On September 05 2012 01:29 SHODAN wrote: I'll try to explain the strategy behind triple OC tanks, and economy-heavy builds in general. Early tanks allow you to play as economy-heavy as possible. It lets you take a faster 3rd OC, faster 3rd and 4th gas, and faster upgrades (double engi bay at 6:50). Your early tank count then allows you to land your third much faster (9:30 mark), take the 5th and 6th gas faster, start 2-2 much faster. By 11:00 you will have 1 viking and 2 medivacs. After the viking has finished clearing nearby overlords, you use it as a spotter to track the location of his army. 2 medivacs and 16 supply of bio can freely clear up creep, then lift-off when you're in danger. This also opens up the threat of drops, whether you choose to commit to them or not. At the 13:30 mark, you will be at 180 supply, 6-7 tanks, 4 medivacs, 2-2 finished, 10 naked raxx, 1 with a techlab (only add reactors after you start 3-3). You attack at 2-2, shut down Z's fourth if it exists, trade well with his army and keep trading... I don't rely on gimmicky shit to get an early advantage over zerg. I try to outplay Z in the midgame with core units, drops, and tank lines. Z can safely make 75 drones no matter what opening you do. What really matters is gas, and Z can never have more gas than you with an economy-heavy build.
I know perfectly well both the reasoning and the benchmarks since, as stated some pages ago, I used to play this kind of build. Of course Zerg will reach 75 drones, that's not the point—the question is how much time until he gets that, and how much he has to dedicate to his defence. With triple OC tanks answers are “as fast as possible” and “zero,” which is not good since both creep and Zerg economy are exponential. You might not like Hellions/Banshees openings but at least they don't allow Zergs to reach 3-bases saturation uncontested. Zergs have to be careful against Hellion openings, they have to get Spores against Banshees follows-ups, they can't just sit at the edge of the creep with their extra Queens and allocate all their energy to new Creep Tumors. There's no doubt you hit a powerful 200 2/2 max army by the 14' mark, but it doesn't matter since creep prevents you from using it to its full potency, thus allowing the Zerg to welcome you with unchallenged BLs or 5/3 Ultralisks. Not saying this build will auto-lose but Zergs should definitely have the advantage against it should they go standard macro.
The question for me is how long can I keep zerg in the dark about what I'm doing. If Z sees my natural walled off with 3 raxx, Z must prepare at least for the possibility of bio pressure. Z will most likely take gas, upgrade zergling speed, save larvae for zerglings, and try to sack an overlord. If I feel Z is playing too greedy, I can fake pressure with marines & a bunker at his third; even in small numbers, marines trade really well with slow lings. IM_Yoda faked a 3 tank push against Symbol earlier today and I think it too would work great with this build (G2 Ohana). The maps that work best for 3OC tank (Ohana, Entombed, Daybreak) are also perfect for 2rax pressure. How about 2rax pressure ---> QXC bunker ---> 3OC/tank? How about 3OC/tank ---> tank drop the ledge of Z's main on Antiga? And as I said, medivacs give you the mobility to control creep in the midgame.
As you mentioned the "goto" build in tvz right now is the 1 rax fe into reactor hellion into cloak banshee. Is it a must for terrans to cut scvs and marines to get both gases quickly and then build the depot and continue production or is it ok to first build the depot and continue production and then add the 2 gas? With the later version the first 2 hellions will pop out at 6:35, but i feel more safe against the initial 6 lings since you cut your marine production really long with the first version.
Another question. How many banshees and hellions do you build, when you want to transition into marine tank? Is there a "standard" number or do you do it when you think your pressure isn´t going to work anymore? I heard about 8 hellions and 3 banshees somewhere on reddit, what´s with this number?
And last. What is the normal way to transition into marine,tank,medivac? Do you build your CC, then double gas, then double ebay + tl on rax, then start all 3 upgrades, then adding additional barracks? When is the correct time to begin tank production inbetween them? Do you allways use the tl from the starport to transition? Can you advice with some good replays where the transition is nicely done?
Ver, you've mentioned that something which makes Korean Terrans so dangerous is their ability to all-in -- something that a few other notable posters have said as well. Do you have any nice all-in builds you could recommend to us?
Thank you for making this thread, the wealth of knowledge in such a small thread is something i'm sure we all appreciate.
I'm struggling vs protoss.
I always 1 rax FE, into a +1, stim 2 medivac push (no additional add-ons until I push out) I understand the difference between adding on 2 additional rax or going for my 3rd cc. I understand that if he goes collosus, you should have an opportunity to drop somewhere with the first push. HT, not so much.
What is the primary objective of this 'push out' ? To secure your third? To actually do damage? Don't I have to be the aggressor?
If I just go home and macro, I would say 99% of my games end in 1 way, 1 large battle and usually the toss steam rolls me (I'm mid diamond, my micro is at best average in large engagements). Even if I try to emp as much as I can, I just get destroyed with the additional warp ins of charge lots or any left over HT which turned into archons.
Rather than going into the analysis on why I get rolled late game, I'd rather ultilize my skills (decent multitasker and able to do multi- pronged attacks) with better mid-game agression.
Is there a build, which allows me to apply a much stronger mid-game pressure?
Is a 3 rax before expo pressure considered all in? What about the 3 hellion drop? (too gimmicky?) 1-1-1 with cloak banshee? (that's a 1-base all in technically right?)
Are there any decent 2-base allin? I'm looking for strong mid-game final blow. Since I lose late game all the time, if it's all-in that's not much of a worry for me.
I always lose late game and rather than dealing with it, I'd rather just avoid it. I know that's not ideal, but it is what it is.
On September 05 2012 22:51 RefusedAmen wrote: Thank you for making this thread, the wealth of knowledge in such a small thread is something i'm sure we all appreciate.
I'm struggling vs protoss.
I always 1 rax FE, into a +1, stim 2 medivac push (no additional add-ons until I push out)
You should make add-ons #2 and #3 when you make your Starport at the latest.
On September 05 2012 22:51 RefusedAmen wrote: What is the primary objective of this 'push out' ?
To keep the Protoss player modest, to prevent his third from going up too fast and to scout his army composition / tech path. Opportunities to end the game right there should never happen in an even game; whenever you deal heavy damage you can be sure the Protoss failed something or was too ambitious / used a bad build. Basically this push acts as a soft contain.
On September 05 2012 22:51 RefusedAmen wrote: Is a 3 rax before expo pressure considered all in?
