a better player".
Look, you know the merits of 60+ APM, because you yourself generally average 100+. So I know that you understand when I say that 20 average APM throughout a 10 minute game in high masters is not enough, and that if you know an all-in is coming and you prepare to defend it, you should win and hold.
It doesn't matter if the person is GM level and is doing a double starport banshee build - you know the right response, now you know the 100% surefire way to always scout it, and you will beat it everytime now.
I do not say anything rude here, and I would appreciate if you leave out the ad homs. This is my personal opinion on the game that was presented, i really think the opponent's skill level was not the issue, and I really do believe that 20 APM was a major factor in why you made such glaring mistakes like not sending a single overlord ever, towards his base.
Please. Understand you were the one that lost, not that he won. There was no fight or struggle in this game, he literally rolled over you because you got supply blocked at 28, 36, and had 20 APM and not enough spare APM to scout at all, and have an overlord by his natural's gas.
If you want to disagree with my analysis, that's fine. If you refuse to acknowledge why you lost, or how to improve, that's fine. This is just my analysis, I'm sure others will say 20 APM against a high masters will be more than enough APM, and that blindly being so greedy and skipping ling speed and making a lair and not scouting his natural's gas at all is okay.
On May 05 2012 20:39 Belial88 wrote: I'm glad you like the guide, thanks.
I've been struggling recently with the gasless FFE into early +1 zealots and got all the answers when coming to this thread, and I'm pretty sure I'll be back every once in a while :D
You talk about how macro is so important here, and I think it's not only macro. I noticed my play getting significantly better every time I improved my basics:
- A year ago I spent lots of time working on my APM's, and even if I wasted most of it eventually it helped me tremendously at getting a step higher - After that I worked on my creep spread and keeping all those bases saturated, and forced myself to play really simple and standard BO's - After that I practiced a lot to find out most suitable way to keep up with larvae injects, and worked on a routine to do all the usual tasks without feeling overwhelmed - Right now working on the concept of always trading armies to keep the other on a low count of key units (still new at this, lost too many games because I want to macro hard and let the other guy alone to build up his death ball) - And finally I'm looking into BO's to diversify my play and the current meta game as well (your thread is an amazing resource for this especially)
Amazingly enough, the trick that helped me a lot recently to improve my play was the Shit + # to add eggs to my groups of units, super important for reinforcements... How have I been able to play before without that? :D
Note that even though I don't know jack about BO's and stuff, all the basics I have covered got me fairly easily into top diamonds (though I have to admit the main reason why I can't get higher than diamonds is because my macro is pretty bad, always end up with thousands in the bank :/).
Amazingly enough, the trick that helped me a lot recently to improve my play was the Shit + # to add eggs to my groups of units, super important for reinforcements... How have I been able to play before without that? :D
haha, before i knew about that I would ctrl+click eggs I morphed, so all teh overlords or unused larva would be added and id just have random shit with my army. I really like the shift+ctrl+click, to remove a class of units froma control group (accidental overlords, when queens somehow get hotkeyed with my mutas or hatches, setting up flanks or just parts of my army to go counterattack, banelings and morphing bane sin ling/bane).
Thanks for your feedback! if you have any trouble in zvp, post a rep in here.
On May 06 2012 00:29 Belial88 wrote: Well, you stated in your first post that you thought that 6/7/8 gate all-in counters fast third play, so what I took from that was that on a strategic level, there is no way to hold it. Quite the contrary, 6/7/8 gates are seen as somewhat outdated these days, and considered quite basic and easy to hold (i think when losira when fast third against anypro or hongun or whatever all-in cheese toss it was on dual sight small map, and held his 7 gate all-in, it validated for everyone that if you can hold this most extreme of all-ins in such a convincing and straightforward way, you can hold anything, and so it became standard, and I'm sure every zerg in the world was practicing how to hold it as every toss learned to execute it).
So I pointed out 2 problems in your play - not just the strategic, but your macro as well, as that is probably a big reason why you thought it was such a bo counter.
I don't know what other stuff you think is rude or presumptuous. I'm just pointing out the flaws in that game. I don't think I've impressed upon you how basic and major a problem it is that you didn't redirect your original overlord to go towards his base, and you make your 28 overlord ridiculously early and your 36 overlord majorly supply blocked (essentially, you got supply blocked every single time after your 18 overlord). I also pointed that I thought 20 APM was a big factor in this, you largely made drones and put your buildings down on time, and you weren't THAT far behind at 8:00, so you kept up in those areas while faulting in scouting and overlord production. Did you have an off game, or you normally play with 200+ apm, cool, whatever, then I'm sure you've held this all-in easily many times before.
I mean, in my guide, I state very clearly - send overlord to natural gas, when to make your first 5 overlords, how to hold a 6/7/8 gate all-in, and a vod example. It was clear you didn't read my guide at all, because this are very simple issues that you could easily identify and have fixed if you read the guide.
I'm not trying to talk shit about you, you are obviously a competent player to be in masters and to face such an opponent. I never said any ad hom toward you, I'm sorry you feel that way. I really can't stress enough though, how macro was such a huge factor in your loss, that you had 20 apm and that may explain a lot, and that you made some very basic mistakes.
I mean, maybe you could have said "obviously i was off in this game, so I get why I lost, but I was wondering what to do when I come across this build" or something. I don't mean to bash on you because you actually provided a replay, but in terms of "why did I lose this game?", i think it should have been very clear.
I really do think you lost because of your APM. I don't know what to tell you, 20 APM average for a 10 minute game is where you see bronze-gold sit at. Day9 has said that it takes about 50 apm to pull off basic building, and that this is very low. I really think you made such major mistakes like not sending your initial overlord toward his base when you figured out where he spawned, losing your scouting drone to a probe, not sending a single overlord towards his base, not putting a ling in front of his base, and ALWAYS having idle larva, is a result of your 20 APM in that game. I don't think APM is important, but at masters, I think you really need at least 60 APM to get by.
I mean you never made drones. You just had idle larva all the time, missed injects, overlords were always off. That's fine, everyone does that except high masters, but you literally always had idle larva.
Yea, he outplayed you. But I think the problem with your attitude here, is that you are so intent on focusing on him. But that's not the issue. If YOU had fixed some very minor issues, you could have won.
I mean, if the guy went 5 gate robo, or 5 gate blink, into third, or immortal/sentry all-in, or some type of build where skill can come through, that's fine.
But he literally did a very plain build order, very much similar to a 4 gate, and he won the game instantly. At 8:30 he was in your base, and you just lost right then and there. There was no contest at all, any diamond could have won just by executing the same build, even if his attack came 30 seconds later.
You would have beat this very, very high ranked player, if you just: 1. put an overlord over his natural's gas 2. Not get supply blocked at 28 and 36.
That's it! Very straightforward. But you blindly made a lair, when making a lair before speed is an autoloss to 6/7/8 gate. You never watched his gas, which means you autolose to 6/7/8 gate because you really have to prepare against it by taking very extraordinary measures (no lair, earlier roach warren, no droning after 8).
The game would have been very clear cut if these mistakes didn't occur. You would have gotten ling speed, and would have made enough units to hold his initial 8:30 hit, no matter how good he is.
Maybe if it was be more of an issue if it was a 3 minute back and forth. But it wasnt. He followed a build order. At 8:30 he was at your base. You instantly lose your third because you blindly made a lair and skippped ling speed.
Please. It's not about your opponent at all in this game. If it was a macro game, sure. But he basically did a 4 gate. Now, yea, we would all lose to naniwa's 4 gate. But none of us would lose to someone who was favored on ladder, doing a 4 gate, if we scouted it, by just seeing he doesnt take his gas - or at least, not immediately.
I also think its hilarious as all hell that you now state "Armed with new knowledge, you will beat any ex-GM with even 20 apm on an off day who tries to do a 7 gate allin next time." Really? Initially I must have equal or greater APM to my opponent but now APM doesnt matter!
You stated that normally, your APM is 130+, and that your basic macro mistakes were things that *never* happen to you,and you were just off. Trusting what you say here, I think with 130+ APM you will be more than capable of holding any 6/7/8 gate, even a GM level one, when you have an overlord over his natural seeing he didn't take gas, and pumping roaches and lings after a 6:30 roach warren (or even 7 minute) while you skipped lair in favor of ling speed. If anything, it'll be a close game where you lose after a very intensive battle.
You certaintly won't just flop the second you see his units the first time.
I appreciate the advice (and keep in mind that I am heeding the actual advice given), but what you are saying is just rude and in the case of APM and my opponents level is just wrong.
I don't know if you are heeding them. You had some very real problems in this game, and I feel you aren't taking them seriously at all. You are just writing them off by saying "oh, he was such a better player".
Look, you know the merits of 60+ APM, because you yourself generally average 100+. So I know that you understand when I say that 20 average APM throughout a 10 minute game in high masters is not enough, and that if you know an all-in is coming and you prepare to defend it, you should win and hold.
It doesn't matter if the person is GM level and is doing a double starport banshee build - you know the right response, now you know the 100% surefire way to always scout it, and you will beat it everytime now.
I do not say anything rude here, and I would appreciate if you leave out the ad homs. This is my personal opinion on the game that was presented, i really think the opponent's skill level was not the issue, and I really do believe that 20 APM was a major factor in why you made such glaring mistakes like not sending a single overlord ever, towards his base.
Please. Understand you were the one that lost, not that he won. There was no fight or struggle in this game, he literally rolled over you because you got supply blocked at 28, 36, and had 20 APM and not enough spare APM to scout at all, and have an overlord by his natural's gas.
If you want to disagree with my analysis, that's fine. If you refuse to acknowledge why you lost, or how to improve, that's fine. This is just my analysis, I'm sure others will say 20 APM against a high masters will be more than enough APM, and that blindly being so greedy and skipping ling speed and making a lair and not scouting his natural's gas at all is okay.
Once again, I havent disagreed with your analysis or anything of the sort. I also didnt disagree with you that higher than 20 apm is necessary. However, I do have higher than 20 apm and I show it when I need it.
Remembering to scout is not an APM block, it is a memory block. It is not REMEMBERING to scout. I REMEMBERED to scout late, it is not I didnt have enough APM to scout. If not scouting were an APM block (for me), then I would never have scouted at all instead of scouted late at a more APM intensive time in the game.
Also, you dont have a single vod demonstrating a hold against an 8 gate +1 timing. You dont even mention an 8 gate +1 timing that involves stalkers. In fact, I will go so far as even argue that you in fact can not drone until 8 minutes against such an all in (not this one that is actually refined down to pulling workers off gas). So I actually disagree with your assessment. If you do a traditional drone until 8 minutes against the build that he did, with a 7:15 roach warren you will have maybe 5 roaches popping at the 3rd and 10 popping at your main and natural (at most, actually unlikely) while he has 11 zealots and 7 stalkers already with +1 hammering away at your 3rd. You can not put a roach warren down that late against this build, as you will always at least lose your 3rd.
Hell, your VOD of losira holding a 7 gate +1 was entirely different and completely unrefined compared to what I went against (in addition to him hitting the natural instead of the 3rd)... trickster had 5 sentries, 4 zealots 1 stalker and no proxy pylon at 9 minutes (compared to a much more brute force of 13 zealots and 7 stalkers that I faced, with a proxy pylon finished). Thats most of a production round difference in timing for roaches coming out.
I am going to go ahead and say that even with perfect execution of the 3 hatch roach build you can not roach warren at 7:15 against this specific build. You MUST get a roach warren down at 6:40 by the latest and start producing roaches when it finishes. He is very all in so there is no reason that you cant roach earlier to hold it.
I reject everything that you say. I already admitted my mistakes, and youre being an ass about it. You may not realize it, but you are. While I may have played quite shitty that game, and it happens when you have little experience with a build on top of not playing the game in nearly a month... I still think you put yourself above where you actually are. Get your ego in check and get some manners.
On May 06 2012 00:29 Belial88 wrote: Well, you stated in your first post that you thought that 6/7/8 gate all-in counters fast third play, so what I took from that was that on a strategic level, there is no way to hold it. Quite the contrary, 6/7/8 gates are seen as somewhat outdated these days, and considered quite basic and easy to hold (i think when losira when fast third against anypro or hongun or whatever all-in cheese toss it was on dual sight small map, and held his 7 gate all-in, it validated for everyone that if you can hold this most extreme of all-ins in such a convincing and straightforward way, you can hold anything, and so it became standard, and I'm sure every zerg in the world was practicing how to hold it as every toss learned to execute it).
So I pointed out 2 problems in your play - not just the strategic, but your macro as well, as that is probably a big reason why you thought it was such a bo counter.
I don't know what other stuff you think is rude or presumptuous. I'm just pointing out the flaws in that game. I don't think I've impressed upon you how basic and major a problem it is that you didn't redirect your original overlord to go towards his base, and you make your 28 overlord ridiculously early and your 36 overlord majorly supply blocked (essentially, you got supply blocked every single time after your 18 overlord). I also pointed that I thought 20 APM was a big factor in this, you largely made drones and put your buildings down on time, and you weren't THAT far behind at 8:00, so you kept up in those areas while faulting in scouting and overlord production. Did you have an off game, or you normally play with 200+ apm, cool, whatever, then I'm sure you've held this all-in easily many times before.
I mean, in my guide, I state very clearly - send overlord to natural gas, when to make your first 5 overlords, how to hold a 6/7/8 gate all-in, and a vod example. It was clear you didn't read my guide at all, because this are very simple issues that you could easily identify and have fixed if you read the guide.
I'm not trying to talk shit about you, you are obviously a competent player to be in masters and to face such an opponent. I never said any ad hom toward you, I'm sorry you feel that way. I really can't stress enough though, how macro was such a huge factor in your loss, that you had 20 apm and that may explain a lot, and that you made some very basic mistakes.
I mean, maybe you could have said "obviously i was off in this game, so I get why I lost, but I was wondering what to do when I come across this build" or something. I don't mean to bash on you because you actually provided a replay, but in terms of "why did I lose this game?", i think it should have been very clear.
I really do think you lost because of your APM. I don't know what to tell you, 20 APM average for a 10 minute game is where you see bronze-gold sit at. Day9 has said that it takes about 50 apm to pull off basic building, and that this is very low. I really think you made such major mistakes like not sending your initial overlord toward his base when you figured out where he spawned, losing your scouting drone to a probe, not sending a single overlord towards his base, not putting a ling in front of his base, and ALWAYS having idle larva, is a result of your 20 APM in that game. I don't think APM is important, but at masters, I think you really need at least 60 APM to get by.
I mean you never made drones. You just had idle larva all the time, missed injects, overlords were always off. That's fine, everyone does that except high masters, but you literally always had idle larva.
Yea, he outplayed you. But I think the problem with your attitude here, is that you are so intent on focusing on him. But that's not the issue. If YOU had fixed some very minor issues, you could have won.
I mean, if the guy went 5 gate robo, or 5 gate blink, into third, or immortal/sentry all-in, or some type of build where skill can come through, that's fine.
But he literally did a very plain build order, very much similar to a 4 gate, and he won the game instantly. At 8:30 he was in your base, and you just lost right then and there. There was no contest at all, any diamond could have won just by executing the same build, even if his attack came 30 seconds later.
You would have beat this very, very high ranked player, if you just: 1. put an overlord over his natural's gas 2. Not get supply blocked at 28 and 36.
That's it! Very straightforward. But you blindly made a lair, when making a lair before speed is an autoloss to 6/7/8 gate. You never watched his gas, which means you autolose to 6/7/8 gate because you really have to prepare against it by taking very extraordinary measures (no lair, earlier roach warren, no droning after 8).
The game would have been very clear cut if these mistakes didn't occur. You would have gotten ling speed, and would have made enough units to hold his initial 8:30 hit, no matter how good he is.
Maybe if it was be more of an issue if it was a 3 minute back and forth. But it wasnt. He followed a build order. At 8:30 he was at your base. You instantly lose your third because you blindly made a lair and skippped ling speed.
Please. It's not about your opponent at all in this game. If it was a macro game, sure. But he basically did a 4 gate. Now, yea, we would all lose to naniwa's 4 gate. But none of us would lose to someone who was favored on ladder, doing a 4 gate, if we scouted it, by just seeing he doesnt take his gas - or at least, not immediately.
I also think its hilarious as all hell that you now state "Armed with new knowledge, you will beat any ex-GM with even 20 apm on an off day who tries to do a 7 gate allin next time." Really? Initially I must have equal or greater APM to my opponent but now APM doesnt matter!
You stated that normally, your APM is 130+, and that your basic macro mistakes were things that *never* happen to you,and you were just off. Trusting what you say here, I think with 130+ APM you will be more than capable of holding any 6/7/8 gate, even a GM level one, when you have an overlord over his natural seeing he didn't take gas, and pumping roaches and lings after a 6:30 roach warren (or even 7 minute) while you skipped lair in favor of ling speed. If anything, it'll be a close game where you lose after a very intensive battle.
You certaintly won't just flop the second you see his units the first time.
I appreciate the advice (and keep in mind that I am heeding the actual advice given), but what you are saying is just rude and in the case of APM and my opponents level is just wrong.
I don't know if you are heeding them. You had some very real problems in this game, and I feel you aren't taking them seriously at all. You are just writing them off by saying "oh, he was such a better player".
Look, you know the merits of 60+ APM, because you yourself generally average 100+. So I know that you understand when I say that 20 average APM throughout a 10 minute game in high masters is not enough, and that if you know an all-in is coming and you prepare to defend it, you should win and hold.
It doesn't matter if the person is GM level and is doing a double starport banshee build - you know the right response, now you know the 100% surefire way to always scout it, and you will beat it everytime now.
I do not say anything rude here, and I would appreciate if you leave out the ad homs. This is my personal opinion on the game that was presented, i really think the opponent's skill level was not the issue, and I really do believe that 20 APM was a major factor in why you made such glaring mistakes like not sending a single overlord ever, towards his base.
Please. Understand you were the one that lost, not that he won. There was no fight or struggle in this game, he literally rolled over you because you got supply blocked at 28, 36, and had 20 APM and not enough spare APM to scout at all, and have an overlord by his natural's gas.
If you want to disagree with my analysis, that's fine. If you refuse to acknowledge why you lost, or how to improve, that's fine. This is just my analysis, I'm sure others will say 20 APM against a high masters will be more than enough APM, and that blindly being so greedy and skipping ling speed and making a lair and not scouting his natural's gas at all is okay.
Once again, I havent disagreed with your analysis or anything of the sort. I also didnt disagree with you that higher than 20 apm is necessary. However, I do have higher than 20 apm and I show it when I need it.
Remembering to scout is not an APM block, it is a memory block. It is not REMEMBERING to scout. I REMEMBERED to scout late, it is not I didnt have enough APM to scout. If not scouting were an APM block (for me), then I would never have scouted at all instead of scouted late at a more APM intensive time in the game.
Also, you dont have a single vod demonstrating a hold against an 8 gate +1 timing. You dont even mention an 8 gate +1 timing that involves stalkers. In fact, I will go so far as even argue that you in fact can not drone until 8 minutes against such an all in (not this one that is actually refined down to pulling workers off gas). So I actually disagree with your assessment. If you do a traditional drone until 8 minutes against the build that he did, with a 7:15 roach warren you will have maybe 5 roaches popping at the 3rd and 10 popping at your main and natural (at most, actually unlikely) while he has 11 zealots and 7 stalkers already with +1 hammering away at your 3rd. You can not put a roach warren down that late against this build, as you will always at least lose your 3rd.
Hell, your VOD of losira holding a 7 gate +1 was entirely different and completely unrefined compared to what I went against (in addition to him hitting the natural instead of the 3rd)... trickster had 5 sentries, 4 zealots 1 stalker and no proxy pylon at 9 minutes (compared to a much more brute force of 13 zealots and 7 stalkers that I faced, with a proxy pylon finished). Thats most of a production round difference in timing for roaches coming out.
I am going to go ahead and say that even with perfect execution of the 3 hatch roach build you can not roach warren at 7:15 against this specific build. You MUST get a roach warren down at 6:40 by the latest and start producing roaches when it finishes. He is very all in so there is no reason that you cant roach earlier to hold it.
I reject everything that you say. I already admitted my mistakes, and youre being an ass about it. You may not realize it, but you are. While I may have played quite shitty that game, and it happens when you have little experience with a build on top of not playing the game in nearly a month... I still think you put yourself above where you actually are. Get your ego in check and get some manners.