It's not all-in, but like many pressure builds the game will often come down to whatever happened with your pressure move. You should move out at 5'20 with 10 Marines; pull 2 SCVs for standard pressure (to attempt to Bunker his expand), more if you want to commit. Your expand should be made around this time too, between 5'20 and 5'45. You will be able to kill the Nexus if the Protoss didn't play safe enough, i. e. Nexus before Core, or Robo before Gates #2 & #3, or too many Chronoboosts on Probes and not on Warpgate, etc., or alternatively if he went Sentries as his first units. If he goes Zealot Stalker Stalker with enough Chronoboosts to have a 3 Stalkers warp-in ready when you reach his natural, you will be thwarted. I play 3 rax expand a lot so if you're interested in more details (follow-ups, common threats, etc.), I can provide them.
On September 05 2012 22:51 RefusedAmen wrote: What about the 3 hellion drop? (too gimmicky?)
Again it comes down to how much damage you do with that Hellion drop (and Marines if you used them offensively too). Common follow-ups are 3-1-1 all-in, or simple expand, or double expand. Be careful if you expand, you want 1-2 Bunkers ready or you may die to a random proxy Pylon as you will only have Marines to defend Stalkers.
On September 05 2012 22:51 RefusedAmen wrote: 1-1-1 with cloak banshee? (that's a 1-base all in technically right?)
You can also play double expand behind it with 2-3 Tanks to defend while you tech bio, Taeja does this sometimes (e. g. Taeja vs Harstem on Metropolis @ ASUS ROG Summer).
On September 05 2012 22:51 RefusedAmen wrote: Are there any decent 2-base allin?
Well, there are 2-bases version of the Marines/Tanks/Banshees (with bio tech) all-in, or Marines/Thors/Banshees builds à la JJakji, but to be honest if Protoss scouts right I don't see them working reliably. Another option is to go standard 1 rax FE → 3 rax medivacs → 5 rax and to just pull SCVs at some point like Mvp does sometimes (e. g. Mvp vs Feast on Entombed Valley @ IEM Cologne), but again it should not work against a Protoss player properly scouting your lack of third and thus not trying to take a third base.
On September 05 2012 22:51 RefusedAmen wrote: If I just go home and macro, I would say 99% of my games end in 1 way, 1 large battle and usually the toss steam rolls me (I'm mid diamond, my micro is at best average in large engagements). Even if I try to emp as much as I can, I just get destroyed with the additional warp ins of charge lots or any left over HT which turned into archons.
I know you didn't want advice about late game tvp, but it's something you need to learn. It's kind of simple actually -- bio steamrolls gateway units so hard it's not even fair. The problem is colossus and storm -- if you let the protoss get off 1 good storm, you basically lose. Make sure you get cloak and about 10 ghosts -- if he's splitting templar, use snipe; if his templar are clumped then use EMP. If you don't have 3 vikings/colossus, you lose. In general, I like to have 4 medivac, 10 viking, and 10 ghost, since most protoss get 3 colossus and then switch to templar tech.
Also, in terms of when terran can be aggressive -- I think terran starts to have an advantage from the 10min, peaks around terran's 160 supply (when protoss starts to tech switch to templar), then starts to even out after that (as charge/blink/storm finishes). I like to be aggressive as much as possible before templar tech comes out, because once storm finishes, it's so so hard attacking into them.
On August 31 2012 14:08 Shankapotamus wrote: When I try to play mech TvZ, I keep getting counterattacked by his whole army, his roach army is so much faster than my mech and i can't catch up. I even tried blocking off paths with PFs but since he has his whole army 1a'ing all over the place, i have no idea what i can do to win. How do i kill the zerg without dying to nonstop counterattacks with his whole army?
You have three deterrents to stop them from doing this:
1) Banshees. If he doesn't make muta or corruptor he can't effectively kill them, meaning if he attacks he has to tank a lot of fire. This is why you keep making them like MVP has shown.
2) Sensor towers and scouting units/scans to know that he is doing this and seeing where his army is. Remember if you see his army on a map like Daybreak, meet it partway and force him to retreat, then just 1a after it, it will take him far too long to circle all the way around the map and counter.
3) Hellion counters. The most effective measure yet hardest to grasp/execute. Whenever they send out their roaches that means they don't have anything left at home to stop hellions from decimating all the drones. Here's a tourney game from the tespa csl lan as an example:
Although my opening this game was poor and I got far behind, I managed to make an enormous comeback using this same philosophy. Every time he moved out with roaches my hellions counterattacked, sometimes via runby and sometimes by drop.
On September 01 2012 22:59 Sp4cem4nSpiff wrote: Cool thread idea!!
How should you handle a base race situation? Should you keep attacking their base, or come back and try to salvage some of your own??
Too vague to really answer....depends so much on matchup/situation. There is a certain logic to base trades but it's tricky to learn if you don't play them regularly. In general though you want to commit enough to disable their production, then try to save your own if you have a lead.
On September 01 2012 23:11 Sp4cem4nSpiff wrote: Having a lot of trouble vs Ling/ Infestor....Any suggestions? Thanks again
If you are 100% sure your opponent will go ling infestor you can do MVPDream's blueflame/bio (with heavy marauder emphasis) which simply shreds it. Look at his games vs Golden and Sleep from GESL and other Zergs in Spring Arena. In general ling/infestor has severe issues with a screen of 6-8 marine/marauder in front of your army, and cannot handle large numbers of medivacs whatsoever. You can push a lot faster against pure ling/infestor than other comps if you are diligent about stimming a small number ahead and pre spreading behind.
Lastly if you do go tanks, make sure to target fire infestors in every battle. See Flash vs Soulkey on Antiga from the KESPA Invitational at MLG Anaheim for a strong example of this.
On September 01 2012 22:59 Sp4cem4nSpiff wrote: What builds do you recommend in the current metagame for each Match Up?
TvZ:
Taeja's triple cc -> hellion/banshee or Ghostking's 14cc hellion/banshee he used vs nestea
TvT:
Rax CC/14cc into 2 rax tech to medivac. Vs another rax cc, triple cc.
TvP:
Taeja's rax cc 2 rax double gas to medivac with a faster than normal armory
On September 01 2012 23:12 NiNjAPlation wrote: Hi I'm a mid platinum Protoss, having trouble with the PvT matchup, I'm not necessarily having trouble with any specific comp or build. I'm having trouble dealing with the harrass Terran can do wether it's an early game banshee, blue flame hellions, or just your standard medivac drop. So my question to you is, what's the best way to help deal with this harrass?
Best way to handle the harass is just become comfortable facing it and know beforehand what you need to do. When a pro gets harassed like that he never gets flustered, he just plays on through predeveloped patterns.
Vs Banshee- 3 Stalker/1 obs at each min line, positioned along the main safe aerial entrance to min line to intercept any banshees. Vs hellions- Proper building placement to form walls and make it difficult to run around, and warp in stalkers when needed. Stalkers/cannons in min line stop it very easily. Vs Drops- Make multiple observers and patrol them along the likely drop paths, making sure you always know where they come. 1-2 Templar + blink stalkers at the important spots will nullify any drop.
On September 02 2012 02:16 ShnAndrei wrote: @Ver
Thanks a lot for the replay tips! I think I will go with MVP's TvZ builds@IEM in which he opened straight into bio, and when I will get to the necessary level of multitasking, I will move on to the hellion-banshee opening.
You wrote earlier, regarding TvT: "A much safer opening against almost any 1 base play is Rax cc double gas depot fact port, getting reactor on rax, tech lab on fact, and making marine/tank/viking."
This seems like an awesome TvT build to try. I always wanted to do a FE but I was afraid of my opponent getting fast tanks with vikings and just stepping over me. But this seems rather safe to try. Could you please point me to a replay in which this was carried out?
P.S. Thanks a lot for this fantastic thread! If every top player shared his tips and secrets with the rest of the community the way you do, we would have faced a whole other level of gameplay and skill on ladder. I think this is the most interresting thread I've ever read on teamliquid.net.
I'm glad you like it iirc Kas used it against MVP on Entombed Valley in IEM group stages, however he did a weird transition and blew his lead.
On September 02 2012 05:36 Whatson wrote: Is marauder hellion actually a viable unit composition in the early game TvP? I know in theory the hellions can kill the zealots so the marauders can kill the stalkers/sentries. If it is, are there any specific scenarios where its better than regular bio?
On September 02 2012 05:36 Whatson wrote: Is marauder hellion actually a viable unit composition in the early game TvP? I know in theory the hellions can kill the zealots so the marauders can kill the stalkers/sentries. If it is, are there any specific scenarios where its better than regular bio?
Hellion/Marauder is a pressure opening (there are all-ins variations though, especially against Nexus first), you can find a guide here. Usually this kind of pressure build is used in a BoX series because Protoss cannot know for sure what you're up to upon scouting you went 13 gas.
Yeah it only works in small numbers as a way to coinflip greedy expansion builds because of how fast hellions reinforce. In the midgame it's poor efficiency is not worth any tradeoff.
On September 02 2012 06:41 dOraWa wrote: I've been trying to expand my repertoire of cheese (mid masters terran here) and I'd really like to learn how to use the 1-1-1 effectively vs toss (with or without SCV pull and cloak). I'm aware that it was quite an effective strategy vs toss for a long time, but every time I try to execute it I get stopped very easily.
Which build order(s) with cloak and without cloak would you recommend? Any replays you can direct me to? Thanks
On September 02 2012 08:11 Jermman wrote: The 1-1-1 is mostly figured out by protoss these days (as far as I know). Correct me if im wrong ver.
Yup! Way too easy to stop so long as the Protoss doesn't take banshee damage and properly stalls/flanks the eventual attack. I highly recommend not learning this build.
On September 02 2012 14:22 Natespank wrote: What is the best stutter step method? Is it to use the stop command and right click, hold position and right click, or a-move and right click, or stop-command and a-move? Also, what is the best way to learn to perfect it, especially with stimmed marines?
I personally alternate between right click and hold pos or attack move, depending on how I want my units to move/what formation to take. However as I have no experimented with the others I cannot make claims of their validity or not.
Best way to learn to perfect it is to watch vods of the best players with this, probably MKP, to see exactly what it looks like. Then you want to just practice reaching his level in micro trainers, first getting the timing right, then practicing that timing over and over until it's subconscious. Finally you want to bring that into your real games with the focus of that game on not winning but simply microing correctly.
On September 02 2012 20:44 Sianos wrote: Would you recommand the 10 minute timing attack with marines,hellions,medivacs,combat shield,stim and +1/+1 and followed up by a 3rd base and the transition inot marine,tank as a standard tvz opener for ladder? Or would you recommend to use the hellion and cloak banshee build followed up with a 3rd cc, double upgrades and the transition into marine,tank? Or both as good as the other or would you say that one of those is the better way to go?
The hellion/banshee/3 cc is far better without question, as in it will place you in a good position even if the Zerg knows it is coming and knows how to play against it.
The marine/hellion/medivac attack with +1 attack (the general pro perception is that +1 armor delays the timing too much) will beat a lot of people who don't know how to react to it but once you get to a higher level it will flop badly. My ex-teammate Mystik (who hadn't played for a month) easily beat Illusion 2-0 at WCS NA who did the two versions of this build, because Mystik knew how to stop it and I don't think it's possible to execute it much better than Illusion did. If you want to practice it you can, as it is easier to execute, but obviously for the long term it is much worse.
Korean ladder is better in every way than NA (which has improved a lot) and EU. You actually get garbage TvT practice outside of Korean ladder, as foreigner Terrans don't know how to scv cut. So yes, this is true. The gap between Kor T/Foreigner T and the equivalent from other races is much larger. This is mostly due to Korean Terrans being more willing to coinflip/rely on luck based allins to win, and Terran being so utterly unforgiving. Terran can compete but only if they can go through the game without ever getting behind (which is why Taeja wins). Foreigner Terrans don't have that extreme solidity in their play. In addition, Terran really has only survived until this point by creativity, and Korean Terrans are much more creative than foreigners. Other races don't need to innovate as much, so this effect is less pronounced.
What did you mean by this?
1) Can you give some examples of a proper SCV cut? Is it only a current meta thing, where 1 base openings are prevalent?
2) KOR Terrans are using coinflip builds..... only because Terran it self is unforgiving? Correct me if I'm wrong, but I believe you're talking about TvT here.
3) Is there a "science" behind Korean coinflip builds? Why would they take such huge risks like that? Is it because the coin is "weighted," so to speak, in their favor?
NA 1200 master Terran TT
PS I love this thread btw :D
1. Sorry for not being more precise: I meant in 2 base scenarios or with quick triple cc where people will halt scv production for 1-2 rounds at a time to get your 4th/5th rax quicker. In TvT before a large number of tanks are in play having a few more marines is a lot more important than more scvs. Some Korean Terrans on ladder will do this to an excessive degree, and if you don't cut some as well they'll just 1a through you if you opened similar builds.