Your mistakes are what cost you the game not his super gosu strategy. Because of the gaping flaws in your basic play no amount of drone cutting or timings was going to save you at that point. You dont ask for help on a game you know you played poorly, ask for help on a game where you thought you did everything right but still lost.
On May 06 2012 00:29 Belial88 wrote: Well, you stated in your first post that you thought that 6/7/8 gate all-in counters fast third play, so what I took from that was that on a strategic level, there is no way to hold it. Quite the contrary, 6/7/8 gates are seen as somewhat outdated these days, and considered quite basic and easy to hold (i think when losira when fast third against anypro or hongun or whatever all-in cheese toss it was on dual sight small map, and held his 7 gate all-in, it validated for everyone that if you can hold this most extreme of all-ins in such a convincing and straightforward way, you can hold anything, and so it became standard, and I'm sure every zerg in the world was practicing how to hold it as every toss learned to execute it).
So I pointed out 2 problems in your play - not just the strategic, but your macro as well, as that is probably a big reason why you thought it was such a bo counter.
I don't know what other stuff you think is rude or presumptuous. I'm just pointing out the flaws in that game. I don't think I've impressed upon you how basic and major a problem it is that you didn't redirect your original overlord to go towards his base, and you make your 28 overlord ridiculously early and your 36 overlord majorly supply blocked (essentially, you got supply blocked every single time after your 18 overlord). I also pointed that I thought 20 APM was a big factor in this, you largely made drones and put your buildings down on time, and you weren't THAT far behind at 8:00, so you kept up in those areas while faulting in scouting and overlord production. Did you have an off game, or you normally play with 200+ apm, cool, whatever, then I'm sure you've held this all-in easily many times before.
I mean, in my guide, I state very clearly - send overlord to natural gas, when to make your first 5 overlords, how to hold a 6/7/8 gate all-in, and a vod example. It was clear you didn't read my guide at all, because this are very simple issues that you could easily identify and have fixed if you read the guide.
I'm not trying to talk shit about you, you are obviously a competent player to be in masters and to face such an opponent. I never said any ad hom toward you, I'm sorry you feel that way. I really can't stress enough though, how macro was such a huge factor in your loss, that you had 20 apm and that may explain a lot, and that you made some very basic mistakes.
I mean, maybe you could have said "obviously i was off in this game, so I get why I lost, but I was wondering what to do when I come across this build" or something. I don't mean to bash on you because you actually provided a replay, but in terms of "why did I lose this game?", i think it should have been very clear.
I really do think you lost because of your APM. I don't know what to tell you, 20 APM average for a 10 minute game is where you see bronze-gold sit at. Day9 has said that it takes about 50 apm to pull off basic building, and that this is very low. I really think you made such major mistakes like not sending your initial overlord toward his base when you figured out where he spawned, losing your scouting drone to a probe, not sending a single overlord towards his base, not putting a ling in front of his base, and ALWAYS having idle larva, is a result of your 20 APM in that game. I don't think APM is important, but at masters, I think you really need at least 60 APM to get by.
I mean you never made drones. You just had idle larva all the time, missed injects, overlords were always off. That's fine, everyone does that except high masters, but you literally always had idle larva.
Yea, he outplayed you. But I think the problem with your attitude here, is that you are so intent on focusing on him. But that's not the issue. If YOU had fixed some very minor issues, you could have won.
I mean, if the guy went 5 gate robo, or 5 gate blink, into third, or immortal/sentry all-in, or some type of build where skill can come through, that's fine.
But he literally did a very plain build order, very much similar to a 4 gate, and he won the game instantly. At 8:30 he was in your base, and you just lost right then and there. There was no contest at all, any diamond could have won just by executing the same build, even if his attack came 30 seconds later.
You would have beat this very, very high ranked player, if you just: 1. put an overlord over his natural's gas 2. Not get supply blocked at 28 and 36.
That's it! Very straightforward. But you blindly made a lair, when making a lair before speed is an autoloss to 6/7/8 gate. You never watched his gas, which means you autolose to 6/7/8 gate because you really have to prepare against it by taking very extraordinary measures (no lair, earlier roach warren, no droning after 8).
The game would have been very clear cut if these mistakes didn't occur. You would have gotten ling speed, and would have made enough units to hold his initial 8:30 hit, no matter how good he is.
Maybe if it was be more of an issue if it was a 3 minute back and forth. But it wasnt. He followed a build order. At 8:30 he was at your base. You instantly lose your third because you blindly made a lair and skippped ling speed.
Please. It's not about your opponent at all in this game. If it was a macro game, sure. But he basically did a 4 gate. Now, yea, we would all lose to naniwa's 4 gate. But none of us would lose to someone who was favored on ladder, doing a 4 gate, if we scouted it, by just seeing he doesnt take his gas - or at least, not immediately.
I also think its hilarious as all hell that you now state "Armed with new knowledge, you will beat any ex-GM with even 20 apm on an off day who tries to do a 7 gate allin next time." Really? Initially I must have equal or greater APM to my opponent but now APM doesnt matter!
You stated that normally, your APM is 130+, and that your basic macro mistakes were things that *never* happen to you,and you were just off. Trusting what you say here, I think with 130+ APM you will be more than capable of holding any 6/7/8 gate, even a GM level one, when you have an overlord over his natural seeing he didn't take gas, and pumping roaches and lings after a 6:30 roach warren (or even 7 minute) while you skipped lair in favor of ling speed. If anything, it'll be a close game where you lose after a very intensive battle.
You certaintly won't just flop the second you see his units the first time.
I appreciate the advice (and keep in mind that I am heeding the actual advice given), but what you are saying is just rude and in the case of APM and my opponents level is just wrong.
I don't know if you are heeding them. You had some very real problems in this game, and I feel you aren't taking them seriously at all. You are just writing them off by saying "oh, he was such a better player".
Look, you know the merits of 60+ APM, because you yourself generally average 100+. So I know that you understand when I say that 20 average APM throughout a 10 minute game in high masters is not enough, and that if you know an all-in is coming and you prepare to defend it, you should win and hold.
It doesn't matter if the person is GM level and is doing a double starport banshee build - you know the right response, now you know the 100% surefire way to always scout it, and you will beat it everytime now.
I do not say anything rude here, and I would appreciate if you leave out the ad homs. This is my personal opinion on the game that was presented, i really think the opponent's skill level was not the issue, and I really do believe that 20 APM was a major factor in why you made such glaring mistakes like not sending a single overlord ever, towards his base.
Please. Understand you were the one that lost, not that he won. There was no fight or struggle in this game, he literally rolled over you because you got supply blocked at 28, 36, and had 20 APM and not enough spare APM to scout at all, and have an overlord by his natural's gas.
If you want to disagree with my analysis, that's fine. If you refuse to acknowledge why you lost, or how to improve, that's fine. This is just my analysis, I'm sure others will say 20 APM against a high masters will be more than enough APM, and that blindly being so greedy and skipping ling speed and making a lair and not scouting his natural's gas at all is okay.
Once again, I havent disagreed with your analysis or anything of the sort. I also didnt disagree with you that higher than 20 apm is necessary. However, I do have higher than 20 apm and I show it when I need it.
Remembering to scout is not an APM block, it is a memory block. It is not REMEMBERING to scout. I REMEMBERED to scout late, it is not I didnt have enough APM to scout. If not scouting were an APM block (for me), then I would never have scouted at all instead of scouted late at a more APM intensive time in the game.
Also, you dont have a single vod demonstrating a hold against an 8 gate +1 timing. You dont even mention an 8 gate +1 timing that involves stalkers. In fact, I will go so far as even argue that you in fact can not drone until 8 minutes against such an all in (not this one that is actually refined down to pulling workers off gas). So I actually disagree with your assessment. If you do a traditional drone until 8 minutes against the build that he did, with a 7:15 roach warren you will have maybe 5 roaches popping at the 3rd and 10 popping at your main and natural (at most, actually unlikely) while he has 11 zealots and 7 stalkers already with +1 hammering away at your 3rd. You can not put a roach warren down that late against this build, as you will always at least lose your 3rd.
Hell, your VOD of losira holding a 7 gate +1 was entirely different and completely unrefined compared to what I went against (in addition to him hitting the natural instead of the 3rd)... trickster had 5 sentries, 4 zealots 1 stalker and no proxy pylon at 9 minutes (compared to a much more brute force of 13 zealots and 7 stalkers that I faced, with a proxy pylon finished). Thats most of a production round difference in timing for roaches coming out.
I am going to go ahead and say that even with perfect execution of the 3 hatch roach build you can not roach warren at 7:15 against this specific build. You MUST get a roach warren down at 6:40 by the latest and start producing roaches when it finishes. He is very all in so there is no reason that you cant roach earlier to hold it.
I reject everything that you say. I already admitted my mistakes, and youre being an ass about it. You may not realize it, but you are. While I may have played quite shitty that game, and it happens when you have little experience with a build on top of not playing the game in nearly a month... I still think you put yourself above where you actually are. Get your ego in check and get some manners.
Your mistakes are what cost you the game not his super gosu strategy. Because of the gaping flaws in your basic play no amount of drone cutting or timings was going to save you at that point. You dont ask for help on a game you know you played poorly, ask for help on a game where you thought you did everything right but still lost.
I asked for help because I felt that the standard timings would not work against that build. I realize my mistakes definitely did not help, and I have stated this about a million times. I dont play enough to be able to ask for help on a game that I play well on (and because most of the time when I play well, I dont lose and those that I do lose... I know why I lost). Given the fact that I have now looked a bit into the replay some more and have come to the realization that you are basically guaranteed to at least lose the 3rd against this build if you dont get a roach warren down sooner I have locked in what I need to do the next time I encounter it.
When you get good enough that you can realize and correct your own mistakes, but need to discuss what to possibly do differently in the future to hold something off then you can unabashedly post a game where you played horribly and discuss the things that need to be done against it in the future, regardless of your own play (because you can already correct your own play).
On May 06 2012 09:43 Effay wrote: How exactly do you know to get a 6:30 roach warren to stop the 4 gate +1 attack if you don't overlord scout until 7:00 at the earliest?
Warp gate tech finishes at ~8 minutes for gateway builds off FFE. Dropping it at 6:30 provides enough time for the warren to finish as well as get 4 or so roaches, which should be enough to completely shut down 4 gate aggression.
On May 06 2012 09:43 Effay wrote: How exactly do you know to get a 6:30 roach warren to stop the 4 gate +1 attack if you don't overlord scout until 7:00 at the earliest?
He says in the guide, but no gas at the nat is a tell of some sort of early pressure. You cant distiguish whether its a 4gate +1 or 6/7/8 gate till the 7:30 ish mark when he'll either take 2 more gas or, have none, or your overlord will have scouted everything.
Belial what do you think about a 2 hatch opening that uses an early ling speed, lair, and hyrdas for defending toss pressure? Day9 did a daily on it I think back in Feb. Defends 4gate +1 with pure ling, and any 8-10 minute blink stalker/immo sentry push with ling hyrda.
On May 06 2012 12:38 Buckshot9044 wrote: Belial what do you think about a 2 hatch opening that uses an early ling speed, lair, and hyrdas for defending toss pressure? Day9 did a daily on it I think back in Feb. Defends 4gate +1 with pure ling, and any 8-10 minute blink stalker/immo sentry push with ling hyrda.
Wasn't this the daily about Zenio? I remember Day[9] talking about how Zenio built a preemptive hydralisk den while waiting for mutalisks in order to fend off any P aggression.
On May 06 2012 08:58 TheRabidDeer wrote: Remembering to scout is not an APM block, it is a memory block. It is not REMEMBERING to scout. I REMEMBERED to scout late, it is not I didnt have enough APM to scout. If not scouting were an APM block (for me), then I would never have scouted at all instead of scouted late at a more APM intensive time in the game.
Maybe you don't understand what I was saying. I was saying you had an average of 20ish APM that whole game. Maybe your APM can spike to 300+, whatever, but having an average of 20 APM in a 10 minute game in high masters, is really low. Diamonds generally average around the 60's.
You never did scout at all... that's the thing. That's why I think your APM was a factor in your loss here. You never, ever sent an overlord towards his base, until after the 7:00 mark, and from your main, which was way too late. You should have known that sending an overlord the opponent's base, is the first thing you should do with your first 2-3 overlords. You didn't scout late, you just made zero effort the entire game to ever scout.
I've had enough of this APM discussion. We've both said what we believe in, you obviously aren't going to agree that having 20 average APM throughout a 10 minute game is too low in a high masters game, and that this contributed in you losing your scouting worker to probes, not sending a single overlord toward his base, not checking the front of his base with a ling, never making any effort, ever, to scout (an overlord at 8:00 when stalkers in your base, sent from your main, is not helpful), and always having idle larva.
edit: i think in the day9 'probes and pylons', he showed that basic macro was easy and could be done just using the mouse and no keyboard, and that it only took 50 APM. How in the world do you have 20 average apm in masters, seriously. that blows my mind, platinum dont have 20 apm in 10 minutes of a game. But whatever, you already stated your actual average APM is always 100+, you were just somehow really off in this game...
Also, you dont have a single vod demonstrating a hold against an 8 gate +1 timing. You dont even mention an 8 gate +1 timing that involves stalkers. In fact, I will go so far as even argue that you in fact can not drone until 8 minutes against such an all in (not this one that is actually refined down to pulling workers off gas). So I actually disagree with your assessment. If you do a traditional drone until 8 minutes against the build that he did, with a 7:15 roach warren you will have maybe 5 roaches popping at the 3rd and 10 popping at your main and natural (at most, actually unlikely) while he has 11 zealots and 7 stalkers already with +1 hammering away at your 3rd. You can not put a roach warren down that late against this build, as you will always at least lose your 3rd.
Hell, your VOD of losira holding a 7 gate +1 was entirely different and completely unrefined compared to what I went against (in addition to him hitting the natural instead of the 3rd)... trickster had 5 sentries, 4 zealots 1 stalker and no proxy pylon at 9 minutes (compared to a much more brute force of 13 zealots and 7 stalkers that I faced, with a proxy pylon finished). Thats most of a production round difference in timing for roaches coming out.
uh...what?
6/7/8 gates are all the same. You respond the exact same way, they are scouted the exact same way, and they are all just as all-in. No one responds in a way expecting a 6 gate+1, and then dies because it was actually an 8 gate. All of these builds are the same, with minor pros and cons - an extra gate, less probes and slightly lower production but more forgiving to miss a warp-in, an upgrade, but less units or production, etc. They all hit you are 8:30, they all kill you if you went lair before speed and have no earlier roach warren and pump units.
It has nothing to do with stalkers... You are just outright wrong here. Any Toss can go 8 gate +1, and just make a zealot instead of a stalker. All 6/7/8 gate builds though, have very few sentries, if any at all, and are largely a 50/50 mix of zealot/stalker.
You can disagree all you want, but you are just plain wrong. The build would have been beat if you followed my guide on what I say about 6/7/8 gate all-ins - you see he didnt take gas at his nat, so you dont get lair and instead get speed, and you pump roaches when 6:30 warren, that you made earlier because of toss not having gas, pops, and you continue to pump units if he continues his aggression.
Trickster pushed out and was on Zerg territory by 8:30 in that vod. In your game, your opponent wasn't exactly in your base at 8:30 either. They both were shooting zerg stuf at 8:45ish. Your opponent did not have a proxy pylon set up either.
I can't believe you seriously think it's a different build... You can't afford more than 6 gate +1 or 7 gate - anything else or more, like 8 gate +1, means that he will have much less stalkers and sentries (sometimes they may have 1-4 sentries, but not if you go 8 gate +1), much less production (it's impossible for 40 probes to produce stalkers out of 8 gates...). You hold both builds exactly the same. It's not that hard to understand.
If you had done exactly what Losira did - saw no gas, made earlier warren, not gotten lair, got ling speed, pumped units at 8:00, you would have easily won. You failed to do that, and even if you did do that, your macro was horrible. It's 100% a priority to put an overlord by Toss' gas at the natural, you will see EVERY zerg player do this.
I am going to go ahead and say that even with perfect execution of the 3 hatch roach build you can not roach warren at 7:15 against this specific build. You MUST get a roach warren down at 6:40 by the latest and start producing roaches when it finishes. He is very all in so there is no reason that you cant roach earlier to hold it.
Right. I said that in the guide. Not because of the build though, but because based on what you scout, a 4 gate +1 pressure looks the exact same (JUST NO GAS AT TOSS NATURAL, BECAUSE YOU PUT AN OVERLORD THERE), so you make a 6:30 warren. Thanks for reading the guide.
I reject everything that you say. I already admitted my mistakes, and youre being an ass about it. You may not realize it, but you are. While I may have played quite shitty that game, and it happens when you have little experience with a build on top of not playing the game in nearly a month... I still think you put yourself above where you actually are. Get your ego in check and get some manners.
Okay. I'm going to tell you something.
I listen to the Howard Stern radio show. I love it. Great stuff. There's a staff member called Sal on it. Now, Sal is not an idiot, but he is when it comes to history. As in, he doesn't know if the revolutionary, civil, or ww1, came first. He thinks Buzz Aldrin made the movie transformers. He thinks that Roosevelt got us into Vietnam, and he killed JFK because JFK wanted us out of Vietnam. He's not a retard or anything, he just is totally out of it when it comes to history. Everything else - science, math, he's just average on.
Now, Sal insists that history is not a big deal, he doesn't give a shit about history, who cares, it doesn't affect him, what does this have to do with his job or what he's going to do tommorow.
But what he fails to realize, is that this isn't like a minor thing. This isn't like "oh he's just bad at history". He is completely unaware, but he looks like a jackass for his ridiculousness when it comes to history, and that he sooooo off. It's ironic, because he's the guy who does the 'on the street' style interviews with hookers and models, making fun of them by asking them if they know addition, who the president is, etc.
Well, you are kind of like Sal. You really don't think that 20 APM had an effect on your lack of scouting, or that how to hold a 6 gate +1, is the same as holding any sort of plain, 8:30 gateway all-in. You really don't think it's a big deal, or that you getting supply blocked every single time after 18, is what cost you the game. But it is. It totally is. You have such very basic, major flaws in your gameplay, or at least in this game, that it's just ruining you. You don't send a single overlord toward the natural gas of Toss. You make lair blindly. You never make drones.
It's a problem dude. A huge problem. Watch any game of any pro zerg. You will see they do these things - that they are 65 or higher in supply by 8:00. They don't get supply blocked, ever, at 28 or 36, much less at all in the first 8 minutes. They *always* send their first overlord to the natural gas of Toss. And they all hold the 6/7/8 gate all-in exactly as how I say to - they see no gas at Toss natural, and they make ling speed, they don't get lair at all, and they make a slightly earlier roach warren and pump units as soon as it finishes. You can watch any mid-masters Zerg, they don't get supply blocked at 28, 36, and they send an overlord to Toss natural.
Anyways, it's pretty clear you refuse to learn anything from me. I don't know why you even bothered posting in here. You clearly didn't read a single part of my guide, you refuse to take my advice when everyone here has accepted that it is correct (not because I am saying anything new or novel, but because I'm saying very basic stuff that all higher level people already know, and lower level people generally know but may pick up some extra tips on). I'm pretty sure that anyone else reading out conversation, would side with me over you. If you disagree, feel free to make a thread of your replay, and see the responses you get.
But you are just like the low league people in that "Macro better = Diamond+" thread by Cecil. They insist that the problems they make are that big, and they refuse to acknowledge that macro is a huge part of the game. By having such a negative attitude, they will stagnate and never improve.
That's the end of our talk. You just continue to attack me, when I have made perfectly clear my stance. You played like shit, you admit you played like shit, I don't know why you find this so offensive if you have 100+ APM and you know you played bad that game. You obviously refuse to learn anything - well, this is my guide, my thread. I'm telling you like I see it. If you don't like how I see it, then quit asking me!
I asked for help because I felt that the standard timings would not work against that build. I realize my mistakes definitely did not help, and I have stated this about a million times. I dont play enough to be able to ask for help on a game that I play well on (and because most of the time when I play well, I dont lose and those that I do lose... I know why I lost). Given the fact that I have now looked a bit into the replay some more and have come to the realization that you are basically guaranteed to at least lose the 3rd against this build if you dont get a roach warren down sooner I have locked in what I need to do the next time I encounter it.