In general you can only use 44 scvs to mine on 2 bases and a couple more to build, so it's more important to cut scvs beyond that point to get your 3rd cc, 4th/5th raxes, and 2nd factory faster as the scvs don't provide a benefit until several minutes later.
2. If Koreans win with coinflips they don't have to play normal games where the unforgiveness of Terran is so prominent. On the contrary it's very unforgiving to play against Terran allins.
3. It's not a science. It's partially weighted, but it's also not a coinflip if the opponent doesn't know how to react properly. It's much easier to learn how to beat a macro, reactionary build, with a very narrow and precise timing than doing the former and being able to handle every kind of attack. So Kor T's will get lots of free wins taking advantage of people who have not perfected their builds in all matchups.
On September 04 2012 14:58 ThaSlayer wrote: Im a Masters Terran struggling with T v T. I often lose games when I mech or face mech. 1) As the meching player, when is a good time to switch to sky terran? - I tend to play too passively and sit around my 4th, only to realise all my tanks are dead weight when BCs and Vikings appear 2) When moving out around 16min (the mid game push), what is a good siege tank count to leave at my 3rd to defend? 3) I've been doing a rather awkward cloak banshee (standard 13 gas timing) into expansion, and I as I get my production started around 8-10 min, I am rather vulunerable to drops - What is a good FE build into mech?
Thanks!
1. I don't actually think you want to go battlecruisers unless you have an overwhelming lead and they are going air too or you have an econ lead and they are going bio/tank. You want to try to kill/cripple with a tank army, as tanks beat bio too cost effectively in high numbers better than BCs do.
2. 1-2 at most. Leaving units behind vs mech is actually really bad because otherwise it's too easy to get a giant flank and kill the depleted mech army in the open. It's better to be very proactive about scouting with hellions sensor towers and scans to find out if they are going to backstab and go kill that initially, then move on their weakened army.
3. Many openings work: something that involves hellion/banshee is ideal to give you some control and harass potential (such as rax cc into reactor hellion banshee cloak). It makes it a lot harder to play vs mech if you constantly have to worry about hellion backstabs. Getting +1 attack early is crucial for mech: + armor is best saved for when your position is more stabilized and you have enough tanks.
On September 01 2012 11:13 Natespank wrote: Hello Ver,
What do you think of custom keyboard layouts? I noticed for example that you bind your rax to 7 or 8, so I assume you customized your keys. What do you think of layouts like Darkgrid, Chameleon, the Core, or Artosis's hotkey video from recently?
edit: Do you recommend learning a macro cycle, similar to those described in this thread? I've been considering it for some time, but I want to decide whether I will use a custom keyboard setup first so I don't waste the time training my hands on something I won't use.
Honestly I never put any thought into this and have a mostly standard key setup with the exception of some area hotkeys rebinded. A flexible macro cycle is good because it will let you train unconsciously easier, but at the same time it's not necessary. It's more important I think to watch fpviews in stream or replay of top T pro's to see how they do it, as there are some tricky things you need to learn, such as not macroing while kiting TvP, and how to prequeue units before an engagement.
On September 01 2012 09:59 dynwar7 wrote: Thank you very much Ver for the help.
I know you have lots of questions in this thread and you are busy, but I apologize, please allow me toask 2-3 more questions.
1. Wel, other than that youtube link, is it possible if you upload 2 or 3 replays of you going mech vs bio?
2. I absolutely love ghosts, the reason I play terran. What do you say about using nukes vs tank lines?
I guess that is all. No rush! Thanks again for your help Ver, you are Very Very nice :D
1) Yeah. These are all old games cause I can't find any ladder ones recently but they use the same formula. + Show Spoiler [Reps] +
2. Yeah this is possible, especially with bio vs mech as you can take advantage of their unsiege more directly. I suggest watching bomber vs sjow on testbug from MLG Raleigh 2011.Normally it's more important to get air first but this is an option if you can use it to take a strong position.
On September 01 2012 22:26 Rabiatic wrote: Hey, thanks for doing this Q/A!
I have a question about TvP. I am a top diamond player, and I struggle a lot with 1base openings like 4gate robo warp prism, 4gate blink stalker etc, and I'm wondering what's the best response against these kinds of builds? When I spot 1base play doing 1rax FE, I usually just bunker up my nat, but then they just elevator/blink into my main and I'm screwed.
Should I bunker up my main as well? If so, where should I positon my bunkers? And also, when do I get the addons on my barracks, and when do I get up my factory+starport?
1 base allins can be divided into 1 gas and 2 gas, which is something you can see. Of course they can wait to chase your scv out with a stalker and then plant a 2nd gas but that will delay the allin and make it easier to hold regardless. The list is quite vast and it's pretty close to impossible to be 100% prepared against every one on most maps, making it possible for much worse Protoss players to blind coinflip even the best, though their winrate will be very low. Off the top of my head with vods for how to stop it ->
1 gas-
4 gate (620) -> make more than 1 bunker, bring scvs to nat to mass repair 4 gate warp prism (625-640) -> noblesse vs arcanne, terminus, iron squid regionals 3 gate void w/ zealots -> mma vs hongun, entombed/antiga, recent gsl.
2 gas-
blink obs (745-815~) -> discussed earlier. drewbie vs ddoro, daybreak, NA invite only qualifiers for winter arena. dt (715) -> turret or save scan Delayed 4 gate (can happen anytime with more sentries) dt warp prism (dts slightly later, warp prism a min later) -> white ra vs me, dual sight, mlg orlando Immortal (8-9, not sure exactly) -> Marineking vs Parting, Dual Sight, MLG Arena (not an allin variant, but shows the best way to handle it) Colo (9-930, not sure exactly) Warp prism ramp block (830~) -> wbc vs puma, shakuras, mlg raleigh 2011
I truly honor you for this thread, Ver. Really. With the advice you gave me in just 3 days I made it to platinum, now I'm rank 13 and I'm really close to making it to diamond. I really feel that my mechanics, builds, and just overall mechanics are better than before, and I feel more competitive as the opponents are getting better as well. Getting really close to diamond, and trying my best to improve as much as I can before placement happens again, I want to ask a few questions. 1. This doesn't really have to do with Terran that much, but maybe Terran specific advice could be given. How much of a gap is it from platinum to diamond? Are diamond players' mechanics much better, or are they not too hard of a leap from platinum to diamond? Are the builds they do much faster too? Do they max out much faster than high platinums, and just pay attention to everything a lot better? 2. Sometimes if there is a rush that the other player does, or just something happens that makes me stockpile on 1000+ minerals that I can't really spend, what am I going to be spending it on? I always think that if I make too much more barracks, starports, or factories, that when the surplus of minerals are gone, that's too much production for me, and that I can't get enough minerals to make as much units as the production facilities can hold. I think that it's much better if I expand instead, so that I can get more scvs and have a better economy going, but I always wonder to myself if it's a good idea to get the expansion, because the only reason I'm getting it is because I have a surplus of minerals, and I may not be able to defend the expansion or maybe I should have spend the minerals on production facilities instead. Does the early, mid, or late game come into play when it comes to this situation? Like, if I'm in the early game, I should expand so that I can get better macro going, or mid game, I should get more barracks because I have a decent economy already, or is it more of a "what's better in this situation" kind of thing?