No, you don't. You will have no clue how to face it next time because you don't send an overlord to his natural. You see no gas at his natural, make a 6:30 roach warren, pump units when it pops. That's it. If he's doing a 6/7/8 gate instead of 4 gate +1, you can get away with a 7:00 'standard' roach warren, as long as you are pumping units after 8:00 (lings) - but since 4 gate +1 and 6/7/8 gate look exactly the same with your scouting overlords, you would make a 6:30 roach warren anyways. Just the difference is if you screwed up and forgot it for 30 seconds, you won't lose like you would against 4 gate +1.
How exactly do you know to get a 6:30 roach warren to stop the 4 gate +1 attack if you don't overlord scout until 7:00 at the earliest?
You sacrifice an overlord into his main at 7:00+. The first overlord that you start out with, should be sent directly to his natural's geysers, so you can spot them and how he takes them (2 gas taken right when nexus finishes is a dead ringer to double stargate, 2 gas taken at 7:00 could be many things, most usually gateway all-ins though). You should have an overlord over his natural gas by around 3:00 or earlier depending on map size.
Belial what do you think about a 2 hatch opening that uses an early ling speed, lair, and hyrdas for defending toss pressure? Day9 did a daily on it I think back in Feb. Defends 4gate +1 with pure ling, and any 8-10 minute blink stalker/immo sentry push with ling hyrda.
I think it's a horrible all-in that's extremely outdated, but it has a lot of potential if Toss opens FFE into single stargate. Toss should not go 4 gate +1 if he sees you didn't take fast third, and obviously took gas, and obviously staying on 2 base - he should just macro up, defend, come out way ahead.
Any type of 2 base build, you HAVE to do a huge damage against FFE, or else you are really far behind. In some cases, like 2 base muta on TDA, it's pretty easy to do economic damage and not be too far behind if you play it right and Toss didn't counter your build (either blindly, or because it's obvious when you didn't go fast third, and didn't go roach/ling).
I discuss 2 base hydra in my Zerg All-ins section.
Holding 2 base Toss aggression is not the plan with it though. It should be you doing the aggression and Toss defending. You should start lair ~35 and have hydra den planted ~50-55. It's really all-in, 2 base muta or infestor isn't nearly as all-in (neither will likely kill Toss off straight up though, but won't leave you so far behind if you don't, and is more likely to do at least some damage too).
If you go 2 base lair against FFE, and you are defending, there's no way to win. Toss can simply just take a third as he has total map control and keeps you in your base, and really there's just no way to win from that point. If you go 2 base lair against FFE, you have to be the aggressor, and if you fail to outright kill Toss, you need to have map control and keep him in his base, do at least some damage, and take your third while you still have an army advantage.
I am not going to respond to your entire post, because either I have not spoken a word of english, or you dont listen to me, or you are just trolling me. However, I am going to pick this little bit out because it is so horribly explained
You sacrifice an overlord into his main at 7:00+. The first overlord that you start out with, should be sent directly to his natural's geysers, so you can spot them and how he takes them (2 gas taken right when nexus finishes is a dead ringer to double stargate, 2 gas taken at 7:00 could be many things, most usually gateway all-ins though). You should have an overlord over his natural gas by around 3:00 or earlier depending on map size.
He is asking how you are supposed to know to throw down a roach warren at 6:30 if you dont scout until 7:00. If you follow the guide to the letter, then you will never know to throw down a roach warren at 6:30 because you havent scouted yet... that doesnt happen for another 30 seconds. So to him, it seems like a blind roach warren at 6:30 because scouting has yet to occur.
Your first overlord is watching the gas at his natural all game long. As soon as it finds the opponents base it goes and camps behind the natural geysers. The 7:00 scout is the other overlord checking out his main (or maybe both overlords).
My main concern is that one of the most common builds for the 2-3 immortal 7-gate stays on 2 gas for quite a while, then double gases the natural between 7:00 and 7:15, and this is very different from an aggressive 7-gate. you really don't want an early roach warren against that, because you need every single drone to stop it.
On May 07 2012 00:33 TheRabidDeer wrote: I am not going to respond to your entire post, because either I have not spoken a word of english, or you dont listen to me, or you are just trolling me. However, I am going to pick this little bit out because it is so horribly explained
You sacrifice an overlord into his main at 7:00+. The first overlord that you start out with, should be sent directly to his natural's geysers, so you can spot them and how he takes them (2 gas taken right when nexus finishes is a dead ringer to double stargate, 2 gas taken at 7:00 could be many things, most usually gateway all-ins though). You should have an overlord over his natural gas by around 3:00 or earlier depending on map size.
He is asking how you are supposed to know to throw down a roach warren at 6:30 if you dont scout until 7:00. If you follow the guide to the letter, then you will never know to throw down a roach warren at 6:30 because you havent scouted yet... that doesnt happen for another 30 seconds. So to him, it seems like a blind roach warren at 6:30 because scouting has yet to occur.
Your overlord should, as Belial has said, already be in place to scout gas by 6:30, if you see no gas, you get an early roach warren.
That's it, if he's doing early pressure, you see it and you hold. If he isn't, you can drone up some more. That's all there is to it.
Thanks for the replies Belial and ThomasHobbes. I must have misread the scouting part of the guide, since i was under the impression that you only scouted with overlords (after confirming FFE of course) at the 7:00 mark, by going into main AND checking the nat's gas
I just want to stop by and say thanks to the OP. I've been lost in this matchup for a while and just having the timings more-or-less laid out against FFE has been amazing for me. Again, thanks for the help!
He is asking how you are supposed to know to throw down a roach warren at 6:30 if you dont scout until 7:00. If you follow the guide to the letter, then you will never know to throw down a roach warren at 6:30 because you havent scouted yet... that doesnt happen for another 30 seconds. So to him, it seems like a blind roach warren at 6:30 because scouting has yet to occur.
It says quite clearly in the guide to send your initial overlord to his natural's gas, something that every pro does (maybe not 1st, but one of the first 3 at least, depending on map size). This is something you didn't do, so I can understand why you are so completely confused here.
1 overlord to his natural's gas asap. A different overlord by his main, to be sac'd at 7:00-7:30. You will watch his gas from 3:00+ with the overlord at his natural. You will sac a different overlord into his main at 7:00+.
He is asking how you are supposed to know to throw down a roach warren at 6:30 if you dont scout until 7:00. If you follow the guide to the letter, then you will never know to throw down a roach warren at 6:30 because you havent scouted yet... that doesnt happen for another 30 seconds. So to him, it seems like a blind roach warren at 6:30 because scouting has yet to occur.
It says quite clearly in the guide to send your initial overlord to his natural's gas, something that every pro does (maybe not 1st, but one of the first 3 at least, depending on map size). This is something you didn't do, so I can understand why you are so completely confused here.
1 overlord to his natural's gas asap. A different overlord by his main, to be sac'd at 7:00-7:30. You will watch his gas from 3:00+ with the overlord at his natural. You will sac a different overlord into his main at 7:00+.
It says to keep an overlord at his natural's gas, but not what to look for. It doesnt say what the signal is to put a roach warren down at 6:30. It just says to have an overlord at his naturals gas. That is my point. And you can honestly drop my game completely because I am not arguing with you anymore because you are biased and ignoring just about everything I am saying.
Just to end your idiotic discussion about that game, here is one of me from a month ago http://drop.sc/175142
Not the best game still, but a lot less rusty... and imo not entirely bad for a guy that has less than 500 games played in 2 years.
On May 09 2012 05:49 TheRabidDeer wrote: It says to keep an overlord at his natural's gas, but not what to look for. It doesnt say what the signal is to put a roach warren down at 6:30. It just says to have an overlord at his naturals gas. That is my point. And you can honestly drop my game completely because I am not arguing with you anymore because you are biased and ignoring just about everything I am saying.
You mustn't have read the guide properly because the section on scouting the opponent's gas at his natural is quite detailed. It deals with the various possibilities arising from no gas, 1 gas, or 2 gas and how to react appropriately (the trigger for early roach warren is no gas at the natural).
On May 09 2012 05:49 TheRabidDeer wrote: It says to keep an overlord at his natural's gas, but not what to look for. It doesnt say what the signal is to put a roach warren down at 6:30. It just says to have an overlord at his naturals gas. That is my point. And you can honestly drop my game completely because I am not arguing with you anymore because you are biased and ignoring just about everything I am saying.
You mustn't have read the guide properly because the section on scouting the opponent's gas at his natural is quite detailed. It deals with the various possibilities arising from no gas, 1 gas, or 2 gas and how to react appropriately (the trigger for early roach warren is no gas at the natural).
I did read that, but I am just saying that it doesnt specifically state that you must have a roach warren at 6:30 if no gas... at least I dont remember it stating that when I last read it a while back.
On May 09 2012 05:49 TheRabidDeer wrote: It says to keep an overlord at his natural's gas, but not what to look for. It doesnt say what the signal is to put a roach warren down at 6:30. It just says to have an overlord at his naturals gas. That is my point. And you can honestly drop my game completely because I am not arguing with you anymore because you are biased and ignoring just about everything I am saying.
You mustn't have read the guide properly because the section on scouting the opponent's gas at his natural is quite detailed. It deals with the various possibilities arising from no gas, 1 gas, or 2 gas and how to react appropriately (the trigger for early roach warren is no gas at the natural).
I did read that, but I am just saying that it doesnt specifically state that you must have a roach warren at 6:30 if no gas... at least I dont remember it stating that when I last read it a while back.
I'm not going to take the time to respond to you, you are obviously a troll. It goes quite in-depth that if you see no gas, you should make a 6:30 warren in the "What zero gas means" in the scouting section, "how to hold 6/7/8 gate/4 gate+1" section, and "Roach Warren/Evo Chamber" section.
Yea, maybe you should read the actual guide. Thanks.
Nice post, I still think 15p/17h is better than 14/15 or 14/16, but that's personal preference.
That's perfectly fine. I'd like to see someone actually test it (14 vs 15 pool, toss puts down a pylon, zerg puts down hatch when pylon is killed by 4 lings). I just said I 'felt' 15p was worse due to later hatch and later lings, but wasn't really sure what was better.
Well, I think the reason for it is that your hatch is delayed longer, so the 14p ends up even economically (imo). And you should be able to afford the queen - you will have 1 larva, so you make 2 lings + queen, then 2 more lings. Or if you stop droning at 15 instead of 16 (like what I do at least), you will just have the money because you didn't make that drone.
It's all preference, but we should get some testing on it to see which is really better when a pylon block occurs. I'll try to test the builds later to see which is better against a pylon block.
On May 10 2012 00:47 Belial88 wrote: Well, I think the reason for it is that your hatch is delayed longer, so the 14p ends up even economically (imo). And you should be able to afford the queen - you will have 1 larva, so you make 2 lings + queen, then 2 more lings. Or if you stop droning at 15 instead of 16 (like what I do at least), you will just have the money because you didn't make that drone.
It's all preference, but we should get some testing on it to see which is really better when a pylon block occurs. I'll try to test the builds later to see which is better against a pylon block.
I edited my post. Do you always get lings to clear natural instead of taking third before natural?
Because if you're not placing a hatchery before pool pops you'll ofc be able to afford the lings and queen but if you're doing
Yea, it really just depends if toss can get that pylon up or not. But you can't know before you decide to throw down the pool if you can get that hatch up. It's just a gamble to take.
Hey, i have two questions. 1. What should I do on Tal'Darim Altair? Shall I take a 3rd like always? Although it's quite hard to defend it? 2. How to scout gas, if you send your Overlord to the wrong base and then you have to go cross the map to get to the gases. for example on Antiga Shipyard? I don't think the Overlord will be their in time so you know when to take the Roach Warren.
I really like your guide, but I hit 65 supply at 8 minutes and I'm not high masters, the Probes and Pylons theory is not 100% accurate if you play Zerg.
hey belial this is looking very cool but i have a question: if i get pylon blocked on lets say ohana... i cant get my drones to destroy the pylons... if i wish to nydus bust as fallow up from ONE base... what is the best way to do it because i tried it and didnt had hydras and he just killed my army with 2 voids. otherwise i feel if i go hydra off one base it might be to late to kill him
On March 12 2012 17:36 Belial88 wrote: Hey Belial88, I'm just starting out, any tips? Focus on your macro. When you are below Masters, you don't really realize how horrible your macro is, so you may want a higher level person to analyze it for you, and you can never underestimate how important it is, because if you just macro 'okay', not good, but 'okay (avoid ever being supply blocked, don't bank 400+ money ever, never have a lapse of more than a second in idle larva or worker/unit production), and understand basic scouting (did he expand or not, does he have a lot of basic army/macro or could he be teching?), you would just simply skyrocket to Diamond (with a slight improvement, maybe a minor understanding of the game as anyone would have by diamond, then you would be masters soon enough). All of the goofy, late strats and attacks you see below masters, all become unviable as good macro simply hard counters them (3 void rays at 11:00? u leik 170/200 roaches?).
When you are above Masters, you realize how shitty your macro is. Especially if you play against a high masters or GM - you really don't appreciate how good a pro player is until you play one once you are at a level where you can execute basic macro. The efficiency they make a single unit have, is ridiculous (below masters, it's unanimous that people just hurt themselves every time they macro or attack because of lack of multi-task, let alone macro).
US Server?
Also shouldn't that be micro a few times instead of macro?
Your OP is over 23,000 words. I just wanted to say thanks for posting something that in-depth and I can't wait to read it all when I have time. I'm trying to get back into SC2 after several months off and this is exactly the kind of guide I was hoping to find!
Also shouldn't that be micro a few times instead of macro?
The scary thing is no, your macro. Because then you'll realize how much they'll rape you with good micro, on top of good macro.
It's ridiculous what units are capable of doing in this game. It's not something you really see exploited until the higher levels, when macro becomes better.
Hey, i have two questions. 1. What should I do on Tal'Darim Altair? Shall I take a 3rd like always? Although it's quite hard to defend it? 2. How to scout gas, if you send your Overlord to the wrong base and then you have to go cross the map to get to the gases. for example on Antiga Shipyard? I don't think the Overlord will be their in time so you know when to take the Roach Warren.
btw.: Thanks for this great guide.
Veto the map. Taking a fast third is standard play in ZvP, so if you can't even play a macro game, then the map is crap.
Pick something from the Zerg all-in section. However, mutas are so strong on TDA that 2 base muta is almost just as strong as a macro build, even against Toss who bo counter it with quick blink or mass gateway.
But you can't really defend your third against a strong immortal/sentry or blink all-in (some gateway all-in slightly heavier than just plain 6/7/8 gate) because you either hurt your econ tearing down rocks or your third is so far away and rallying reinforcements is impossible due to distance as they cut off your third from reinforcements with FF.
Glad you like it.
I really like your guide, but I hit 65 supply at 8 minutes and I'm not high masters, the Probes and Pylons theory is not 100% accurate if you play Zerg.
The 8:00 benchmark is one of many.
65 supply is not impressive at all either. 65 is just 'okay'.
So yea, it is accurate. I don't know where in the guide you got the impression that 65 was good performance in a real game, because it isn't. It's 'acceptable' in a real game, but it's not 'good'.
Did you hit 65 in a real game? Or against AI?
But thanks for trying.
hey belial this is looking very cool but i have a question: if i get pylon blocked on lets say ohana... i cant get my drones to destroy the pylons... if i wish to nydus bust as fallow up from ONE base... what is the best way to do it because i tried it and didnt had hydras and he just killed my army with 2 voids. otherwise i feel if i go hydra off one base it might be to late to kill him
I state pretty clearly in the guide what to do in regards to pylon blocks - don't let it happen. I already stated that you can't get your drones to destroy the pylons.
Just patrol a worker, don't let a ramp block occur.
If you are stuck on 1 base, you aren't going to win against toss unless he makes a mistake. The only reason you aren't gg'ing out is because you are hoping the toss is dumb enough to let you get a nydus up and they don't respond correctly.
Your OP is over 23,000 words. I just wanted to say thanks for posting something that in-depth and I can't wait to read it all when I have time. I'm trying to get back into SC2 after several months off and this is exactly the kind of guide I was hoping to find!
Was wondering what the reaction to a Protoss taking a really early third is? I looked at the How to React section and it only covers various all-ins/pressures deeply.
I tried doing the Stephano Roach max against it, and they've been able to hold it off and attack afterwards with a higher tech army. I'm thinking I should take 4th, maybe 5th(just for geysers) drone up, spine up and rush out infestor broodlord. Should this be my plan?
Um, well that's a little deeper than this thread but I suppose you can 'infer' what it means. So we know the gameflow is.
.... Defend third, Take map control and deny/delay toss third Delay toss' 3 base deathball stalker/colossus+some push to get BL/Infestor in time
So the consequences of Toss taking a fast third means: - Third is succesfully defended (ie you don't need to defend it) - You can possibly straight up 'deny'/kill toss third as he's weaker, whereas normally Toss takes their third in a moment of strength, or at least growing strength to overcome your mass roach - Toss' 3 base deathball push will come quicker
So what this allows you to do sort of is: 1. Drone up more 2. Mass roaches or make mutas/infestors without making many roach/ling at all and be aggressive 3. Tech up hard and quick
Mutas can be really good against a toss who takes a third, as he's spread out and won't have as many gateways and blink to defend against the mutas (i really, really like going mutas against fast third toss, if I know mass roach won't work, like on entombed valley where a single FF can protect third). Going for infestors and teching super fast to hive is also a good choice.
I mean basically your third is 'defended', so you move right into the next stage of the game - you have map control. Do what you want. You didn't have to make many roaches or lings to protect yourself, so this allows you to get those mutas/infestors/hive much quicker, or even mass roach as you were allowed to drone a little more. Depending on the map, and how quickly you noticed what was going on, can make a big impact on what choice is best.
But yea in short, the whole 'defend your third' is really short and simple, and you have map control. You can either press hard with this map control against a toss you think is being too greedy (good on maps like, say, daybreak, where there's lots of room, and bad on entombed valley where the third is easily defendable and close to their natural), or tech up as you know they won't attack anytime soon (either to muta/infestors or all the way up to hive).
On May 10 2012 22:36 Destroyr wrote: hey belial this is looking very cool but i have a question: if i get pylon blocked on lets say ohana... i cant get my drones to destroy the pylons... if i wish to nydus bust as fallow up from ONE base... what is the best way to do it because i tried it and didnt had hydras and he just killed my army with 2 voids. otherwise i feel if i go hydra off one base it might be to late to kill him
I played two ZvPs on Ohana yesterday where they messed around with pylons, so I thought I might as well link them to you. This is diamond level and the opponents were pretty stupid, overcommitting like hell and suiciding units but at least it should show that protoss tech gets delayed quite a bit when they do these kind of stuff.
On May 12 2012 18:44 Belial88 wrote: Um, well that's a little deeper than this thread but I suppose you can 'infer' what it means. So we know the gameflow is.
.... Defend third, Take map control and deny/delay toss third Delay toss' 3 base deathball stalker/colossus+some push to get BL/Infestor in time
So the consequences of Toss taking a fast third means: - Third is succesfully defended (ie you don't need to defend it) - You can possibly straight up 'deny'/kill toss third as he's weaker, whereas normally Toss takes their third in a moment of strength, or at least growing strength to overcome your mass roach - Toss' 3 base deathball push will come quicker
So what this allows you to do sort of is: 1. Drone up more 2. Mass roaches or make mutas/infestors without making many roach/ling at all and be aggressive 3. Tech up hard and quick
Mutas can be really good against a toss who takes a third, as he's spread out and won't have as many gateways and blink to defend against the mutas (i really, really like going mutas against fast third toss, if I know mass roach won't work, like on entombed valley where a single FF can protect third). Going for infestors and teching super fast to hive is also a good choice.
I mean basically your third is 'defended', so you move right into the next stage of the game - you have map control. Do what you want. You didn't have to make many roaches or lings to protect yourself, so this allows you to get those mutas/infestors/hive much quicker, or even mass roach as you were allowed to drone a little more. Depending on the map, and how quickly you noticed what was going on, can make a big impact on what choice is best.
But yea in short, the whole 'defend your third' is really short and simple, and you have map control. You can either press hard with this map control against a toss you think is being too greedy (good on maps like, say, daybreak, where there's lots of room, and bad on entombed valley where the third is easily defendable and close to their natural), or tech up as you know they won't attack anytime soon (either to muta/infestors or all the way up to hive).
Ah, thanks for the advice, great help. I needed a plan vs this.
Toss who oppens gateway and then sees you went hatch first can just throw down a reactive forge and cannon rush you + zealot. This should be an autowin. there's a gsl game on crossfire where this hapened too. This is something higher level people have told me as an absolute, so if I'm wrong, please let me know, but as I was told and understood it from examples, toss scouts with probe to see hatch first, then reactively throw down a forge, chrono zealots out, and you can't keep the hatch.
I have a question about the overlord timings mentioned in your post. I apologize if this has been mentioned elsewhere, I only noticed the initial comments on the first page where you further stated to make overlords at 26/28, 50/54, etc, because doing so earlier hurts your drone count. This was something I was examining in my own replays to try and hit the 8:00 benchmark.
I watched replays of DRG, and replays that you posted, and I noticed that these overlord timings are usually not followed. Here are three screenshots from one game of DRG vs Parting, from (I believe) MLG Spring Arena 1.
I noticed similar timings in your own replays posted in your guide. Are these timings in replays just mistakes, or is there a different optimal time to make overlords before 50 supply?
Thank you very much, and thanks again for the guide!
They were pretty general overlord timings, and just what I seemed to do at the time. The point was more on where to send overlords, rather than when to make them, so people would know what overlord I was talking about (is 3rd overlord the 3rd one you make, or the 2nd one you make?)
When you take your third, cannon/pylon block/pressure, and how early you want to start you queens, can affect overlords timings.
In the case of the game you referenced, I'm not sure what happened, but DRG clearly took a third later than 24 (or something else happened, like shenanigans at the nat). He had to make his ~33 overlord at a different time because he didnt have 2 extra supply granted by the hatch that you would usually have at that time. His ~42 overlord is also affected - generally, you have 2 injects popping right when you are at 36 supply (which is exactly when your ~33 overlord pops and your third has been done), which means you have like, hundreds of minerals, and 0 larva to 10 larva instantly. It makes no difference if you make your overlord between 36 to 44, because you are morphing like 8 larva all at the same time.
Making an overlord at 42 is a bit too early, I've learned now. 44 sounds a bit early, I'm not sure. He has enough money to make 5 drones in that pic, so in the production of that overlord he has to make 5 more drones, which I think he probably could do.
Nice post though, thanks for the effort. cool stuff. I mean DRG could be making macro slips due to it being a real game, and making an overlord just a supply or two late by the 40+ mark is okay, it's when you make it really early or really late, or do it earlier than 40 supply, that it's a bigger deal. But after 40+, as long as you aren't blocked/early by , i dont know, 5+ seconds, you should be okay i think. Generally speaking though, I think DRG took a later third (not a bad thing, just later end than most people do in that game, or shenanigans occurred) in that game and yea, his 44 overlord is better than 42 overlord.
Question about macro benchmarking. I'm ~Platinum skill Zerg and I hit 70 supply @ 8 minute mark vs Toss. Almost all of the supply count is drones. Did you mess up the benchmark perhaps? Because I consider my macro to be bad and I do have idle larva from time to time.
EDIT: Oh, and by the way! Thanks for this greeeaaat guide! This is going to help me a ton, especially the scouting guide. Gonna go ahead and read your Terran guide too. Thanks alot!
On May 28 2012 01:54 vonKlaust II wrote: Question about macro benchmarking. I'm ~Platinum skill Zerg and I hit 70 supply @ 8 minute mark vs Toss. Almost all of the supply count is drones. Did you mess up the benchmark perhaps? Because I consider my macro to be bad and I do have idle larva from time to time.
EDIT: Oh, and by the way! Thanks for this greeeaaat guide! This is going to help me a ton, especially the scouting guide. Gonna go ahead and read your Terran guide too. Thanks alot!
If Toss doesn't attack you at all this is pretty standart I think. But I think the point Belial wanted to state is, that no matter what the Toss tries you should be able to achieve this benchmark by that time.
I've learned something recently about hitting the 70 supply mark. Apparently, the best way to maynard workers, is to rally them to new hatches as they pop. So instead of taking half of your main's workers when the natural pops, you just rally new drones to it, then when your third pops, just rally new drones to it. The timing works out well (rally to third when it pops and your main and nat are roughly balanced, and if not, just take the gases all in one base, or build roach warren/evo using drones from the heavier mineral line) that it's all pretty intuitive (main will be heavier, but make both gases in main). That's why I always struggled to hit 70 myself, I figured splitting everytime was the best way to maynard. It makes a difference of about 5-10 supply by the 8:00 mark, now I consistently hit 65+.
Question about macro benchmarking. I'm ~Platinum skill Zerg and I hit 70 supply @ 8 minute mark vs Toss. Almost all of the supply count is drones. Did you mess up the benchmark perhaps? Because I consider my macro to be bad and I do have idle larva from time to time.
EDIT: Oh, and by the way! Thanks for this greeeaaat guide! This is going to help me a ton, especially the scouting guide. Gonna go ahead and read your Terran guide too. Thanks alot!
Then you have amazing macro, post the rep. I probably have something to learn from you if you can hit 70+ every game, in a real game.
I struggle to hit 70, but I always hit 60's. Usually 65 is where I hit. Starting to hit 70's now that I learned better ways to maynard.
Also, it's not so much about toss attacking you, since units count for supply too. I'm not talking about drone counts, I'm talking about supply. Having 70+ supply of 'stuff' at 8:00 is a good thing.
wow, you have 2 comprehensive MU guides? that's amazing, it's well organized and has a lot of really useful specific information
On May 28 2012 19:57 Belial88 wrote: Then you have amazing macro, post the rep. I probably have something to learn from you if you can hit 70+ every game, in a real game.
Unfortunately, I don't save my replays so I don't have the replays I watched anymore. I just use the recent replays-feature. But I have been looking through quite a few of my recent game just to check my benchmarks, and in those couple of replays where I hit 70 I must have had some kind of divine inspiration. Most of my games I'm at around 50 supply @ 8 minutes. Some games I'm at 60, some games I'm at 40+. Maybe I was just watching some pro-game without realizing it...
^ Just post the rep of the next time you play, I'm sure myself and others can point out some problems that you could easily fix.
When I go fast third in ZvP, all I think about is macro macro macro. I swear, the first 8:30 always feels like the final leg of a marathon, I'm just spamming S all the time and trying to make the larva appear quicker. It's so tiring. I don't know what FFE is like, but god damn, trying to hit 70+ is a fucking accomplishment when going fast third.
Struggling to apply this guide.. I'm a platinum league zerg and whenever I scout the gasses (Let's say he has no gasses at natural and has warp gates finished in his main) I assume a timing attack, so I prepare with 6:30 roach warren and make the proper defense.. At 8 minutes when the attack should come...nothing comes... He sits on 2 base for 15 minutes and I have no idea what to do
^ Then go fucking kill him, you should easily overpower him when he has no sentries, and your roach speed kicks in.
All pressure/all-ins are timing attacks, because they only work during a certain, critical times. A 6 pool doesn't work if you have 70 drones and can reactively make 50 roaches. Similarly,a Protoss who is all-inning by lacking tech, because he doesn't take gas at his natural, will lose his timing to do damage hopefully (for him) before you have enough roach/ling, and the timing is completely over when roach speed kicks in (and things like burrow as well) and you can roll plain gateway armies.
Post the replay, there are probably macro problems in your play, but if you massed roaches and lings against a gasless Toss who simply doesn't move out, that means he has no sentries, and you can just go kill him with roach speed.
A 2 base Toss will never be able to take his third base against 200/200 roaches unless he has lots of sentries and immortals, and he goes for that third BEFORE you are maxed out (or, if he does wait that long, you will just have a powerhouse of an economy and have hive soon coming, or mass infestors or mutas). A 2 base Toss will also never be able to handle a zerg who has 15+ mutas, which you can get quite quickly off of only 6 gas.
The information is in the guide though - massing roaches is the early game, making mutas or infestors is the midgame, and making broodlords is the endgame. You can stay longer or shorter on each phase, depending on the dynamics of the game and what you want to do stylistically. If you've massed roaches against Toss, who still doesn't have his third and is just sitting on 2 base, the next natural step, if killing him doesn't work (and I strongly advise against busting up a ramp against a 2 base turtle, just forever deny his third and starve him to death), is get higher economy - but oh, you say, you already have 3 bases and a 4th coming just for gas, against a 2 base toss? Well, then your only next choice is tech!
But wait, I said infestors are bad, and hive is bad too! Well, infestors are just bad in that you can die while teching to them. If you clearly have map control with mass roach, going for infestors is fine. You've already 'won' the game, just don't throw it away by running up a choke or fast tracking to hive with no army out. If Toss moves out with a clearly intidimating sentry/immortal/blink 2 base deathball, just base trade, 200/200 roaches will beat a base trade against sentries and immortals and zealots. Mutas are a great choice though, personally when Toss is stuck on 2 base like this, I go for mutas, and force the base trade, mass spines, and then win the base trade.
Generally, you should never attack someone when YOU have the economic advantage - but if you have a key timing, take advantage of it! Toss has no sentries, that means roach pressure will just be beastly, and worst case scenario, you trade roaches for his entire army or colossi (remember, his income is limited to X bases, but yours is potentially infinite when you have map control - map control is everything in this game, and that's why zerg race is best race, speed=power). If Toss is on 2 base and doesn't have HT, mutas will wreck him! If Toss massed phoenixes, fast tech to hive and infestors because colossi will be late! Also, there's a big difference between attacking, and denying bases. You don't want to attack a turtling opponent when you have the macro lead, but you DO want to deny bases and make sure he can never step out of his little turtle shell.
Read the "Avoiding Ridiculous Losses section" for more clarification, specifically:
This is not some ambiguous thread. It's extremely useful. If you are on 3 bases vs 2, or 5 bases vs 3, your "E" is good. Now, work on on Tech and Army. If you lost a stupid game even though you were way ahead in econ, check the army values when you lost the game. Chances are, you were way behind in army value, and 1k less in roaches will never beat a higher value army of colossi, +3 upgrades, etc.
Basically, once you've secured your third, and held off that Toss aggression with such a huge lead, you no longer need to focus on your econ. You already have more bases. You don't need more drones, especially once you've reached 70. You need to match what Toss is doing. Match your tech, to his tech. Match your army, to his army. With a higher econ, but even army and tech, there is no way to lose. Too many times, people lose because of their 6 bases vs a 2 or 3 base Toss, because they only have roaches against sentry/colossi/blink/upgraded deathball. Unless you have lots of mutas and Toss doesn't have 3 bases with HT and lots of blink stalkers, you will never beat colossi without broodlord tech.
Once Toss makes the move toward more econ, then you can get that 4th, or 5-6-7 bases. Just because you aren't growing your econ, doesn't mean you aren't growing. Tech and army are also great ways to 'expand'..
On May 31 2012 13:07 Piwi22 wrote: Struggling to apply this guide.. I'm a platinum league zerg and whenever I scout the gasses (Let's say he has no gasses at natural and has warp gates finished in his main) I assume a timing attack, so I prepare with 6:30 roach warren and make the proper defense.. At 8 minutes when the attack should come...nothing comes... He sits on 2 base for 15 minutes and I have no idea what to do
When protoss sits on two base and makes colossus for 15 minutes, the best thing to do is to sit in front of his base and deny any third. He will be forced to do one big allin. Since you are on 3base eco vs 2base eco for like 7 minutes, you will have a much larger bank, while he cannot reinforce.
Everything in the macro stages still holds true, except when you realize hes not pushing you should stop making roachling and drone to 70-75, then resume roachling to max.
Theres two ways to play this, either make a huge spine wall and remax on roachling infestor over and over and counter until he runs out of steam, or make a huge spine wall and delay with your roachling infestor until you can replace your supply with broodlords. Both involve a huge spine wall, and both involve sitting on 3base with 70-80 drones.
It really depends on what you like and what they do.
On May 31 2012 14:41 Belial88 wrote: do you like my guide jombo?
Hey Belial88, anything else? Thanks NrGmonk to changing title. Thanks to UCSD SC2 for letting me know how shitty I am, I was quite unaware
Anything by Belial88 Highmasters w/ Star icon!
EDIT: In all seriousness, this is starting to become okay even for me. Try looking on some of the more newschool styles like the resurfacing 2base into 5base muta basetrade and 3hatch fast gas basetrade styles. The standard 3hatch is starting to get figured at the KR pro level, but still 100% viable at NA/EU pro levels.
In all seriousness, this is starting to become okay even for me.
The guide or me saying I'm high masters?
Try looking on some of the more newschool styles like the resurfacing 2base into 5base muta basetrade and 3hatch fast gas basetrade styles. The standard 3hatch is starting to get figured at the KR pro level, but still 100% viable at NA/EU pro levels.
It's new, nestea is the only person to do it in the GSL (only very recently though) so far. I am watching it's progress. I continue to update this guide, ive added reps recently for 4 gate immortal all-in and zenio's 3 hatch bane bust.
I covered the old builds already, but I am considering changing a few parts of the guide already, for 2 reasons: 1. We are seeing some zergs go infestors off 3 hatch against FFE, as their opening tech of choice. This goes against my 'advice', where I see I 'prefer' not to get infestors against 2 base Toss (except against double stargate). So far, every time someone did this, it has been against double stargate or lost, but it's obviously worked for suhosin before. I suppose I'm just saying it's a personal preference, but I am watching to see if anyone else does this. 2. 2 base infestor as a macro based build. As far as I've seen, this has always resulted in a loss, and there was a mass infested terran spam timing attack (that usually failed too). As far as I see, this build is being played the same as it was before (and with less success so far).
I am keeping an eye on old and new builds coming up, and will updated the guide as necessary.
Your guide is okay, not that you're high masters! I think the rough definition of high masters is around 1200+ with bonus pool right now, considering the cutoff for GM adjusted is around 1420 right now, so 1200 is when you would start matching everyone in the masters range and the occasional GM. I don't think you can be considered when you cant reach another masters level player even in the favored range.
Just IMHO.
I would say the 2base to 3base muta opener should be discussed more. Its still a "standard" strategy nowadays. Its used significantly more on the KR ladder than in pro games (anecdotal, just from watching KR streams of ToD and NightEnd).
Hey Belial i have a question. Its about this part in how to counter 2 stargate
Finally: Make sure to trade with his void rays before he gets 6+ critical mass that will kill anything you could make. That's a stupid way to lose. You don't need to be cost efficient, but 8+ void rays with phoenix support and ever increasing numbers will destroy anything but mass infestor, and you will not get mass infestors in time when he already has 4+ void rays and lots of phoenixes by the time your lair is done, unless you went straight for infestors. This is fairly easy to do with Corruptors, as phoenix just tickle them and you can snipe the voids very quickly with them once you have 8+.
How can i make sure do trade with his voids? I usually let him come at me, but if he doesnt? I just go with my queens to his base?
I have a question concerning creep spread. I dond't think it's mentioned in you guide. I could be mistaken, though... So, my question is "Is it a good idea to add a queen dedicated to creep spread? And if yes, when?" Or in a more general way "How do I spread creep"? Because when I have 3 hatches and 3 queens, all my queen energy is devoted to injects, and I sometimes find myself in a late-game position where I can't spine up effectively, due to the lack of creep...
How can i make sure do trade with his voids? I usually let him come at me, but if he doesnt? I just go with my queens to his base?
You used queens/spores to defend at first. You made a hydra/corruptor/infestor pit so you could move out on the map against him - be aggressive, deny his third. He will have no choice but to use some of his air units to secure his third and protect his base.
I have a question concerning creep spread. I dond't think it's mentioned in you guide. I could be mistaken, though... So, my question is "Is it a good idea to add a queen dedicated to creep spread? And if yes, when?" Or in a more general way "How do I spread creep"? Because when I have 3 hatches and 3 queens, all my queen energy is devoted to injects, and I sometimes find myself in a late-game position where I can't spine up effectively, due to the lack of creep...
Thanks!
Of course spreading creep is good. A dedicated creep queen when you can afford it, or getting overlord speed because of how often base trades occur in this match-up and having vision of army movement and using a few overlords to creep out right outside your base in lategame so you just bring a queen over to plant a few tumors down to solidify a huge creep area immediately, having good macro with your creep spread, etc...
Eventually lategame you'll have some extra energy on your queens. Just use that to spread a bit of creep, if not get a dedicated creep queen.
It's not in my guide particularly because there's nothing to really say about it - of course it's good to spread creep. It's up to you how to do that obviously, and I do mention with the infestor style, you will need creep spread to plant mass spines. What I do is just use overlord speed to lay out a bunch of creep out front, then drag a queen over to plant tumors all over it and wala, instant huge spread of creep in front of your base that you push out aggressively with.
Firstly, thanks so much for your guide. I just have a little question, about dealing with Immortal/Sentry. You've mentioned that use lots of ling/roach to shut it down, but Immortal just deal tons of damges to roaches, and force filde is just annoying. (You've mentioned if macroing well, it should able to shut it down, but is there an easier way to deal with?) I've seen Day9 talking about roach free hydra openings, and I find out that Ling/hydra is good at against nearly everything before colossi and HT. So, is it possible to go ling/hydra before colossi, and switch back to roaches/corruptors/infestors after colossi pop out? (this spend same amount of gas compare to roaches, so it won't affect the tech after, but need more larva and supply) (I've try this, but it doesn't work every time as my macro is not solid enough, and I hope that you can give me some suggestions, thanks)
How to deal with One Base Colossus? I've scouted it but I really didn't knew what to do.... his army easily overwhelmed me... Here is a replay: http://drop.sc/192611
So, is it possible to go ling/hydra before colossi, and switch back to roaches/corruptors/infestors after colossi pop out? (this spend same amount of gas compare to roaches, so it won't affect the tech after, but need more larva and supply) (I've try this, but it doesn't work every time as my macro is not solid enough, and I hope that you can give me some suggestions, thanks)
No. You can't really afford the tech. And making so many hydras leaves you vulnerable to colossi switches. Which would be fine if you were the one being aggressive (like if he went stargate), but against immortal/sentry, you are being defensive. You need a large army to beat immortal/sentry, zerg doesn't quite work like counter x with y. Immortal/sentrry will rape hydras in equal numbers, hydras really only work if they have an overwhelming roach/ling army to sort of tank for them. You can't really afford all of that in time.
So when I first wrote this guide, I was really struggling to hit 70+ on my own. Hell, I even struggled to hit 60+. I also thought "wow, some pros really screw up their macro, they make THREE overlords early on in the 50s! wtf?". I was a little confused why I struggled to hit 60+, even though I thought I knew so much. And you know, i am super smart, but I was too smart, that was the problem, hehe.
So here's some observations I made of pros, that I incorporate in my play now, and I can hit 70+ easily now in build order testers, and 60+ regularly in games (seriously, being only at the 60's, is SO hard to deal with things like immortal/sentry all-in or fast thirds, and just impossible to win if you are in the 50's, because you either have no econ, or you are making drones when you need to be making units to hold pushes or punish fast thirds).
- Pros rally drones to new bases, they don't maynard. Consistently, what I see of pros, especially ones who regularly hit 70+, is they will not maynard drones from their main to their natural, or main/natural to third. What they do, is when their natural pops, they rally new drones to the natural (only exception is maybe if natural is really late, which can happen if pylon blocked and you don't take your third right away). Maynarding drones from your main/nat to third also really hurts.
What pros do, is they just have their hatches rally new drones to the natural. Then, when the third pops, they rally new drones to the third. The timing works out in a way, that generally, when you start ralling to your third, your main/nat are even (after taking both gas in your main, since main will be oversaturated, which means make your buildings in the main, so it even outs, especially if you take both gas there), and then you just need to really rally to your third for a second, since when it pops, you usually ahve 2x injects to go there.
So what I do now, is rally drones to natural, and then when third is about to pop, I make a ton of drones with the larva injects, and just rally those eggs collectively to the third until it's about 12 drones, as that would be even in all 3 bases, and just keep the rally points from each hatch to their respective bases.
- Pros tend to make 2-3 overlords all at once at ~50/54 supply. You would think this is bad macro, and well, it's not perfect, but you literally have a ton of things going on all at once here. You tend to lose an overlord at this point, which is 7:00, to sacrifice as a scout. Secondly, you just have a ton of things going on right now - you are dealing with zealot/stalker pressure, dealing with probes trying to set pylons, scouting Toss, making lair, roach warren, evo, speed, and 3xinjects popping. So you really need these 3 overlords all at once, as you suddenly have a huge boost in larva.