On September 05 2012 01:29 SHODAN wrote: I'll try to explain the strategy behind triple OC tanks, and economy-heavy builds in general. Early tanks allow you to play as economy-heavy as possible. It lets you take a faster 3rd OC, faster 3rd and 4th gas, and faster upgrades (double engi bay at 6:50). Your early tank count then allows you to land your third much faster (9:30 mark), take the 5th and 6th gas faster, start 2-2 much faster. By 11:00 you will have 1 viking and 2 medivacs. After the viking has finished clearing nearby overlords, you use it as a spotter to track the location of his army. 2 medivacs and 16 supply of bio can freely clear up creep, then lift-off when you're in danger. This also opens up the threat of drops, whether you choose to commit to them or not. At the 13:30 mark, you will be at 180 supply, 6-7 tanks, 4 medivacs, 2-2 finished, 10 naked raxx, 1 with a techlab (only add reactors after you start 3-3). You attack at 2-2, shut down Z's fourth if it exists, trade well with his army and keep trading... I don't rely on gimmicky shit to get an early advantage over zerg. I try to outplay Z in the midgame with core units, drops, and tank lines. Z can safely make 75 drones no matter what opening you do. What really matters is gas, and Z can never have more gas than you with an economy-heavy build.
I know perfectly well both the reasoning and the benchmarks since, as stated some pages ago, I used to play this kind of build. Of course Zerg will reach 75 drones, that's not the point—the question is how much time until he gets that, and how much he has to dedicate to his defence. With triple OC tanks answers are “as fast as possible” and “zero,” which is not good since both creep and Zerg economy are exponential. You might not like Hellions/Banshees openings but at least they don't allow Zergs to reach 3-bases saturation uncontested. Zergs have to be careful against Hellion openings, they have to get Spores against Banshees follows-ups, they can't just sit at the edge of the creep with their extra Queens and allocate all their energy to new Creep Tumors. There's no doubt you hit a powerful 200 2/2 max army by the 14' mark, but it doesn't matter since creep prevents you from using it to its full potency, thus allowing the Zerg to welcome you with unchallenged BLs or 5/3 Ultralisks. Not saying this build will auto-lose but Zergs should definitely have the advantage against it should they go standard macro.
The question for me is how long can I keep zerg in the dark about what I'm doing. If Z sees my natural walled off with 3 raxx, Z must prepare at least for the possibility of bio pressure. Z will most likely take gas, upgrade zergling speed, save larvae for zerglings, and try to sack an overlord. If I feel Z is playing too greedy, I can fake pressure with marines & a bunker at his third; even in small numbers, marines trade really well with slow lings. IM_Yoda faked a 3 tank push against Symbol earlier today and I think it too would work great with this build (G2 Ohana). The maps that work best for 3OC tank (Ohana, Entombed, Daybreak) are also perfect for 2rax pressure. How about 2rax pressure ---> QXC bunker ---> 3OC/tank? How about 3OC/tank ---> tank drop the ledge of Z's main on Antiga? And as I said, medivacs give you the mobility to control creep in the midgame.
Dwf already covered almost everything so I'll just mention a few points:
-The Yoda/MMA games are irrelevant because the Zergs in question had 2010 level creep spread. You can't fake any kind of pressure when creep is at your watchtower.
-"faking pressure" with marines/bunker at his 3rd does absolutely nothing, as he just pulls his new queens and kills it before the bunker ever gets up.
-Zergs can't be kept in the dark except on cloud kingdom. If they see 3 cc and siege tanks, or even a tech labbed fact alone, all they have to do is take a really fast 4th hatch at a 4th, drone uninterrupted to 85-90, and rush the fastest possible hive. By the time the push reaches a dangerous area after clearing acres of creep they will already be producing their hive unit of choice (or done) and have thousands of minerals in the bank to use however they want. Although the build order is different, the creep spread weaker, and the game prepatch, heavenslight vs select from lonestarclash is a similar example of what happens vs these passive builds.
-Sadly Zerg is ridiculous enough that even the triple cc tank almost auto loses to certain coinflips, namely the slayers 1 or 0 queen baneling bust that hits before 550 and suffers severe damage vs the DRG 630 35-40 ling raid. Reactor hellion holds vs either.
-While early roach attacks can coinflip reactor hellion banshee builds in certain spwans/bunker placements, I've found they also can be mitigated. I've gotten a relatively even position vs Violet and Suppy who tried early roach timings on close Antiga/cross Entombed respectively. The key is to make 2 extra orbitals asap and use your cloaked banshees to deny his 3rd. That way he's stuck on 2 bases until he can get mutas, while you can replenish the scv deficit with 4 OC on 3 bases.
The other option is to make a marauder from the tech labbed rax, make the tech lab on the port on its own, and hold the ramp/bunker area with the marauder wailing on the roaches. See MKP vs Symbol, Antiga, MLG Anaheim.
-No idea what day9 has to do with anything, as I haven't watched him since BW, but Taeja is the only Terran in the world who reliably enters the midgame from an even or superior position against every Zerg he's played regardless of what they do (unlike Gumiho), so it's clear he's doing something right nobody else is.
On September 05 2012 03:24 Sianos wrote: As you mentioned the "goto" build in tvz right now is the 1 rax fe into reactor hellion into cloak banshee. Is it a must for terrans to cut scvs and marines to get both gases quickly and then build the depot and continue production or is it ok to first build the depot and continue production and then add the 2 gas? With the later version the first 2 hellions will pop out at 6:35, but i feel more safe against the initial 6 lings since you cut your marine production really long with the first version.
Another question. How many banshees and hellions do you build, when you want to transition into marine tank? Is there a "standard" number or do you do it when you think your pressure isn´t going to work anymore? I heard about 8 hellions and 3 banshees somewhere on reddit, what´s with this number?