Pros really do this a lot, and quite frankly, I think it's smart. 3 overlords x 8 supply is 24, bringing you from 54 cap, to 78. You tend to lose one of these overlords, so it puts you ave 70 (so make a 4th soon after if you are about to lose one to sacrifice). You can't really make all drones, then make an overlord from natural larva, it really is best just to make these 3 overlords in a very, very short span of eachother.
Just a note. I think I'll edit the macro section on this. It's what's helped me personally, a lot to make 3x overlords at 51/54, and don't maynard, but instead rally drones. You balance the mineral fields out by smart rallying, and making buildings/gases at mineral lines that are a bit oversaturated.
Thanks! I've been getting quite a lot of feedback about this guide recently, I didn't realize how many people found it so helpful.
I've heard a few people say to me on ladder that "lol your guide sucks" but i dont know why. havent heard any criticism besides 'you suck scrub bad guides newb' and then i own them and tell them "hey you clearly should read my guides" hahaha
I've been benefiting from this guide for a while, but finally registered so I thought I would say thank you for posting it, it's helped quite a bit.
Also, I know this has been edited since I last read it, and this is a decent length thread, so apologies if it's been mentioned before, but I would recommend emphasizing just how reliant on smooth macro this build is, as that is what I was initially struggling with, and I know many others have. If you aren't hitting your injects, you will outright lose to someone of your own level, and if you have the mental habit of 'scaling back' benchmarks to what would be appropriate in your league, it can be very difficult to understand what you're doing wrong.
Also, when I have more time, I would love to contribute to the streamlining debates when I have time, as after working on that slightly, I quickly went from struggling to break 60 supply, to coming up to 80 and over-droning in practice situations. Unfortunately for now it looks like I have a lovely ZvT guide that I had missed (yours), and so need to go make that my top matchup.
Crazy right? As impossible as it sounds, and as amazing as it seems already, DRG's macro has actually improved a lot in the last couple months.
So we saw in the GSL Genius vs DRG finals that DRG was consistently reaching 70+ supply, although he had a few games where he was below 70, I think 68 on daybreak due to early aggression.
On a recent stream game, however, he reached 81 supply at the 8:00 mark, and this is when a 4 gate style pressure of some sort hit around 7:30, and forced him to make roaches and units around then and he had to pull drones off his third around 7:30 (he keeps it, but loses due to close spawn entombed about 10 minutes later).
In the MLG finals, DRG was consistently 70+ supply, unlike the GSL finals, and he hit 75+ most of them, and 80+ a few times. If you go over his recent GSTL games and GSL games, he is appearing to hit over 75+ in all his ZvP now, and 80+ many times.
I didn't think it was possible to go 80+, and it was quite the benchmark when DRG his 78 in the genius finals.
Here's some observations I made of his games: - Getting pylon blocked costs zerg (and toss) about 3-5 supply - DRG does not make a 3rd queen pre-emptively, but rather, when the third finishes, he makes the 3rd queen there. This is different from what he did in the Genius vs DRG GSL finals, when he made his queens early. - DRG tends to only make 2 lings at the start instead of 4, but tends to make 4 more lings around 30 supply. A little risky if someone pylon blocks your third, but a gamble he takes... - DRG makes his 2nd queen before third. DRG seems to take a little bit later of a third than most zergs, as most take third before 2nd queen. DRG sends a drone out around 20 supply to the third, and takes it at 22-24. - DRG takes 3xGas at 6:30+, usually 6:40-6:45. I think this really boosts his econ more than the 2x gas style, because you'll notice DRG hits 50+ supply at the 6:00 mark, whereas most zergs hit that at 7:00 mark. - Most Zergs make 3 overlords at once at ~50 supply.
He also seems to make overlords, at 36+ supply, when he reaches cap. So an overlord at 36/44, then when he hits 44/44, he makes another one, and he's jsut at 44/44 for a split second as that earlier overlord finshes, and he constantly makes an overlord when he reaches cap from this point onward.
and as I stated earlier, DRG does not maynard workers, he rallies from main to nat right before it pops so drones arrive when it finishes, then rallies to third instead of maynarding.
As a toss i completely agree with the points related to zerg macro
Basically, the way i go is +1 4 gate attack - i think this is by far the best toss opener as you HAVE to force something other than drones. All other openers (4 gas, 3 gas) are basically waaay too passive and put the ball in zergs court.
I think in general, you can't allow zerg to have 70+ supply by 8 minutes or you are totally fucked as toss.
I pretty much benchmark all my games vs zerg by how much supply they have when i attack at just before 8 minutes and i pretty much lose any game where they have good macro. I don't think i've EVER won a game where i do some kind of strong 2 base attack where i don't push out before 8:00 and zerg macro's correctly.
For now it's ok, there are ALOT of bad macro zergs high up on ladder, but i absolutely think that ZvP becomes a MUCH easier matchup when you are capable of getting 70+ supply at that mark. As in, you literally are able to wipe the floor with almost any toss out there, and you at the very minimum garuantee yourself an easy path to infestor BL lategame.
On June 25 2012 15:55 Belial88 wrote: MLG Finals DRG vs Alicia Set1.
DRG REACHES 80 SUPPLY!
Crazy right? As impossible as it sounds, and as amazing as it seems already, DRG's macro has actually improved a lot in the last couple months.
So we saw in the GSL Genius vs DRG finals that DRG was consistently reaching 70+ supply, although he had a few games where he was below 70, I think 68 on daybreak due to early aggression.
On a recent stream game, however, he reached 81 supply at the 8:00 mark, and this is when a 4 gate style pressure of some sort hit around 7:30, and forced him to make roaches and units around then and he had to pull drones off his third around 7:30 (he keeps it, but loses due to close spawn entombed about 10 minutes later).
In the MLG finals, DRG was consistently 70+ supply, unlike the GSL finals, and he hit 75+ most of them, and 80+ a few times. If you go over his recent GSTL games and GSL games, he is appearing to hit over 75+ in all his ZvP now, and 80+ many times.
I didn't think it was possible to go 80+, and it was quite the benchmark when DRG his 78 in the genius finals.
Here's some observations I made of his games: - Getting pylon blocked costs zerg (and toss) about 3-5 supply - DRG does not make a 3rd queen pre-emptively, but rather, when the third finishes, he makes the 3rd queen there. This is different from what he did in the Genius vs DRG GSL finals, when he made his queens early. - DRG tends to only make 2 lings at the start instead of 4, but tends to make 4 more lings around 30 supply. A little risky if someone pylon blocks your third, but a gamble he takes... - DRG makes his 2nd queen before third. DRG seems to take a little bit later of a third than most zergs, as most take third before 2nd queen. DRG sends a drone out around 20 supply to the third, and takes it at 22-24. - DRG takes 3xGas at 6:30+, usually 6:40-6:45. I think this really boosts his econ more than the 2x gas style, because you'll notice DRG hits 50+ supply at the 6:00 mark, whereas most zergs hit that at 7:00 mark. - Most Zergs make 3 overlords at once at ~50 supply.
He also seems to make overlords, at 36+ supply, when he reaches cap. So an overlord at 36/44, then when he hits 44/44, he makes another one, and he's jsut at 44/44 for a split second as that earlier overlord finshes, and he constantly makes an overlord when he reaches cap from this point onward.
and as I stated earlier, DRG does not maynard workers, he rallies from main to nat right before it pops so drones arrive when it finishes, then rallies to third instead of maynarding.
As always, thanks for your work Belial !
Now, we should ask ourselves what are the pros & cons of DRG's macro boost style besides the Stephano's one ? Like the 3 delayed gas ? What are the risks doing something like this ? If i understand, he takes the 3 gas and the RW at the same time ?
If you scout +1 zealot warpgates coming, how many drones do you expect to cut (or what supply at 8:00?)
Also, what if he goes nexus forge gate (all on 17) and gets out his zealot at 7:30, whats your plan then? How many drones/supply do you expect to have then?
He didn't take the roach warren, he saw with his 2 overlords sacrfice (yes on top of that 80 supply in 8 minutes, he's able to afford 2 overlords sacrifice !) the robotic and the 2 gas in the natural. He decided that making roach wasn't the best answer for this protoss build and thus decided to plan mutas.
On June 25 2012 15:55 Belial88 wrote: MLG Finals DRG vs Alicia Set1.
DRG REACHES 80 SUPPLY!
Crazy right? As impossible as it sounds, and as amazing as it seems already, DRG's macro has actually improved a lot in the last couple months.
So we saw in the GSL Genius vs DRG finals that DRG was consistently reaching 70+ supply, although he had a few games where he was below 70, I think 68 on daybreak due to early aggression.
On a recent stream game, however, he reached 81 supply at the 8:00 mark, and this is when a 4 gate style pressure of some sort hit around 7:30, and forced him to make roaches and units around then and he had to pull drones off his third around 7:30 (he keeps it, but loses due to close spawn entombed about 10 minutes later).
In the MLG finals, DRG was consistently 70+ supply, unlike the GSL finals, and he hit 75+ most of them, and 80+ a few times. If you go over his recent GSTL games and GSL games, he is appearing to hit over 75+ in all his ZvP now, and 80+ many times.
I didn't think it was possible to go 80+, and it was quite the benchmark when DRG his 78 in the genius finals.
Here's some observations I made of his games: - Getting pylon blocked costs zerg (and toss) about 3-5 supply - DRG does not make a 3rd queen pre-emptively, but rather, when the third finishes, he makes the 3rd queen there. This is different from what he did in the Genius vs DRG GSL finals, when he made his queens early. - DRG tends to only make 2 lings at the start instead of 4, but tends to make 4 more lings around 30 supply. A little risky if someone pylon blocks your third, but a gamble he takes... - DRG makes his 2nd queen before third. DRG seems to take a little bit later of a third than most zergs, as most take third before 2nd queen. DRG sends a drone out around 20 supply to the third, and takes it at 22-24. - DRG takes 3xGas at 6:30+, usually 6:40-6:45. I think this really boosts his econ more than the 2x gas style, because you'll notice DRG hits 50+ supply at the 6:00 mark, whereas most zergs hit that at 7:00 mark. - Most Zergs make 3 overlords at once at ~50 supply.
He also seems to make overlords, at 36+ supply, when he reaches cap. So an overlord at 36/44, then when he hits 44/44, he makes another one, and he's jsut at 44/44 for a split second as that earlier overlord finshes, and he constantly makes an overlord when he reaches cap from this point onward.
and as I stated earlier, DRG does not maynard workers, he rallies from main to nat right before it pops so drones arrive when it finishes, then rallies to third instead of maynarding.
I've seen kinda the same thing in Stephano stream, he rally his drone to the new base when it's necesarry, and re rally when he takes gas or make structure. The only time I've seen him maynarding is when he has to take the third before the second base, depending on how much delayed his base was. About the third queen timing, it might gives you a better economy, but I have the feeling that it muse really delay your creep spread, which could be problematic against stargate.
Basically, the way i go is +1 4 gate attack - i think this is by far the best toss opener as you HAVE to force something other than drones. All other openers (4 gas, 3 gas) are basically waaay too passive and put the ball in zergs court.
This attack puts you REALLY far behind if Zerg responds correctly with just a 6:30 roach warren when he sees no gas at your natural.
I wouldn't say it's an autowin for Zerg, but I'd say it's about as close as you can come to one. There's a reason you don't see it in pro play anymore.
That said, if it works for you, keep doing it. It's definitely strong, and there is a reason why it was popular for a while
Like the 3 delayed gas ? What are the risks doing something like this ? If i understand, he takes the 3 gas and the RW at the same time ?
Just stylistic, for the most part. I think your first 100 gas will be a little later with this style, but just by a tiny bit.
Also quick question to belial
If you scout +1 zealot warpgates coming, how many drones do you expect to cut (or what supply at 8:00?)
Also, what if he goes nexus forge gate (all on 17) and gets out his zealot at 7:30, whats your plan then? How many drones/supply do you expect to have then?
If you scout +1 zealot 4 gate pressure coming, you probably cut drones around 50. You still should be around 70 supply, but you just have roaches and lings instead of drones. If you respond correctly though, Toss will be really far behind, as he has no tech, and only 4 gates, and no sentries.
If you go nexus forge gate and gets out a zealot at 7:30.... i dont know what you mean, the 4 gate +1 will still hit at about 8:00 since you delayed your tech for more econ, and the core still is down at the same time as forge/nexus.
He didn't take the roach warren, he saw with his 2 overlords sacrfice (yes on top of that 80 supply in 8 minutes, he's able to afford 2 overlords sacrifice !) the robotic and the 2 gas in the natural. He decided that making roach wasn't the best answer for this protoss build and thus decided to plan mutas.
really? im going to look again at those games, very cool.
Dam, dude, I'd buy you a beer if you were in Ukraine.
I couldn't take a game off protoss for two seasons, even though I didn't practice much, I played well when I did play. I slid from higher masters to mid to low to diamond because I just lost every game against toss (and also because of Skyrim and D3). Somehow I was convinced that 60 probes was enough against protoss. As I think about it right now I don't even know where I got this idea, maybe because while expecting an attack I started stacking combat units early. Anyway, I fixed my drone count, got to 80 supply in 8 minutes (yeah I can do that because diamond tosses are pretty nub and 4 zerglings is enough to terrorize them for 8 minutes). And then at a 10 minute mark protoss looked like a baby to me. My BLs came out before he had the second colossus. The other game I went up to 100 drones against toss and had the most fun when I ran over zealots+archons with 4 waves of 200 zerglings :D
You brought SC fun back into my life, thank you!
P.S. Where did the term "maynard" come from? Keynes reference? I'm just curious, can't find it in any dictionary.
P.S. Where did the term "maynard" come from? Keynes reference? I'm just curious, can't find it in any dictionary.
It came from a BW player called Maynard who was the first to bring drone to his new base.
Your welcome
Pffft lol thanks. At least they didn't name Attack-Move command after a guy who first did that. If his name was Jack... don't jack your zealots onto the queen!
I expected something more sophisticated. I thought it was after John Maynard Keynes, the father of macro-economics. It would be cooler that way.
Basically, the way i go is +1 4 gate attack - i think this is by far the best toss opener as you HAVE to force something other than drones. All other openers (4 gas, 3 gas) are basically waaay too passive and put the ball in zergs court.
This attack puts you REALLY far behind if Zerg responds correctly with just a 6:30 roach warren when he sees no gas at your natural.
I wouldn't say it's an autowin for Zerg, but I'd say it's about as close as you can come to one. There's a reason you don't see it in pro play anymore.
That said, if it works for you, keep doing it. It's definitely strong, and there is a reason why it was popular for a while
If you scout +1 zealot warpgates coming, how many drones do you expect to cut (or what supply at 8:00?)
Also, what if he goes nexus forge gate (all on 17) and gets out his zealot at 7:30, whats your plan then? How many drones/supply do you expect to have then?
If you scout +1 zealot 4 gate pressure coming, you probably cut drones around 50. You still should be around 70 supply, but you just have roaches and lings instead of drones. If you respond correctly though, Toss will be really far behind, as he has no tech, and only 4 gates, and no sentries.
If you go nexus forge gate and gets out a zealot at 7:30.... i dont know what you mean, the 4 gate +1 will still hit at about 8:00 since you delayed your tech for more econ, and the core still is down at the same time as forge/nexus.
He didn't take the roach warren, he saw with his 2 overlords sacrfice (yes on top of that 80 supply in 8 minutes, he's able to afford 2 overlords sacrifice !) the robotic and the 2 gas in the natural. He decided that making roach wasn't the best answer for this protoss build and thus decided to plan mutas.
really? im going to look again at those games, very cool.
What i meant was if he has 4 warping zealots with +1, warping in at 7:30 (instead of 8:00)
It's disappointing to hear it's basically unused now. I hate playing in such a way i allow zerg to macro nothing but drones freely until 8 minutes, it just feels like an absolute losing proposition to me.
What do you reccomend protoss play for the first 8-9 minutes should be geared towards?
^ You should submit a replay, because I've never seen anyone warp in a full 30 seconds earlier than 8:00 off a FFE...
There's nothing wrong with 'allowing' zerg to macro freely until 8 minute, you can't really stop it. That would be like zerg saying "I hate letting toss take an expansion off just a forge". Besides some drastic all-in that can be completely stopped when scouted, there isn't anything you can do to stop it.
You know, I've been running into Toss on certain maps, like Cloud Kingdom, who take a super fast third off 1 gate or 1 gate 1 robo, and use 2-3 cannons and wall-offs to secure their third that's taken at like 7-9 minute. I always felt there was nothing I could do to stop it, and I can't, as zerg drones up to 8:30.
I would 'recommend' for Toss to do some sort of robo opening, either 6-8 gates for a timing or 3-5 gates for an expansion, and play off zerg not knowing exactly which variation you are doing, and push out with your first two immortals around 9:30 or take your expansion around 8-9:00. That seems to be the standard these days. Toss can definitely keep up with Zerg in the macro game, I'm pretty sure most Zergs complain more about lategame Toss than mid-game Toss.
I didn't really optimise this (i forgot probes for a short bit after my gates, i also forgot to spend some chrono on forge), but you get the idea. A good player can get the warpin maybe even earlier, but its definitely around a 7:30 warpin with this build. Cons are obviously, no room to pylon block the hatchery
On June 27 2012 04:24 Belial88 wrote:You know, I've been running into Toss on certain maps, like Cloud Kingdom, who take a super fast third off 1 gate or 1 gate 1 robo, and use 2-3 cannons and wall-offs to secure their third that's taken at like 7-9 minute. I always felt there was nothing I could do to stop it, and I can't, as zerg drones up to 8:30.
I can see this pattern playing out to its logical extreme. When protoss realized they couldn't deny zerg's fast expand off one base, they started fast expanding. When zerg realized they couldn't deny protoss's fast expand off two base, they started taking a fast third. As protoss is realizing they can't deny zerg's fast third, they are starting to take a fast third. Zergs will learn that the best answer is to rush a fourth base. By next year, I expect to see the matchup devolve into who can be the first to hit 200 workers.
I didn't really optimise this (i forgot probes for a short bit after my gates, i also forgot to spend some chrono on forge), but you get the idea. A good player can get the warpin maybe even earlier, but its definitely around a 7:30 warpin with this build. Cons are obviously, no room to pylon block the hatchery
You went nexus, forge, gate, 2nd pylon. That seemed greedy as fuck. I think a standard 14 pool could punish that?
I don't know. Maybe try the build going a more mild, nexus first, forge, cannon, then gate and 2nd pylon? I mean I know you could maybe get away with that in a real game, but I don't know. That could definitely be a large part of how you hit 7:30 complete warp gate.
You also didn't make a 2nd unit out of your gateway. You would've just died if anyone did any sort of all-in, roach or baneling based. I would imagine in a real game, you would probably make a sentry, or send out the zealot and then make a 2nd zealot.
Your second warp in is at like 7:50, which is a more reasonable timing. Your initial warp in of 4 zealots is kind of like if you just pushed out with 3 units from gateway.
i dont know. very cute though, i think a combination of how many corners you cut and how greedy your build was allowed you to shave a few seconds off, but in a real game I think zerg who has roaches popping at 8:00 from his 6:30 roach warren will still defend it, just the same as if it was a 4 gate +1 done from a more standard/safe opening. Interesting to see how tangible that nexus first and greedier stuff was though, I'll definitely take that into account in future games when playing Toss who go nexus first!
I didn't really optimise this (i forgot probes for a short bit after my gates, i also forgot to spend some chrono on forge), but you get the idea. A good player can get the warpin maybe even earlier, but its definitely around a 7:30 warpin with this build. Cons are obviously, no room to pylon block the hatchery
You went nexus, forge, gate, 2nd pylon. That seemed greedy as fuck. I think a standard 14 pool could punish that?
I don't know. Maybe try the build going a more mild, nexus first, forge, cannon, then gate and 2nd pylon? I mean I know you could maybe get away with that in a real game, but I don't know. That could definitely be a large part of how you hit 7:30 complete warp gate.
You also didn't make a 2nd unit out of your gateway. You would've just died if anyone did any sort of all-in, roach or baneling based. I would imagine in a real game, you would probably make a sentry, or send out the zealot and then make a 2nd zealot.
Your second warp in is at like 7:50, which is a more reasonable timing. Your initial warp in of 4 zealots is kind of like if you just pushed out with 3 units from gateway.
i dont know. very cute though, i think a combination of how many corners you cut and how greedy your build was allowed you to shave a few seconds off, but in a real game I think zerg who has roaches popping at 8:00 from his 6:30 roach warren will still defend it, just the same as if it was a 4 gate +1 done from a more standard/safe opening. Interesting to see how tangible that nexus first and greedier stuff was though, I'll definitely take that into account in future games when playing Toss who go nexus first!
*always* nexus first is safe vs 14 pool
As it stands, your cannon is no later if you get nexus, forge, gate, pylon all on 17 ( as you saw i built my cannon as soon as the forge finished)
Also in terms of unit's built, i can get a sentry, or a stalker or w/e. There is plenty of room to allocate resources however you want. There is also the option to pull guys off gas if you want to have more minerals for more units (not reccommended). I havn't optimised it. Usually i feel safe just getting zealot zealot or zealot stalker when seeing a 3rd base, otherwise i'll do something else
One of the good things about the build is the obvious fast warpgate which is handy vs a lot of all in variants, depending on untis built.
I also personally think nexus first should be automatic, and that you should be delaying your cannon as late as possible to get down an earlier gateway always. You can always drop a forge and wall off on a scout, the only thing the force does is let you get a cannon in time - so technically as long as your cannon is in time your forge can come whenever you want.
Vs hatch first i often go nexus - gate - forge, and i might even skip the cannon until later and you can get some 7:00 +1 warpin timings instead.
^ Nexus first is definitely safe against 14 pool. Casters are dumb when they say it isn't, or "oh snap hes being greedy!", but going nexus, forge, gate, 2nd pylon, is as greedy as you can be and is not safe.
There are smarter and better Toss players than me out there, but I think you probably need to incorporate something like a pylon block to make sure lings dont come in and screw you up. You couldn't do that greedy of a nexus first against a zerg who makes 6 lings when he sees you make a gateway instead of cannon first.
I dont think anything you are saying is wrong, I just think that nexus, forge, gate, 2nd pylon, then cannon, is a bit too greedy and lings will get in against that. If you reactively make a cannon earlier, that's fine, but I don't think the exact build, as you did it, would work in a real game. I could be wrong, but I know I have reactively made lings against people who didn't make a cannon after their nexus, forge, and got the lings in. Things like nexus, forge, gate.
On June 27 2012 16:51 Belial88 wrote: ^ Nexus first is definitely safe against 14 pool. Casters are dumb when they say it isn't, or "oh snap hes being greedy!", but going nexus, forge, gate, 2nd pylon, is as greedy as you can be and is not safe.
There are smarter and better Toss players than me out there, but I think you probably need to incorporate something like a pylon block to make sure lings dont come in and screw you up. You couldn't do that greedy of a nexus first against a zerg who makes 6 lings when he sees you make a gateway instead of cannon first.
I dont think anything you are saying is wrong, I just think that nexus, forge, gate, 2nd pylon, then cannon, is a bit too greedy and lings will get in against that. If you reactively make a cannon earlier, that's fine, but I don't think the exact build, as you did it, would work in a real game. I could be wrong, but I know I have reactively made lings against people who didn't make a cannon after their nexus, forge, and got the lings in. Things like nexus, forge, gate.
Whether i make the gate or pylon after have absolutely no impact on the cannon timing because the cannon is build as soon as the forge is finished...?
General rule of thumb is if they pool/hatch w/o getting lings out first you're safe to delay cannon. For the most part zergs will be hunting a probe and not sending all of them straight to the base regardless if they haven't found the probe yet, and even in the instance that they do P should have a probe blocking a zealot space to drop a pylon to delay if need be.
On June 27 2012 17:52 Belial88 wrote: Your cannon is not built when the forge is finished.
Depends on the build. If u chrono probes 3 times and build the nexus on 17 and forge on 17, u can put a pylon on 17 and a gate on 18 before the forge is finished. Your forge finishes on 18-19 supply right when you have 150 minerals. As far as its safety is concerned, if u build ur 2nd pylon as part of ur wall, u can block one space entrance long enough for cannon to be ready. But if u go for a forge gate cyber wall, if the zerg goes 6 lings off the 14 pool and send them straight to your base, they can get past before the cannon finishes unless they build a 2nd gate to block it off. The second gate hurts ur build quite a bit and if u are doing the standard nexus first gas timings and constant probes, u will only have about 70-80 minerals when the lings arrive so it's difficult to do reactively without cutting probes in case he does try to go for the run-by.
I go triple chrono, 17 nexus, forge, gate, pylon. On most maps you can complete the wall with another pylon if the lings really are too fast, though off a 14 pool this shouldn't happen.
You have to do a little bit of trickery with your mineral usage as to have enough to get double gas + cyber, and your chrono's are to be saved for your cyber and forge
MC's one gas 7 gate warps in just after 7:30, and could make it down to that mark I believe. He used it against Stephano at Red Bull in games 2 and 3, or 3 and 4? don't quite remember.
As to the Maynarding, I've never really seen why it's done so frequently. If you transfer 5 drones at one, or rally them as they build, you are losing the same amount of mining time, only that you are losing it all at once, and much earlier in your build by maynarding. Of course it still has uses, but I rally whenever possible.
I know i can reactively make a cannon when his lings come out depending on the number of lings and a pylon wall but yeah T_T
Right, and then you aren't doing nexus, forge, gateway, 2nd pylon, then cannon, you are doing nexus, forge, cannon or nexus, forge, gateway with paused probe production into cannon or something. And that would affect the timing.
As to the Maynarding, I've never really seen why it's done so frequently. If you transfer 5 drones at one, or rally them as they build, you are losing the same amount of mining time, only that you are losing it all at once, and much earlier in your build by maynarding. Of course it still has uses, but I rally whenever possible.
Maynarding is only bad with fast third because of 3 bases taken in quick succession, and the way zerg's macro works with making multiple drones at once instead of one at a time. It's better to maynard with other races, or if you take your natural and third later (like if you get pylon blocked and do some sort of 2 base lair style, although you might have the income and larva by then to just instantly saturate your third instead of having to maynard too). The idea behind it is still good though. I think the reason maynarding hurts so much when going fast third, is because of the quick succession of 2 extra bases, zerg making multiple drones at once with larva inject, and because it's so early in the game that a small difference of mining time and drones in the early game leads to a huge impact at the 8-9:00 mark, so it's worth it to hurt your mining later on, if it allows you just a little more mining early on (ie not transfering workers but instead rallying new ones).
I know i can reactively make a cannon when his lings come out depending on the number of lings and a pylon wall but yeah T_T
Right, and then you aren't doing nexus, forge, gateway, 2nd pylon, then cannon, you are doing nexus, forge, cannon or nexus, forge, gateway with paused probe production into cannon or something. And that would affect the timing.
As to the Maynarding, I've never really seen why it's done so frequently. If you transfer 5 drones at one, or rally them as they build, you are losing the same amount of mining time, only that you are losing it all at once, and much earlier in your build by maynarding. Of course it still has uses, but I rally whenever possible.
Maynarding is only bad with fast third because of 3 bases taken in quick succession, and the way zerg's macro works with making multiple drones at once instead of one at a time. It's better to maynard with other races, or if you take your natural and third later (like if you get pylon blocked and do some sort of 2 base lair style, although you might have the income and larva by then to just instantly saturate your third instead of having to maynard too). The idea behind it is still good though. I think the reason maynarding hurts so much when going fast third, is because of the quick succession of 2 extra bases, zerg making multiple drones at once with larva inject, and because it's so early in the game that a small difference of mining time and drones in the early game leads to a huge impact at the 8-9:00 mark, so it's worth it to hurt your mining later on, if it allows you just a little more mining early on (ie not transfering workers but instead rallying new ones).
3 chrono, 17 nexus 17 forge 17 gateway 17 pylon cannon as soon as the forge is done 1 gas before gateway done cyber as soon as gateway is done
^ msg me belial.869@na and we can test this real quick. but this is regardless anyways, a 7 pool autowins against you too. my point is you are doing a greedy opener, that's why your wg is quicker.
On June 28 2012 05:59 ImUnemployed wrote: As to the Maynarding, I've never really seen why it's done so frequently. If you transfer 5 drones at one, or rally them as they build, you are losing the same amount of mining time, only that you are losing it all at once, and much earlier in your build by maynarding. Of course it still has uses, but I rally whenever possible.
I suspect that a lot of ex-broodwar players maynard simply out of habit, without reevaluating why they do it, and people who learn from them pick up the habit without evaluating if it's good or not.
Maynarding was a lot more important in Broodwar because the dropoff in mining from oversaturation wasn't as binary. You started losing efficiency earlier and capped out later. Therefore, maynarding workers was the most effective way to mitigate the economic bump you take from expanding.
In SC2, I feel like there is much more incentive to time your expansions together with the saturation of your bases because the benefit of workers past saturation is so small, and there is almost no dropoff before the hard cap, which makes maynarding something you just do when your expansion timing has been delayed.
Unlike Broodwar, the mining of my 9th-16th workers isn't any better at my natural than it is at my main, while I get almost 0 benefit from my 17th worker if I keep it at my main.
And because of this, it follows that Terran and Protoss will have more opportunities to maynard than zerg will, because they will be forced to keep building workers past saturation even if their expansion isn't finished building yet, while zerg doesn't have to start droning a base until it's built.
On June 28 2012 17:44 Belial88 wrote: ^ msg me belial.869@na and we can test this real quick. but this is regardless anyways, a 7 pool autowins against you too. my point is you are doing a greedy opener, that's why your wg is quicker.
Doesn't a pylon scout arrive in time to not nexus first on a 2 player map? That's been my experience as zerg anyway, maybe a 7pool is faster than I am imagining in my head.
^ Maybe, but you certainly aren't going to do a nexus first anymore, and if you did, you would autolose. Which is why I think his build isn't too honest, he cut just a shitton of corners.
Hi, I really love this guide, but always struggled in the lategame. Now as I switched over to the Infestor/Spine turtle play style I got much better in this aspect, but much worser in defending all-ins. I just never know, whether to build Spines/Infestation Pit or pump out Zergling/Roach. I also don't know wether I should get +1 Melee or +1 Range. As it seems to be better to get +1 Range to defend the all-ins but +1 Melee is better for the turtle style.
So should I always get the infestation pit right as the lair finishes? Or just when the Protoss takes the third? Which upgrade should I get first? What should I do if I crush his 3rd with some lings and Roaches? Should I throw down the Infestation Pit, or just Max Out on Roaches and Lings and continue attacking and containing? Because it often happens that the Protoss transitions into an 2 base all-in after his 3rd gone down. But sometimes they will try to put it down again. So I really don't know how to act in such a situation.
I watched the Day9 daily on the Stephano turtle style and it seems like he throws down the infestation pit always immediatly.
how would I defend an 6gate robo all in? It seems that +1 zealots just massacre my lings when i get them near the toss while my roaches get stuck in FFs and get hell rained on them by those immortals... and i can't even retreat cuz half of my units get stuck in the ffs....
Check out the "Nestea anti-immortal/sentry build' thread. I don't think there's a common answer yet on immortal/sentry all-ins yet. I'm not sure it's completely figure out, but... seems to be a few ways.
1. I talk about immortal/sentry in this guide - you just max out on roach/ling at 8-8:30+, engage well. 2. When he pushes out, the roach/ling you have just base trades while you mass spines in your nat and give up your third (pull the drones when he arrives). 3. Rush mutas (as talked about in that thread) with the same idea - mass spines in nat, base trade.
But i talk about it in this guide, there's also a very good replay example in this thread on how to beat it.
max at 200? wow lol at 8 mins i usually only have 120-160 ish supply. meybes it is my macro So its pretty much like against marine medic in BW? jsut mass stationary def
On July 13 2012 03:53 STYDawn wrote: max at 200? wow lol at 8 mins i usually only have 120-160 ish supply. meybes it is my macro So its pretty much like against marine medic in BW? jsut mass stationary def
He means at 8-8:30 just constantly produce roach/ling, not actually max out at that time (which is impossible )
On July 25 2012 22:07 InfCereal wrote: Is the 200/200 push still viable?
I never did it when it was working, and I'm lost in zvp right now.
Yep still works, stephano still uses it to great effect.
Anything in particular you need to do, or just a move and wait out forcefields?
Never get supply blocked or miss an inject or this style does not work.
If he's going sentry immortal try to fight anywhere but your base.
Always have lings in your composition, pure roach isn't nearly as good as roach/ling. (it's really easy to make nothing but roaches and feel you're macroing well, when you're not)
Is it still true that most masters struggle to get beyond 60 supply by the 8 minute mark when going Fast Third vs. FFE? I don't really understand this, since I'm in Gold and manage to get to 70 supply half the time and basically always hit 60. I'm not saying I have great macro: in fact, it seems more likely that players have progressed greatly since this guide was written, instead of me somehow managing to play the early game far above my league.
The quality of my opponents is probably a big thing too. I basically never get harassed/poked by the 8-minute mark, which would naturally slow me down.
On August 10 2012 19:43 Salivanth wrote: Is it still true that most masters struggle to get beyond 60 supply by the 8 minute mark when going Fast Third vs. FFE? I don't really understand this, since I'm in Gold and manage to get to 70 supply half the time and basically always hit 60. I'm not saying I have great macro: in fact, it seems more likely that players have progressed greatly since this guide was written, instead of me somehow managing to play the early game far above my league.
The quality of my opponents is probably a big thing too. I basically never get harassed/poked by the 8-minute mark, which would naturally slow me down.
no you need to really fuck up to get less than 60 supply at 8min.
I'd rather hear the answer from Belial tbh, given he was the one who originally made the claim, and basically everything else in this guide is correct, so I'm inclined to trust him.
I don't think any masters struggle to get past 60, I think they struggle to get past 70 is what I said (meant?i think i meant past 60, as in struggle to get past the 60s and into the 70s). If you are below 60 supply you should just leave the game because it's really far behind, I don't think you'll hold any timing (although if he's going fast third you could be greedy as fuck and make up for it by 15:00). 70+ is decent macro, 75+ for good macro, 80+ for really good. I think most lower masters struggle to hit 70+ regularly though, so if you are hitting 70+ half the time as gold you are doing very good for yourself (assuming you have roach/evo/lair/3-4gas/speed/3queens)!
But as the guy said, yea, you'll really need to fuck up to get less than 60 supply by 8 min. Something like losing an initial overlord to a cannon in the natural or to a stalker, or just getting fucked up by zealot/stalker harass.
I've done some pretty big updates on the "How to React" section, as well as adding/cleaning a bit of the Playing the Game and Identifying the Build.
I've completely overhauled the replay section, and added a couple of replays on how to beat immortal/sentry all-in. I constantly add new replays every week, so if you haven't looked over this guide in a while, check that section out!
I've added a ton of in-depth content on Immortal/Sentry all-ins as well. Every other day someone asks about them in the Zerg Help Me Thread, so I've done a ton of research and playing around with that.
One of the best threads. Very in-deapth and comprehensive. I like the update as well.
Some of my thoughts on your immortal/sentry base race with roach/ling section: I am not sure massing 8+ spines in the natural works consistently. Natural is wide on some maps, especially Ohana where there are 2 entrances to the natural. If you can hold natural, of course that would be great, but making spines on top of main ramp where it is guaranteed choke for Protoss army would be better in some cases. As long as you kill protoss main base, you don't need your natural either. It is more importrant to make sure that your spines actually go up in time and fight in good location. I would probably put spines at natural on maps like Cloud Kingdom but at main on maps like Ohana.
Start spewing creep for all of your overlords and escape them to air space as well. From time to time, you get in situations where both sides lose all bases and literally trade bases. Protoss has to rebuild pylons while you don't have to remake overlords if you escape them to air space. Also, losing your lair doesn't stop overlord creeping as long as you started spewing before lair died.
^ The time it takes for Toss to bust the natural would be enough time to get mutas out. It wouldn't really make sense for Toss not to take down the third at the same time he's going for your natural. I mean, the third and rocks at the third take about the same time to kill, that's all you really need. All you need to make sure of, is that Toss doesnt race up into your main and kill the spire before you've been able to pump about 25+ mutas. I mean theoretically, if you put your spire in your natural, and then toss goes move-command straight into the main, or goes to the natural, kills that, then main, you'd be fine too. Hell, you could actually proxy the spire, and not make any spines at all in the natural, instead making them at the furthest base Toss attacks first to buy time, although that doesnt really matter too much either.
You aren't making the spines to hold the Toss push. You are making the spines to discourage Toss from going move-command into your main, and killing the spire there. Theoretically, if you proxy the spire, you don't need to make spines at all, or if you make the spines at your third and Toss goes to your main instead.
But you can't know for sure where Toss is going to go, and Toss will figure you put the spire in your main, making your main the go-to if he knows you are going quick mutas. So that's why you put spines in your natural, to buy time for the spire to finish, so you can get those mutas out. By putting spines in the natural, you force Toss to make additional warp-ins before he can bust up the ramp (or else you can come home with the lings and just kill his entire army of sentry/immortal with only a single warp in of units, as well as when the mutas pop). Thereby, forcing Toss to go to the third in the meantime. I mean it'd be awesome if Toss just sat afk in front of you base, waiting for his army to get large enough to bust up.
On Ohana, if he tries to go around and bust the rocks, you just re-root the spines to that side. If he moves toward the main ramp, uh oh, you just re-root the spines! And while this goes on for about 40 seconds, Toss has already lost every base, and you've been able to fully mine gas off 3 bases and will have 25 mutas very, very quickly, and it'll be much easier.
The base trade is a race against the clock, and Toss needs to do 2 things - kill the spire asap, which he can't if you make it in the main and make a ton of spines in the natural to at least buy time, if not kill his army if he's dumb enough to go straight into the natural before he has enough warp-ins, and cut off mining at as many bases as possible, which won't happen if he has to go through rocks or spines.
Really, going quick muta base trade with mass spines in the natural just hard counters not only immortal/sentry, but robo play in general. Hence, that's why you don't do immortal/sentry all-ins, it's prone to just being countered by preparation. If you actually go gas, lair, spire quick enough, you don't even need the spines (although that may require the spire be proxied).
Yes, making sure all overlords escape is a great tip when going base trade against immortal/sentry. That's something I'll add. If you watch the reps, you'll see I make a conscious effort to start moving all overlords over dead air as soon as I know it's a robo build. And yea, spewing creep will continue if you start it before lair goes down.
but it takes a long time to work down rocks. You just play a game of uproot spines over and over, and as you mine more and more minerals as toss' army does less and less, you can make more spines, more mutas, more units. Immortal/sentry is a timing, if he doesn't accomplish enough damage by a certain time, he just has no chance. With your spire up, and 15+ mutas started all at once when spire finished, the clock ticks from you having 15 mutas which won't beat the immortal/sentry army straight up, to 25+ mutas, which will just a-move against any immortal/sentry army and crush it easily.
^ The time it takes for Toss to bust the natural would be enough time to get mutas out. It wouldn't really make sense for Toss not to take down the third at the same time he's going for your natural. I mean, the third and rocks at the third take about the same time to kill, that's all you really need. All you need to make sure of, is that Toss doesnt race up into your main and kill the spire before you've been able to pump about 25+ mutas. I mean theoretically, if you put your spire in your natural, and then toss goes move-command straight into the main, or goes to the natural, kills that, then main, you'd be fine too. Hell, you could actually proxy the spire, and not make any spines at all in the natural, instead making them at the furthest base Toss attacks first to buy time, although that doesnt really matter too much either.
You aren't making the spines to hold the Toss push. You are making the spines to discourage Toss from going move-command into your main, and killing the spire there. Theoretically, if you proxy the spire, you don't need to make spines at all, or if you make the spines at your third and Toss goes to your main instead.
But you can't know for sure where Toss is going to go, and Toss will figure you put the spire in your main, making your main the go-to if he knows you are going quick mutas. So that's why you put spines in your natural, to buy time for the spire to finish, so you can get those mutas out. By putting spines in the natural, you force Toss to make additional warp-ins before he can bust up the ramp (or else you can come home with the lings and just kill his entire army of sentry/immortal with only a single warp in of units, as well as when the mutas pop). Thereby, forcing Toss to go to the third in the meantime. I mean it'd be awesome if Toss just sat afk in front of you base, waiting for his army to get large enough to bust up.
On Ohana, if he tries to go around and bust the rocks, you just re-root the spines to that side. If he moves toward the main ramp, uh oh, you just re-root the spines! And while this goes on for about 40 seconds, Toss has already lost every base, and you've been able to fully mine gas off 3 bases and will have 25 mutas very, very quickly, and it'll be much easier.