And last. What is the normal way to transition into marine,tank,medivac? Do you build your CC, then double gas, then double ebay + tl on rax, then start all 3 upgrades, then adding additional barracks? When is the correct time to begin tank production inbetween them? Do you allways use the tl from the starport to transition? Can you advice with some good replays where the transition is nicely done?
You don't need to double gas before depot, except if they do gas before pool. Taeja often will make 3 cc's before any gas if he spots no gas from the Zerg.
6 hellion/2 banshee is the standard due to trade offs in either direction, but as it is the standard, Zergs will be more prepared for it. 8/3 will let you bully the standard 6 queens around and delay creep considerably, but will set back your initial move out with marine/medivac a bit. 4/2 or 4/1 is mostly there to catch the Zerg off guard with a really early 11-12 min pressure but isn't the most reliable.
If the Zerg hasn't taken a third by 7 mins or so and you don't see a third hatch, you can't make more than 6/2 due to the threat of 2 base muta allin and you need to prioritize retaining everything (Life/Keen, TSL, Cloud Kingdom).
Yes you start 3rd cc/upgrades first, then add more barracks. You make tanks when you can, basically. They can be delayed or rushed, depending on how aggressive you want to be. Using the tech lab on starport for the fact is generally a good idea, yes. The rax should make its own earlier.
Taeja vs Shine from TLopen on Antiga is a good example replay.
On September 05 2012 07:51 happyft wrote: Ver, you've mentioned that something which makes Korean Terrans so dangerous is their ability to all-in -- something that a few other notable posters have said as well. Do you have any nice all-in builds you could recommend to us?
TvZ: Proxy 11/11, pull 3 scvs or 5. (Taeja vs Life Antiga, TAC; Keen vs Life on Ohana, TSL.) TvT: cloak banshee into marine tank allin (polt vs mvp, dual sight, gstl) TvP: - gas rax reactor hellion marauder (mkp/killer,mlg anaheim, antiga) -11/11 or 12/14, with 1 rax hidden (Alive, MLG Columbus, Daybreak games in particular, and Alive/Parting on Metro from IPL 4). -Proxy hellion drop (MKP/PArting, Antiga, GSL; Taeja/Finale, TL open, Antiga).
On September 06 2012 10:38 Absurd Bunny wrote: I truly honor you for this thread, Ver. Really. With the advice you gave me in just 3 days I made it to platinum, now I'm rank 13 and I'm really close to making it to diamond. I really feel that my mechanics, builds, and just overall mechanics are better than before, and I feel more competitive as the opponents are getting better as well. Getting really close to diamond, and trying my best to improve as much as I can before placement happens again, I want to ask a few questions. 1. This doesn't really have to do with Terran that much, but maybe Terran specific advice could be given. How much of a gap is it from platinum to diamond? Are diamond players' mechanics much better, or are they not too hard of a leap from platinum to diamond? Are the builds they do much faster too? Do they max out much faster than high platinums, and just pay attention to everything a lot better? 2. Sometimes if there is a rush that the other player does, or just something happens that makes me stockpile on 1000+ minerals that I can't really spend, what am I going to be spending it on? I always think that if I make too much more barracks, starports, or factories, that when the surplus of minerals are gone, that's too much production for me, and that I can't get enough minerals to make as much units as the production facilities can hold. I think that it's much better if I expand instead, so that I can get more scvs and have a better economy going, but I always wonder to myself if it's a good idea to get the expansion, because the only reason I'm getting it is because I have a surplus of minerals, and I may not be able to defend the expansion or maybe I should have spend the minerals on production facilities instead. Does the early, mid, or late game come into play when it comes to this situation? Like, if I'm in the early game, I should expand so that I can get better macro going, or mid game, I should get more barracks because I have a decent economy already, or is it more of a "what's better in this situation" kind of thing?
1. I'm glad you've gained so much from it! Honestly I have no idea; I've never played below masters. However, if this is anything but idle curiosity, you should not be thinking about trivialities like this. Your own worth or skill is not determined by those around or, or at least it is detrimental to think in such a way. Best focus on only yourself, and you'll grow in ranks naturally without much effort.
2. This really depends, but in general you should do the build in the order that you went in the game planning on it even if you have excess money. But in general, adding a 3rd cc is a good idea if you're floating that much unless your opponent is doing a dedicated 1 base allin that you don't feel you can hold the next wave of.
On September 05 2012 01:29 SHODAN wrote: I'll try to explain the strategy behind triple OC tanks, and economy-heavy builds in general. Early tanks allow you to play as economy-heavy as possible. It lets you take a faster 3rd OC, faster 3rd and 4th gas, and faster upgrades (double engi bay at 6:50). Your early tank count then allows you to land your third much faster (9:30 mark), take the 5th and 6th gas faster, start 2-2 much faster. By 11:00 you will have 1 viking and 2 medivacs. After the viking has finished clearing nearby overlords, you use it as a spotter to track the location of his army. 2 medivacs and 16 supply of bio can freely clear up creep, then lift-off when you're in danger. This also opens up the threat of drops, whether you choose to commit to them or not. At the 13:30 mark, you will be at 180 supply, 6-7 tanks, 4 medivacs, 2-2 finished, 10 naked raxx, 1 with a techlab (only add reactors after you start 3-3). You attack at 2-2, shut down Z's fourth if it exists, trade well with his army and keep trading... I don't rely on gimmicky shit to get an early advantage over zerg. I try to outplay Z in the midgame with core units, drops, and tank lines. Z can safely make 75 drones no matter what opening you do. What really matters is gas, and Z can never have more gas than you with an economy-heavy build.
I know perfectly well both the reasoning and the benchmarks since, as stated some pages ago, I used to play this kind of build. Of course Zerg will reach 75 drones, that's not the point—the question is how much time until he gets that, and how much he has to dedicate to his defence. With triple OC tanks answers are “as fast as possible” and “zero,” which is not good since both creep and Zerg economy are exponential. You might not like Hellions/Banshees openings but at least they don't allow Zergs to reach 3-bases saturation uncontested. Zergs have to be careful against Hellion openings, they have to get Spores against Banshees follows-ups, they can't just sit at the edge of the creep with their extra Queens and allocate all their energy to new Creep Tumors. There's no doubt you hit a powerful 200 2/2 max army by the 14' mark, but it doesn't matter since creep prevents you from using it to its full potency, thus allowing the Zerg to welcome you with unchallenged BLs or 5/3 Ultralisks. Not saying this build will auto-lose but Zergs should definitely have the advantage against it should they go standard macro.