The base trade is a race against the clock, and Toss needs to do 2 things - kill the spire asap, which he can't if you make it in the main and make a ton of spines in the natural to at least buy time, if not kill his army if he's dumb enough to go straight into the natural before he has enough warp-ins, and cut off mining at as many bases as possible, which won't happen if he has to go through rocks or spines.
Really, going quick muta base trade with mass spines in the natural just hard counters not only immortal/sentry, but robo play in general. Hence, that's why you don't do immortal/sentry all-ins, it's prone to just being countered by preparation. If you actually go gas, lair, spire quick enough, you don't even need the spines (although that may require the spire be proxied).
Yes, making sure all overlords escape is a great tip when going base trade against immortal/sentry. That's something I'll add. If you watch the reps, you'll see I make a conscious effort to start moving all overlords over dead air as soon as I know it's a robo build. And yea, spewing creep will continue if you start it before lair goes down.
but it takes a long time to work down rocks. You just play a game of uproot spines over and over, and as you mine more and more minerals as toss' army does less and less, you can make more spines, more mutas, more units. Immortal/sentry is a timing, if he doesn't accomplish enough damage by a certain time, he just has no chance. With your spire up, and 15+ mutas started all at once when spire finished, the clock ticks from you having 15 mutas which won't beat the immortal/sentry army straight up, to 25+ mutas, which will just a-move against any immortal/sentry army and crush it easily.
I was talking about bace race with roach/ling, but I guess your answer carries over for non-spire scenario. Thank you for quick response.
My question is in regards to ZvP opening, when I open 14 pool -> hatch, when should I be making the 2nd overlord? The creation of the hatch + 4 lings + queen + overlord can't seem to flow smoothly for me.
Should I prefer delaying the hatch to make the OV on 16 then hatch? Should I drop the hatch on 16, skip the 16 drone so I can make 4lings + queen then the OV? ( I assume making the 4 lings ASAP is critical to hold a commited cannon rush or even just a pylon+cannon on the third )
If my hatch get's delayed by pylon how does that effect my OV timing? Should I go for the OV as soon as I see that my hatch won't be going down on 16? I find myself blocked for either the queen or the lings in that situation?
My question is in regards to ZvP opening, when I open 14 pool -> hatch, when should I be making the 2nd overlord? The creation of the hatch + 4 lings + queen + overlord can't seem to flow smoothly for me.
Should I prefer delaying the hatch to make the OV on 16 then hatch? Should I drop the hatch on 16, skip the 16 drone so I can make 4lings + queen then the OV? ( I assume making the 4 lings ASAP is critical to hold a commited cannon rush or even just a pylon+cannon on the third )
If my hatch get's delayed by pylon how does that effect my OV timing? Should I go for the OV as soon as I see that my hatch won't be going down on 16? I find myself blocked for either the queen or the lings in that situation?
Just build everytime overlord before hatch. It wont changes a lot about pylon blocking, but you'll be safe and not be supply blocked.
My question is in regards to ZvP opening, when I open 14 pool -> hatch, when should I be making the 2nd overlord? The creation of the hatch + 4 lings + queen + overlord can't seem to flow smoothly for me.
Should I prefer delaying the hatch to make the OV on 16 then hatch? Should I drop the hatch on 16, skip the 16 drone so I can make 4lings + queen then the OV? ( I assume making the 4 lings ASAP is critical to hold a commited cannon rush or even just a pylon+cannon on the third )
If my hatch get's delayed by pylon how does that effect my OV timing? Should I go for the OV as soon as I see that my hatch won't be going down on 16? I find myself blocked for either the queen or the lings in that situation?
Just build everytime overlord before hatch. It wont changes a lot about pylon blocking, but you'll be safe and not be supply blocked.
Yeah, but if I do that and I DO get my hatch down on 16 the OV comes in way too early and I sacrificed a drone for nothing.
Just read your update on Immortal/Sentry. Dude. How are you not a blue poster yet? Your posts are so good that I literally switched to Zerg to take advantage of them. Thank you so much for all the hard work you put into these guides, and continue to put into them months after they first come out. You may not be the best player on TL, but I think you're the best guide-writer on here. Even CecilSunkure, my Toss idol when I played Toss doesn't come close, even though he's probably higher ranked.
On August 15 2012 20:17 Salivanth wrote: Just read your update on Immortal/Sentry. Dude. How are you not a blue poster yet? Your posts are so good that I literally switched to Zerg to take advantage of them. Thank you so much for all the hard work you put into these guides, and continue to put into them months after they first come out. You may not be the best player on TL, but I think you're the best guide-writer on here. Even CecilSunkure, my Toss idol when I played Toss doesn't come close, even though he's probably higher ranked.
His build knowledge is insane. Team liquid needs to hire this guy for TLO, Haypro, Ret, Sheth, and Zenio so that he can do full-time build research for them.
Where is the update on dealing with immortal/sentry? Is it the bit in the "How to React" section? I could really use some help on combatting this particular build. x.x'
Check out the replay section (Fast third vs FFE) - there's 3 new replays specifically on Immortal/Sentry.
Thanks Orek and Salivanth for the kind words. My ban/warn history is not exactly perfect. I was banned recently for shitting on mac's in some weird thread where the OP said he didn't want anyone shitting on mac's, but why do no pro gamers use macs instead of pcs, and I guess some people didn't want to hear the reason pro gamers dont use macs is because....
This guide was featured in an episode of Pony Tales though, by NrGMonk's recommendation.
I don't think much of what I put in my guides is really anything higher level players don't know already. Literally, the skill difference between me and someone like Haypro is larger than a bronze and myself. Pros have a ridiculous amount of knowledge. I often will PM pros here on TL after I've watched a particularly interesting series with them in it, and they often will reply, and just drop huge knowledge bombs. I've talked to Nerchio, Darkforce, Slush, etc...most of these pro games have a shitton of mental mind games going on between the 2 players in the match, but the observers, casters, and audience are none the wiser about it. I'm sure you guys have heard idra or qxc talk about the game, they aren't particularly smarter than everyone else, it's that every pro gamer has that kind of knowledge going on.
Try PMing a pro zerg on here if you have questions about what he does in his games, I'm sure they'll respond to you if you are nice.
Wow. I knew that a pro would know far more than a mid-high master, but I didn't know it was THAT big, nor did I ever expect them to actually answer those questions! (Though frankly, it has never occurred to me to ask.) Would you say the same thing with the Idra/qxc situation goes on with your rank? Namely, every (or at least most) mid-high master Zergs have the knowledge to write a guide like yours, but you happen to put in the effort required to do so?
Yea, I imagine most 1.2k points Masters Zergs already know everything in my guides. There might be a few tips or tricks that may be new to them, but for the most part they already know all of this.
I imagine most other higher level zergs haven't written out a guide like this because they aren't full of themselves. Also, I don't think they care when people post bad information or shitty guides and then get praised by people simply because it's the only guide on X subject. That is kind of a pet peeve for me.
I lurked TL for a while, but I joined when I needed help to build a PC when I knew nothing about building PCs. I did a ton of research on the subject of how PC's work - I didn't have a big budget, I'm a very picky person, and I wanted the absolute best, because it was my own money. My way of giving back to the community, was writing a thorough guide on how to build a budget PC, and what you need to know to do so. I still think it's a great guide, and despite it being 2 years, it's still all relevant since PC hardware hasn't changed at all in the last 2 years (given how much bulldozer flopped and nvidia/intel stuff that's new is so goddamn expensive), but it was closed by a mod.
I then posted the ling/infestor guide, it's just a short guide that is not viable anymore, when I was in diamond, whatever. I posted the 6 pool guide as a response to Bad Habit's guide on how to 6 pool, because I felt that it was possible to beat any type of 6/7/8 pool with hatch first, it just wasn't proven yet. So I spent about a month of playing with masters volunteers to come up with ways to beat 6/7/8 pool of all types with hatch first.
I wrote this guide in response to someone who wrote a guide on how to go fast third vs FFE, but his guide was complete shit, and he was a total moron, yet people were praising his shitty guide and he was talking shit because he said 'why dont you write a better guide if you are so smart'. He's permabanned now, and people started to post in his thread saying how shitty his guide was after the first page. I eventually totally overhauled this guide to a complete zvp guide, and then I wrote a zvt guide in similar fashion because zvt is pretty easy and simple, imo, in terms of understanding.
Zvz metagame changes way too quickly, so I have written a few guides on it but they get outdated very quickly, unfortunately. that, and the fact that there are 3 viable openings in zvz make just the early game itself incredibly difficult to cover in a well-formatted manner. I'll probably write a guide on early game zvz, and then update it, like this one, to all of zvz.
When you take your gas is a big deal. If you notice you bank gas in your games, chances are you are taking gas too quickly, or too much gas. The standard gas timing tends to be 6:00, or right around when third finishes. (...) If you like to go by supply, that means 2 gas is roughly taken around 50 supply.
might be a typo or so, because I've never ever seen any Zerg hitting 50 supply @ 6:00. Stephano normally hits 40 or so, and I normally hit 36. (actually I have no idea why, because I just can't find the difference) So I don't think it's possible to reach 50 supply @6:00. And getting 2 gases at 50 supply might be a little bit late i think. Because Lair, Speed, and attack will be delayed quite a bit. So I think it should either be 3 gas at 50 supply or 2 gas at 40 supply. Correct me if I'm wrong. I'm only low Masters.
^ No, most Zergs hit a little above 40 around 6:00. But you should be around 50 by 6:30. Zergs these days seem to be erring on even later gas, so if anything, i'd edit it to say 2xgas at 6:30, than 6:00. It's just a rough estimate though for supply, but it's still true - around 50 supply, 2xgas is taken total (3 gas later, 1 gas earlier than another later, 2 gas around a little before or after then, etc).
2 gases at 50 supply these days is actually very standard. Seems most zergs take 2-3 gases at 6:30 or later. I've seen some get 2, some get 3. Your lair, speed, should not be too late, and for the most part, Zerg isn't relying on tech to hold toss pressures, as lair won't be done for most gateway pressures, and those that do hit a bit later, you won't have upgrades or tech out in time likely. Blink all-in, stargate gateway, warp gate pressure, etc, you wont have roach speed if not lair against these pressures most of the time. Hold initially with roach/ling, and when roach speed and ling speed finish up, you are in a position to push back, but you already were if you held the pressure. Econ > tech.
Here are 4 replays of me playing the same person doing the same build (his version of the immortal/sentry all-in) and me failing to defend it with roach/ling. I have read and believe I understand the way this guide explains to deal with the all-in using the method I am attempting to use, but it's still not even close. I am hoping someone can shed some light for me on what I did wrong. I understand my macro is not gosu x2 level, but I really don't think that is the problem, as my opponent's macro isn't pinpoint accurate either. We are both in platinum league, if that matters any. Any advice would really be appreciated.
Now, here is one replay of me attempting to do the muta/base race style with great success. I believe my opponent misclicked his army while attacking my natural, but I don't think it would have made much of a difference. (Also, I forgot ling speed. ;D)
I would really prefer to be able to hold this all-in without base racing, as I feel it is very gimmicky (especially with mutas, as if I can't reliably scout the robo, it's not a reliable way to handle the all-in or other all-ins that may be coming my way). On that note, any help that anyone can give me to holding the immortal/sentry all-in in a "straight-up" manner with roach/ling is greatly appreciated. I know my macro needs to be good, I know I need to waste forcefields, I know I need to flank. I feel like I'm doing these things reasonably well for my skill level, and am still getting steamrolled when I respond in this manner. Please halp.
^ Base racing with mutas is not at all gimmicky, you WANT to base race with mutas, that's the entire point of their existence - if the opponent is dumb enough to move out before he has a shitton of cannons in his bases, and AOE like storm along with a ton of stalkers, you WILL win the base race (assuming you didnt lose too many mutas to something like a clutch storm or blink). What's gimmicky is someone pushing out against mutas (like immortal/sentry all-ins...), not mutas owning in a base trade.
Whenever you go mutas, you WANT a base trade. Most Toss these days are not that stupid, so they know when they see mutas, they need to hunker down, get lots of cannons and storm before pushing out. What mutas are bad at, is fighting straight up. So if you go like 200/200 pure 3/3 muta, if a Toss just turtles forever, he will actually win, even if you deny any base his army isn't at, he'll slowly eventually take a fourth, even if you kill his third in the process, and he'll keep adding onto a large 3/3/3 archon/mothership/storm army that will just annihilate even pure broodlord armies and he can win the game. Here's a game I played where this was a problem:
I think that the muta method is a great way to play against robo openers, as if you went roach tech against a Toss who goes robo expand, then it just kind of sucks that you invested in roach speed and a roach warren against a Toss going for a macro game where you will need broodlords asap vs that eventual colossus 3 base push. But mutas, are strong against both.
It is nice knowing how to hold in both ways though - sometimes toss can do something like a 4 gate +1 which will force roaches and might force you not to get spire in time against his robo follow-up, at which point you'll need to rely on roach/ling.
Lair takes a while to make, so you have plenty of time to get that spire against a robo all-in, it should be obvious when toss hasn't taken a third or pushed out by the time your lair is done, if not seeing his sentry count at the natural with the overlord you keep by his gas that is basically set on patrol command to keep peeking into the nat.
I'll check out the first replay you posted in a bit. But if mutas worked for you, don't need to fix what isn't broken right? Mind saying which of the 4 replays you think would be best to watch or something? And yikes, forgetting ling speed? You should always hotkey your spawning pool so you can get speed as soon as you have 100 (just set it to something far off like 6, I use 6 for my evolution chambers and spires for upgrades, but i'll add key tech structures to it too, like pool for speed since you start speed before even getting evos usually, and you can just tab through or just remove the pool later, add infestation pit later, etc).
I understand that, by going mutas, base trading is ideal. What I'm saying is that I would prefer to not rely on a strategy that plans on base trading. For reasons that you mentioned, mutas won't always be a guaranteed option, based on what the protoss does up to the point that I have lair (i.e., if I have to make roaches). I'd like to be able to hold it with roach/ling. If it's not possible, I think that kind of sucks.
If you are only going to watch one replay, I'd recommend the fourth one in the set of four (on CK). The standalone replay is where I did go mutas. Also, I rarely if ever forget ling speed. I had just decided to attempt a rudimentary version of the fast muta strategy you outline here in the guide one game, and had never practiced it before, which led me to make that silly mistake. The game still turned out in my favor, though.
Added a rep in 1 base, on cannon rushes. I felt the Naniwa vs Nestea replay wasn't a great example because Naniwa doesn't really dedicate that much into it.
On August 24 2012 14:42 Belial88 wrote: Added a rep in 1 base, on cannon rushes. I felt the Naniwa vs Nestea replay wasn't a great example because Naniwa doesn't really dedicate that much into it.
Thankyou for the bump instead of just editing the main post. I dont know what I would do if I forgot about this thread.
^ I update this guide at least once a week, and add about 2-3 new reps a week. This and the ZvT guide. I'll just bump it instead of editing silently. I thought bumping it like that would be bad.
But seriously it'd be bumped like every other day if I bumped whenever I updated this guide. I'll just bump it when major updates happen, something I think is special, or just after a bunch of updates.
So yesterday I checked my replays against your benchmarks and found out a couple of things. First, my macro was worse than even I had thought, which was in itself sort of impressive. My best ZvP with quick 3 base was 34 Drones at the 8 minute mark, wih 54 supply. Second was that in all of my replays I for some reason, rather than finding out what my opponent was doing I kept building Lings, for an attack that literally never comes. So yeah, that was useful to know.
I'm posting though to thank you for the guide, after a bit of practice at spending Larva, injecting and not over building Overlords I just managed to hit 79 supply at the 8 minute mark with 67 Drones (5 on the way) and was maxed at 11:28 which whilst still not the best in the world of course is a full 2 minutes faster than I've ever reached in the past. Also had a better APM than average too with something to focus on. I just need to get those skills in to ladder now as I was playing vs ai in tests, though I did build 4 Lings and take towers and position Overlords, but it is really cool to see it's possible for even a fairly bad Gold leauger like me to hit that kinda benchmark with a bit of practice.
I don't have any questions or anything really to add, I just wanted to thank you for the OP and the benchmarks especially.
On August 24 2012 16:14 Belial88 wrote: ^ I update this guide at least once a week, and add about 2-3 new reps a week. This and the ZvT guide. I'll just bump it instead of editing silently. I thought bumping it like that would be bad.
But seriously it'd be bumped like every other day if I bumped whenever I updated this guide. I'll just bump it when major updates happen, something I think is special, or just after a bunch of updates.
I was being sarcastic. Your over-bumping is self-indulgent and annoying.
I don't really bump it, even after major updates. I don't know why you'd say that, you must be confused on other people bumping this thread. Given the recent posts, like yours, i was going to start to though.
Well I for one genuinely think you should. I really don't know what I'd do if I forgot about this thread No sarcasm! I'd love to get updates on revisions, rather than have to re-skim the guide every few days.
Glad to hear. If you have any suggestions, think the guide is missing anything on reps or a particular attack, I'll be sure to load it up next time i face it on ladder (done correctly) or have someone do it to me.
I was amazed that he indeed fit your description, hitting 78 supply with 61 drones at exactly 8:00 mark. But he was quite ling heavy. Wonder if it would have worked out that way had he made roaches instead of lings. (Though I understand that he had to use some drones for spores) And he was very drone-light throughout the game. Only 64 drones for the rest of the game and I frankly think he was quite sloppy. He was lucky that Jangbi was even sloppier.
How do you deal with blink stalker push ? can u play muta mass spine trade base (or just a shit load of lings ?) or are you forced to get a roach warren ? Do you have critical timing to spot in order to recognize this build ? I have seen a lot of toss do differents things on this build order, like forge activate very early or very late with a lot of chronoboost, 4 gaz at 6m30 or 4 gaz a 7min, zealot push before the blink stalker push, or not ...
oh and thx a lot for this guide btw, it's really interesting =)
How do you deal with blink stalker push ? can u play muta mass spine trade base (or just a shit load of lings ?) or are you forced to get a roach warren ?
You can't get lair tech out against blink push. You'll need roach/ling.
You recognize blink stalkers because he's not making sentries, but rather stalkers, you may see a twilight council, you see lots of stalkers push out around 8-9:00... I mean it should look similar to a 6/7/8 gate but he already has a couple stalkers built up and it's a little later, whereas most Toss are building up sentries. 4 gases, lots and lots of chrono on the forge from early on, and the cybercore. You don't particularly need to know it's a blink push, you just see he's doing a 4 gas warp gate push, and you just get lots of roach/ling against that.
If they do a zealot push before it, that's a totally separate issue. You don't ask "how to deal with reactor hellion into mech", you ask "how to deal with reactor hellion" and "how to deal with mech". Similiarly, you need to learn how to hold the zealot push (before 6:30, use lings made when you see him push out with zealots, always know how many zealots he pushes out with, and make sure he grabs gas at his natural before 6:30 or else get a 6:30 warren and a few roaches), and then blink after that (which shoudl be easy to hold if you hold the zealot pressure).
when the toss use a zealot push before the blink stalker, the blink stalker is delayed and u can't counter zealot +1 with just ling so i think there is a link but nevermind.
one more question, you say to go roach ling against 4 gas, but sentry immo push also use 4 gas and i prefer to deal with it with muta. macro/tech can also be with 4 gas and i prefer to play mutas against it as well so should i just recognize the blink stalker with the fact that protoss doesn't make early sentry ?
^ You should really provide a replay. It sounds like all you want to hear is "your right, you played perfectly, +1 zealot push into blink stalker is totally OP and theres' no way to scout for it"
I already kind of answered your question. No, you counter the +1 zealots with an early roach warren that you made because you saw Toss is gasless. If he took gas, then his +1 zealots should be perfectly counterable with just lings and queens. They aren't going to be warp gate zealots, you should see them moving across the map. These are 2 totally separate things you are talking about, it sounds like you are jsut having trouble with +1 zealot pressure. To which, you have to make a 6:30 roach warren if you see toss has no gas at his nat, and you have to sac an overlord. If you react like it's a 6 gate +1 when instead it's a 4 gate +1 zealot pressure or just double gateway upgraded zealots, you won't have the econ or tech to handle a follow-up like blink stalkers.
As long as you only made like 3-5 roaches, and reactively made units if toss continues pressure, any sort of +1 zealot push opening, 4 gate +1, et cetera, should put toss so far behind that you'll easily crush whatever push he does. Not to mention roach/ling holds any sort of 2 base toss all-in. Colossus? roach/ling. Blink? roach/ling. Immortal/sentry? roach/ling.
one more question, you say to go roach ling against 4 gas, but sentry immo push also use 4 gas and i prefer to deal with it with muta. macro/tech can also be with 4 gas and i prefer to play mutas against it as well so should i just recognize the blink stalker with the fact that protoss doesn't make early sentry ?
Sentry/imortal push is tech. It's a late push that isn't coming anytime soon. That's why people say in order to beat it, you want to be greedy. As in, you don't need a 6:30 roach warren, you won't need to not make lair (ie because if it's a 6 gate +1, you'll die if you go lair before speed basically). You won't need to pump lings early on...
Yea, you should recognize the difference between blink and immortal sentry based on how they take their gas (immortal/sentry either takes 2xgas at around 6:00, or 1 gas earlier/later but basically a similar timing, ie maybe a gas at 5:45 and the 2nd one is taken before 6:30. most usually take 2xgas at ~6:00 though) and if they are making lots of sentries or not. A blink push isn't going to make more than a single sentry that was made just to be safe against early all-ins. If he makes 2 stalkers from his gateway, that should raise questions. It should also be obvious when he pushes out with a bunch of stalkers at around 8:30-9:00, whereas immortal/sentry hasn't pushed out yet. I mean against blink stalker, you should be forced to make roach/ling because you see a move out, whereas immortal/sentry should be pretty obvious.
You really gotta provide a rep though. This is all highly theoretical talk, but in all of my experience I have never seen a game where I didn't know what to do.
I mean even immortal/sentry and blink have the same response to beat - drone hard and tech normally up until 8:00, then pump roach/ling. If you are going quick mutas, you'll just have to accept that blink all-ins will be some trouble to you, and something you should learn to identify. I like going quick mutas too against robo openings, but you can still make a roach warren. I will still make a roach warren most of the time if I can't tell exactly what Toss is doing.
Think of doing things like cutting out the roach waren, getting a quicker lair and gases, no macro hatch, no evo chamber, as cutting corners against immortal/sentry, but you can really go the muta route even if you made a roach warren, evo, et cetera.
thx for your answer, i'll try to post a rep next time i have a question, it's sometimes hard to explain myself in english for such technical things. I don't have a big problem with zealot +1 btw, just with all different kind of build leading to blink stalker (since i like to play muta and need to know if i can go lair and take all my gas, way before 9min). I tend to go gas lair and mutas as soon as i see 4 gas but it's maybe not the good thing to do. anyway thx a lot, again, and i'll come back with some replay if i still have issues with this oh and i don't think there is imba build, i just need to learn how to recognize them and this is where u helped me
The only thing I did not agree with was saying you can not be an econ. major and not be a conservative lol! I'd be inclined to say that you can not be an American economics major and not be a conservative, that I could believe. I met a lot of non republician economic majors and many "right wing" econ majors up here in Canada would be considered decomocrats in the US. =).
But that is another thread for another time, good guide. Thanks!
Lots of economists tend not to be party affiliated, even in america. Personally, I'd just say they are hardcore republicans who won't fess up to it, same thing (libertarian, republican, same thing, really, they just don't have any pride). But they will be conservative, even those up there in canada. They certainly won't be socialists. And people can be conservative and align themselves with a liberal party, everyone's a 1-issue voter when you get down to it, and for example, i don't agree with any of the religious or social stuff the republican party says at all, but my 1 issue happens to be what they agree with more than the democrats. if that makes sense. The majority of economists will be fiscally conservative, basically, is what I'm saying. But yea, canada and other countries tend to have a different center. What's conservative in many countries would be liberal in the US, what's conservative in the US is hardcore in other countries, etc. But they'll lean one way more than the other.
Glad you like the guide. It's no secret that republicans tend to be pro leet masters on sc2
Amazing guide. I was looking for something like this before, but never managed to find a guide that was this good. I think I've finally started figuring protoss out. Just a few questions though; you said if I hadn't seen anything at 10:00, I'd have to throw down spores. Don't you mean spines?
And also, in case of a blink-all in. Shouldn't I almost exclusively be making lings? Or would you prefer mixing in some roaches as well? Also, should I be getting either a range or melee upgrade, or none at all?
Oh, one last thing. How should I react to a ~12:00 min push with colossi in it? If I went ling-heavy, I'm pretty dead aren't I? What's the best army to engage this? Is it roach-corruptor? And do I have enough time to get corruptors out after I spot the first colossus, but do not have my spire yet?
Amazing guide. I was looking for something like this before, but never managed to find a guide that was this good. I think I've finally started figuring protoss out. Just a few questions though; you said if I hadn't seen anything at 10:00, I'd have to throw down spores. Don't you mean spines?
No. Any sort of gateway pushed should have pushed out by 10:00. If he hasn't, you should be able to make enough roach/ling to hold. You also can get spines up reactively when he pushes out, whereas spores you can't exactly start making them when DTs first get in your base or 4 void rays show up at your third.
And also, in case of a blink-all in. Shouldn't I almost exclusively be making lings? Or would you prefer mixing in some roaches as well? Also, should I be getting either a range or melee upgrade, or none at all?
I cover this in the guide, but you always want to be heavier on lings. As the game goes on and toss is clearly 2 base turtling, you go heaviner on roaches.
All 3 upgrades are viable. They each have their own benefits.
Oh, one last thing. How should I react to a ~12:00 min push with colossi in it? If I went ling-heavy, I'm pretty dead aren't I? What's the best army to engage this? Is it roach-corruptor? And do I have enough time to get corruptors out after I spot the first colossus, but do not have my spire yet?
Not really, as long as you have enough roach/ling you should hold it. You can get corruptors if you want, but making just enough roach/ling is really the best way to hold it. I wouldn't really recommend getting spire/corruptors if you scout it as late as seeing his colossus (maybe if you spot the support bay being made), I would only add spire after maxing out on roach/ling and Toss still hasn't pushed out.
You need to set up a flank. Shouldn't be hard to do since he'll be all-inning and pushing toward you. You know he's coming, just set up your army for a flank before hand.
I always overreact with too many roach/lings when toss goes for a gasless attack on my third. So after defending, I just saturate all bases grab a 4th (needlessly) and reinforce the units I made.
At this point, I have a lot of money so I thought making a lot of units to try and kill him would be a good idea. (opened 6:00 3 gas)
This guide is fantastic. I'm currently at the top of diamond league, but it's almost entirely from my ZvT and ZvZ. Those match-ups usually get me wins against low masters players, but I can never advance to masters because my ZvP is such shit (plus the fact that I don't win every ZvT/Z )
I was wondering why the fuck I couldn't win ZvP and after reading thorough this guide and studying some replays, I realized that I have the early game macro of a gold league player. I never doubted my macro to be the issue because, in general, my mid/late game macro is solid. I didn't realize how important the early game macro is. I'm regularly 45-50 supply at 8 minutes. FUCKING TERRIBLE! It won't be hard to fix since I never really put that much care into my early game macro, and now that I know what to do, ZvP should get much easier..... I hope :D
someone probably already mentioned it but that one game in the VOD section DRG vs Squirtle.. its not a 4 gate +1, squirtle actually forgot the +1 upgrade so, kinda of a bad example bcs im always failing against this strat its super strong... 4 gate +1 into 7 gate immortal..
^ Yes, you're right, but DRG's reaction is what is important, and DRG reacts perfectly. If Squirtle remembered +1, DRG would have held just as easily.
I also included a replay of myself holding a 4 gate +1 (where the guy remembers +1), just to make it clear though. I would imagine Squirtle's 4 gate without +1 is scarier than a random ladder masters 4gate +1 anyways.
If you hold a 4gate +1 while adequately having had started lair and droned a bit during the attack as you realize it's 4 gates and not 6/7/8 gates, you should have infestors out in time when Toss pushes and hold easily. And you should really only make infestors against a 2 base toss after you have about 150+ supply in roach/ling (making them too early can result in you dying, but earlier infestors means more energy, so it's a balancing act).
Yay. Belial is finally back. Your contribution to the community is much appreciated. This time, I hope you don't go overly aggressive towards people you disagree with. Welcome back.
You don't completely cover this in the guide...what do I do against an in-base proxy 2 gate (scouted)?
I went 15 pool and figured I'd build 3 spines and pool larvae for lings + a queen...his proxy is so close that he gets to my mineral line well before the crawlers finish. It's the in-base one on Daybreak.
If I scout it, what do I do? Pull drones to force a cancel?
Edit: I feel like such a noob; dying to the silliest cheese in the book, but I honestly don't know what I should've done and it's clearly a hole in my play, especially in a tournament setting (like this game).
If you just overlord scout, it can be deceiving when you see an empty base with maybe no gas being taken, or no wall-in to the ramp at his main, and you haven't seen yet that he has no pylon on the low ground at his natural. While seeing no pylon on the low ground (after seeing he didn't wall-in at ramp) is a dead giveaway to double 10/10 proxy gate, on some maps you just may not see that in time. It's okay, your 14 pool build will be fine, even if you made a hatch at 16.
As soon as your pool pops, make 2 spines in 'uncomfortable', choked up locations like by the geysers, next to your pool that's by a wall/object, or in your mineral line. If you only make 1, Toss can kill it and you will be screwed. Also, grab your gas asap - get ling speed, and you will be able to crush any remaining forces and just run into his base and autowin. I strongly recommend against roaches - he can go for core, stalkers, and torture your slow roaches, especially if the gates aren't proxied.
You should really cancel your hatch if you planted one, but if he's being an idiot and attacks it, let him attack it and just cancel it at the absolute last second. Even if he kills it, you should be fine. However, you don't NEED to cancel it, you will have enough resources to make enough to hold and make mass speedlings and autowin, and if he kills it you can still end up ahead if you make those 2 spines and keep your drones alive and get ling speed. It's up to you - I'd say it's safer to cancel, it's greedier to keep it but really you'll be ahead whether he kills it or doesn't. Or, you could keep or cancel it based on if he attacks it or not.
If he does not proxy the 2 gate, you can definitely keep your hatchery. It will take damage, but you will uproot your spines and put them down in the nat. There are macro oriented 2 gate openings where they expand right afterwards - just make sure he expands (he won't have stalkers) instead of going core, gas, into 4 gate follow-up, and take your third and threaten with your lings and just be so far ahead of him...
If he's proxied it you might be a little tight, just do a bit of drone micro. building 3 spines is too much, just make 2. Or, if you go for 3, dont engage with the drones, just run away, let him focus a spine down, then cancel it and now you have 2 spines. Lings and queen will be up soon so that you can push him back.
Dont pull drones to force a cancel. you might need to do mineral walking against the initial zealots until the spines finish, which may need to be fancier the later you realize it's proxy 2 gate, but you should still hold.
Yeah I saw that quoted text, but I did what you said and died O.o
Like if you watch the replay and see any major mistakes let me know, but I'm confused. I did drone micro, I mineral walked, I even used the "drone flower of death" thing and he just killed too many drones before the spine finished. Running away is an OK idea, but I think he would've been able to kill most of the spines/a lot of drones either way. I'm just not sure where I went wrong, but it's clear I messed up somewhere. It was then 2 spines and a queen and ~7 drones against 7 or 8 zealots, yeah that fight didn't go well.
I think my pool timing was basically a 15 pool, and I did drone scout so I was a drone behind. The guy I played against is a pro, and says he's used it successfully against other pros (only on Daybreak), but I'm just not sure. Zealots just chew through drones.
Edit: watching the replay now, I'll edit in after
Hmmm... I only made 2 lings when pool finished, am I supposed to get 15 overlord? I feel that the extra minerals you get from not dronescouting is what enables you to survive. Also, the proxy is about 10 seconds further away than the one in my game. I guess 2 spines + ling/queen is better than 3 spines + queen. Your drone engage was also in a better position, and you got a full surround with drone/ling. I guess I'll just have to play against it again.
You lost too many drones micro'ing around early on. Against zealots like that you should avoid engaging. Drones fight pretty well against slowlings. They do terribly against zealots. You just want to stay away from the zealots, using mineral walking to sort of run away. Then he has a choice of either chase the drones, or attack the morphing spines. If he attacks the spines, then you attack with the drones. If he attacks the drones, you run away and let the spines finish up. 3 spines is a little much, I'd rather go with 2, as now it's only 9 probes to 12 drones with 3 spines made.
You make 2 spines because your going to be forced to cancel one of them against proxy 2 gate. But at least the other one will finish. I personally like to have 2 spines so I can just counterattack right away when lings first pop but whatever. So maybe make 3 spines, but cancel the 3rd if it's about to finish.
You shouldn't try to drone flower against zealots though. All the fancy micro you try to do results in you losing 7 drones which just put you too far behind to win.
If you know you are being proxy 2 gated, your pool is better placed blocking off the mineral line, so the spines can better cover it (they should finish before he can focus it down anyways).
Ok, thanks for the tips. I was able to hold a proxy 2 gate (one that's literally right beside the mineral line) with 12 drones left by building 2 spines and drone microing the way you suggested. I played 4 games against a friend proxy 2gating and won all of them.
Helped a lot, thanks! Zealots do a shitton of DPS though, it's a little scary how quickly they'll kill drones/queen/spine O.o
I might be participating in a platinum league tournament soon. So, my question is, as a Platinum, what are the better all-ins I can do in this matchup against other Platinums? There's about a dozen all-ins you've provided, and I'm not sure which ones are the strongest.
^ They are all strong. In my personal experience, the ling/baneling all-in has worked the most, because it plays a lot off Toss scouting the third and then thinking they are safe. Which all-in I go for tends to be based on how Toss scouts, but it seems every toss I've played, and all-inned, checks for the third, and you can't really deny a probe mineral walking towards the third. Doing a roach or bane all-in can be great though if you take your third first, and then not take your natural.
Just be careful of what map you do it on, as I've facepalmed once before when I tried to do it on cloud kingdom and saw that a forge/gateway/cyber can make a tight wall-off and the 3 hatch ling/bane all-in only gets like 5 banelings out, which is exactly enough to take down a pylon, and if you wait any longer to get enough banes to break through a bigger building, the sentry is out. Although I've had success a couple times against cannon rushes/ramp blocks with just 10 banelings, getting forcefielded by a single sentry, then just busting down the forge and roflstomping.
Generally, cannons beat roaches, sentries beat banelings. So that's something to keep in mind. Once I played a guy who ramp blocked me, knew I'd be going for an all-in because I had no choice, so he made 3 cannons, but 3 cannons doesn't stop a ling/baneling all-in (it stops a roach all-in cold though). But my overlord was over his natural, I made a roach warren, he made 3 cannons, I cancelled the roach warren when I saw that and decided to go for banes instead. He was a maphacker though, so he made a gateway wall, but the map was condemned ridge and i just needed to bust one pylon to get in and flood his base with lings and just run into his main since cannons don't stop that, only sentries.
But a roach all-in, 2 cannons stop it, sentries and stalkers help against it. Against a baneling all-in, you need sentries or you die. So there's all sorts of room for error there on Toss' part - like he could be being greedy and throw down his tech instead of make that sentry, or he could make a stalker to do pressure/kill overlords (so maybe go for a lign/bane all-in, and float an overlord inside his natural/main right when his core is about to finish so Toss will go for a stalker instead of sentry). And he could just miss the forcefield. And like I said, a couple times I did it when they had sentries but the game was weird because it started with a cannon rush/ramp block and I just had a ton of banes and even though he had sentries, he had spent so much on cannons that he really couldn't have enough in time.
In short to answer your question I like the 3 hatch ling/bane all-in. And when i've done roach all-ins, I've had more success with the IM style ~35 roach warren, speedling all-in instead of the 3-5 roaches from a 20 roach warren, slowling all-in that's more 'common' or leenock styled. Map is an important factor (tighter choke is worse for banes since it means less weak buildings as part of the wall like zealot, pylon).
On October 31 2012 13:20 Belial88 wrote: ^ They are all strong. In my personal experience, the ling/baneling all-in has worked the most, because it plays a lot off Toss scouting the third and then thinking they are safe. Which all-in I go for tends to be based on how Toss scouts, but it seems every toss I've played, and all-inned, checks for the third, and you can't really deny a probe mineral walking towards the third. Doing a roach or bane all-in can be great though if you take your third first, and then not take your natural.
Just be careful of what map you do it on, as I've facepalmed once before when I tried to do it on cloud kingdom and saw that a forge/gateway/cyber can make a tight wall-off and the 3 hatch ling/bane all-in only gets like 5 banelings out, which is exactly enough to take down a pylon, and if you wait any longer to get enough banes to break through a bigger building, the sentry is out. Although I've had success a couple times against cannon rushes/ramp blocks with just 10 banelings, getting forcefielded by a single sentry, then just busting down the forge and roflstomping.
Generally, cannons beat roaches, sentries beat banelings. So that's something to keep in mind. Once I played a guy who ramp blocked me, knew I'd be going for an all-in because I had no choice, so he made 3 cannons, but 3 cannons doesn't stop a ling/baneling all-in (it stops a roach all-in cold though). But my overlord was over his natural, I made a roach warren, he made 3 cannons, I cancelled the roach warren when I saw that and decided to go for banes instead. He was a maphacker though, so he made a gateway wall, but the map was condemned ridge and i just needed to bust one pylon to get in and flood his base with lings and just run into his main since cannons don't stop that, only sentries.
But a roach all-in, 2 cannons stop it, sentries and stalkers help against it. Against a baneling all-in, you need sentries or you die. So there's all sorts of room for error there on Toss' part - like he could be being greedy and throw down his tech instead of make that sentry, or he could make a stalker to do pressure/kill overlords (so maybe go for a lign/bane all-in, and float an overlord inside his natural/main right when his core is about to finish so Toss will go for a stalker instead of sentry). And he could just miss the forcefield. And like I said, a couple times I did it when they had sentries but the game was weird because it started with a cannon rush/ramp block and I just had a ton of banes and even though he had sentries, he had spent so much on cannons that he really couldn't have enough in time.
In short to answer your question I like the 3 hatch ling/bane all-in. And when i've done roach all-ins, I've had more success with the IM style ~35 roach warren, speedling all-in instead of the 3-5 roaches from a 20 roach warren, slowling all-in that's more 'common' or leenock styled. Map is an important factor (tighter choke is worse for banes since it means less weak buildings as part of the wall like zealot, pylon).
you can do a roach all-in in the same style as the 3 hatch baneling all in. 6-7 roaches and 20 speedlings with constant speedling reinforcement. Hit 7:30-7:40ish (depends on map. you can cut a roach or two to get it a little faster, and it usually still works) 3 cannons does not hold it. neither does 3 sentries. with the roaches and speedlings you have enough dps to just rip down the forge and any pylons in the wall and just overrun him with zerglings. i don't know if it's better or worse than baneling, but it still basically automatically kills protoss who don't know it is coming (which is every protoss, because they stop scouting after seeing a hatch at the 3rd).
On ZvP ladder close positions I was planning to try the two base infestor ling timing attack.
My question is, is it better to skip double evo +1+1 for earlier lair and earlier infestors?
It means my attack will come earlier but my lings will have no upgrades.. I'd like to ask which is better in the current metagame (NF + robo + gateways).
it can make for a strong timing attack to hit right when upgrades finish, but it might cost you too much. I dont know about timing, maybe 1/1 will come at sime time you have your 5 infestors at his front of base, maybe it will force you to wait, youll have to test it out. Really depends on how the timings work out, 1/1 is a strong upgrade advantage if you can hit right when they kick in and not be delayed.
i dont think there's a right or wrong on that one, just more about your own style and preference. It's more important that you do it well, than if you get 1/1 or not.
I was told today that the immortal/sentry all-in can be followed up by a 3-base Colossus push if the Protoss keeps 2 Immortals and the Warp Prism alive for prism harass. Is this correct? If not, how do you end the game after successfully holding it off if the Protoss doesn't gg out? Note: This assumes you held it off without losing a lot of drones / your third or anything, only army units.
With the recent infestor nerf, i feel like zerg cant go anymore to BL/infestor/corruptor without having taken a huge lead in the midgame. Playing turtle zerg is not possible anymore, and doing nice transitions while pressuring the toss seems necessary.
My friend, I can consistently hit 70 supply with a 3 base build with 3 gas at 6:10. Its really easy. Does that mean my macro is code A quality? LOL. I am working on making the macro better until maxed out though.
Btw, I think in both WOL and HOTS, its much better to go double upgrades into ling/bane/ultra vs the Colossus mass gateway attack at 15 minutes.