The question for me is how long can I keep zerg in the dark about what I'm doing. If Z sees my natural walled off with 3 raxx, Z must prepare at least for the possibility of bio pressure. Z will most likely take gas, upgrade zergling speed, save larvae for zerglings, and try to sack an overlord. If I feel Z is playing too greedy, I can fake pressure with marines & a bunker at his third; even in small numbers, marines trade really well with slow lings. IM_Yoda faked a 3 tank push against Symbol earlier today and I think it too would work great with this build (G2 Ohana). The maps that work best for 3OC tank (Ohana, Entombed, Daybreak) are also perfect for 2rax pressure. How about 2rax pressure ---> QXC bunker ---> 3OC/tank? How about 3OC/tank ---> tank drop the ledge of Z's main on Antiga? And as I said, medivacs give you the mobility to control creep in the midgame.
Tank dropping the ledge itself doesn't let you get in range of much, and also if you don't have a large army, queens are pretty good at breaking this. Anyway, I just wanted to expand on the 3 oc tanks.
I've seen Guhimo do it on Daybreak, go cc first into 3rd cc before gas tanks. Because his CC go down so quickly, he's able to have the normal 14 minute move out by before 12 minutes, equal upgrades and lots of tanks. Given that there's almost no chance you'll have to fight hive tech here, and instead all the money Zerg was banking turns into banelings/ling to hold this, do you think this is more viable (as compared to a less greedy triple CC tank build)? Even assuming your opponent plays as greedy as possible, you'll be fighting lair tech with a huge 2/2 1/0 marine maruder tank medivac army.
If not all maps, at least specific to Daybreak where the 4th base should be pretty close and such a greedy opener is safer then on other maps?
Edit: this also assumes you don't die to an ultra fast bling bust too, as I concede the hellions are needed to defend those (but also those take a quicker gas that your scouting scv should be able to see, right?).
Hey! this isn't a super specific question to Terran but when I'm playing TvZ I have a lot of probelms with "crisis management"
I'm usually opening the pretty common 1 rax expand into cloaked banshees+hellions into third base then going mech. A lot of Zergs go for Roaches, Lings and Banes and try to bust me and regardless of how well I hold it puts the game on a completely different track than I'm used to and I often end up playing very scrappy, having a weird mineral:gas ration, not being able to support the production I have and so on. I find the same thing happens with drop play, I can hold it off decently but my timings get thrown out of the window from my attacks being delayed, buildings being sniped etc.
I think a lot of people get the same issues in ZvP/TvP where they're harassed by Dark Templar or a surprise warp-in. But whenever I see the top players such as MKP or Stephano play they seem to be able to manage the situation really smoothly even if they're completely unprepared.
Do you have any advice for getting better in situations you're not really used to being in or does it just improve from more practice?
On September 05 2012 01:29 SHODAN wrote: I'll try to explain the strategy behind triple OC tanks, and economy-heavy builds in general. Early tanks allow you to play as economy-heavy as possible. It lets you take a faster 3rd OC, faster 3rd and 4th gas, and faster upgrades (double engi bay at 6:50). Your early tank count then allows you to land your third much faster (9:30 mark), take the 5th and 6th gas faster, start 2-2 much faster. By 11:00 you will have 1 viking and 2 medivacs. After the viking has finished clearing nearby overlords, you use it as a spotter to track the location of his army. 2 medivacs and 16 supply of bio can freely clear up creep, then lift-off when you're in danger. This also opens up the threat of drops, whether you choose to commit to them or not. At the 13:30 mark, you will be at 180 supply, 6-7 tanks, 4 medivacs, 2-2 finished, 10 naked raxx, 1 with a techlab (only add reactors after you start 3-3). You attack at 2-2, shut down Z's fourth if it exists, trade well with his army and keep trading... I don't rely on gimmicky shit to get an early advantage over zerg. I try to outplay Z in the midgame with core units, drops, and tank lines. Z can safely make 75 drones no matter what opening you do. What really matters is gas, and Z can never have more gas than you with an economy-heavy build.
I know perfectly well both the reasoning and the benchmarks since, as stated some pages ago, I used to play this kind of build. Of course Zerg will reach 75 drones, that's not the point—the question is how much time until he gets that, and how much he has to dedicate to his defence. With triple OC tanks answers are “as fast as possible” and “zero,” which is not good since both creep and Zerg economy are exponential. You might not like Hellions/Banshees openings but at least they don't allow Zergs to reach 3-bases saturation uncontested. Zergs have to be careful against Hellion openings, they have to get Spores against Banshees follows-ups, they can't just sit at the edge of the creep with their extra Queens and allocate all their energy to new Creep Tumors. There's no doubt you hit a powerful 200 2/2 max army by the 14' mark, but it doesn't matter since creep prevents you from using it to its full potency, thus allowing the Zerg to welcome you with unchallenged BLs or 5/3 Ultralisks. Not saying this build will auto-lose but Zergs should definitely have the advantage against it should they go standard macro.
The question for me is how long can I keep zerg in the dark about what I'm doing. If Z sees my natural walled off with 3 raxx, Z must prepare at least for the possibility of bio pressure. Z will most likely take gas, upgrade zergling speed, save larvae for zerglings, and try to sack an overlord. If I feel Z is playing too greedy, I can fake pressure with marines & a bunker at his third; even in small numbers, marines trade really well with slow lings. IM_Yoda faked a 3 tank push against Symbol earlier today and I think it too would work great with this build (G2 Ohana). The maps that work best for 3OC tank (Ohana, Entombed, Daybreak) are also perfect for 2rax pressure. How about 2rax pressure ---> QXC bunker ---> 3OC/tank? How about 3OC/tank ---> tank drop the ledge of Z's main on Antiga? And as I said, medivacs give you the mobility to control creep in the midgame.
Tank dropping the ledge itself doesn't let you get in range of much, and also if you don't have a large army, queens are pretty good at breaking this. Anyway, I just wanted to expand on the 3 oc tanks.
I've seen Guhimo do it on Daybreak, go cc first into 3rd cc before gas tanks. Because his CC go down so quickly, he's able to have the normal 14 minute move out by before 12 minutes, equal upgrades and lots of tanks. Given that there's almost no chance you'll have to fight hive tech here, and instead all the money Zerg was banking turns into banelings/ling to hold this, do you think this is more viable (as compared to a less greedy triple CC tank build)? Even assuming your opponent plays as greedy as possible, you'll be fighting lair tech with a huge 2/2 1/0 marine maruder tank medivac army.
If not all maps, at least specific to Daybreak where the 4th base should be pretty close and such a greedy opener is safer then on other maps?
Edit: this also assumes you don't die to an ultra fast bling bust too, as I concede the hellions are needed to defend those (but also those take a quicker gas that your scouting scv should be able to see, right?).
MMA somtimes tank drops on Antiga off of a 1 rax gasless FE. Last time I saw it (S3 Code A) he elevator'd 1 SCV, some marines, and made a bunker in Z's main under cover of tank fire. Pretty cool harass. Bunker usually gets in range of a spawning pool or evo chamber and marines can poke in and out. I'm not sure how it would work with 3OCs though...
Daybreak I started taking the fourth as you mention, and I wall off the choke to my "third" with 4 depots incase of runbys. It's so easy to defend the forward third with tanks because you have xel'naga tower vision. Daybreak is one of those maps where you can get a really solid wall at your natural with only 2 raxx and a forward orbital (once you figure out the layout), and there's that little nook on the high ground for tank fire. The downside to Daybreak is that there is no ramp to Z's nat for a QXC bunker (yeah, I swear by QXC bunkers. I don't know why it was just a fad. No QXC bunker is like not sending your first 2 dragoons to harass terran's bunker in BW!).
I'm sure you'll have your eye on Gumiho next week in GSL, but if you haven't heard, he started streaming on youtube and he does this 3OC tank build all the time:
Hey I have been having problems with TvZ a lot lately and only seem to win against lesser opponents or with blind luck (zerg doesn't scout).
I have attempted the hellion / banshee opening into fast 3rd, but seems like when zerg scout's this they either do all-in (don't have tanks up and even with banshee defending I lose so many scv's) or they go fast muta (against I lose good amount of svc's).
I was thinking maybe forgo the hellion's in order to get faster siege tanks? I am also not that great with hellion's and they never really seem to do much of anything.
Currently I go marine tank in the mid-game and add in marauders and vikings in late game.
Idk I am so lost against Zerg right now - my mmr is high diamond/low masters, I seem to get all-in'ed a lot by Zerg right now - Zerg it seems goes either all-in super aggressive or super greedy and seems coin flip sometimes on what they are doing. Is there a good scout timing / scan timing I should use?
On September 07 2012 23:26 Eshez wrote: Hey! this isn't a super specific question to Terran but when I'm playing TvZ I have a lot of probelms with "crisis management"
I'm usually opening the pretty common 1 rax expand into cloaked banshees+hellions into third base then going mech. A lot of Zergs go for Roaches, Lings and Banes and try to bust me and regardless of how well I hold it puts the game on a completely different track than I'm used to and I often end up playing very scrappy, having a weird mineral:gas ration, not being able to support the production I have and so on. I find the same thing happens with drop play, I can hold it off decently but my timings get thrown out of the window from my attacks being delayed, buildings being sniped etc.
I think a lot of people get the same issues in ZvP/TvP where they're harassed by Dark Templar or a surprise warp-in. But whenever I see the top players such as MKP or Stephano play they seem to be able to manage the situation really smoothly even if they're completely unprepared.
Do you have any advice for getting better in situations you're not really used to being in or does it just improve from more practice?
You have to get caught off guard more. Be able to deal with the unexpected by playing against the unexpected
On September 08 2012 03:26 JIJI_ wrote: Hey I have been having problems with TvZ a lot lately and only seem to win against lesser opponents or with blind luck (zerg doesn't scout).
I have attempted the hellion / banshee opening into fast 3rd, but seems like when zerg scout's this they either do all-in (don't have tanks up and even with banshee defending I lose so many scv's) or they go fast muta (against I lose good amount of svc's).
I was thinking maybe forgo the hellion's in order to get faster siege tanks? I am also not that great with hellion's and they never really seem to do much of anything.
Currently I go marine tank in the mid-game and add in marauders and vikings in late game.
Idk I am so lost against Zerg right now - my mmr is high diamond/low masters, I seem to get all-in'ed a lot by Zerg right now - Zerg it seems goes either all-in super aggressive or super greedy and seems coin flip sometimes on what they are doing. Is there a good scout timing / scan timing I should use?
Thanks.
No don't forgo hellions to get faster tanks. You need to copy the progamer build EXACTLY; don't make changes on your own volition. If you want to do the tank thing you need to do gumiho's build listed. It's really inferior but is much easier to play as well. Hellion/banshee is unfortunately quite difficult to execute, harder than any other opening. Even a top level Terran needs a lot of practice with the build to get the execute good. I highly suggest you go through the build over and over on a map alone until you can macro right and control things a bit better.
Hellion/banshee is really the only build that 'prevents' the coinflip (if they defend perfectly you can get slightly behind vs 3 base full saturation but it will hold allins). If you are miscontrolling your hellions and losing them poorly, then that obviously is going to wreck the foundation of the build. You should generally (map dependent) never scan the Zerg before 13-14 minutes because you desperately need those mules. Rest assured you can hold all the allins: your hellions provide the early warning signal and give you time to react appropriately. If you are really afraid of them you can do the double gas before second depot (after cc) version which gives you the best defense, though weakest econ. I suggest taking scvs off gas once the fact/reactor are started for a certain period, as you get more gas than you can use.
Hellions do several things with this build:
1) Snipe off outlying tumors, control the watchtowers, and limit info for the Zerg. Zergs hate playing against Hellion/banshee because it keeps them in the dark for so long and that is the one thing Zergs feel very uncomfortable with. 2) Prevent queens from camping the edge of the creep and spreading creep too fast by threatening runbys. If the Zerg does not block runbys they will lose far too many drones normally. Only attempt runbys with 4+ hellions, and keep your hellions out of sight range and zoom in and out instead of camping them at the edge. Uncertainty for the Zerg is what you want. 3) Defend against any zergling aggression. You have to be able to micro/kite properly to do this and you cannot lose them carelessly.
It seems to me that gasless builds are simply bad in TvT. How can you stop a gas first banshee from getting 10+ kills? Banshee has more range than marines and is faster, so you will only kill it if your opponent is shit. And if you need to make engieering bay and 3+ turrets wouldn't you just have been better off going for a quick viking?
Also, holding off a 1/1/1 tank push after losing 10 marines to a banshee and not having stim yet? Not to mention gas first builds are so versatile you could lose to so many things. Marine/helion elevator, reactor helion, or cloak/non cloak banshee.
IIRC, MKP said you simply cant go 1 rax expo in TvT anymore, yet your guide says either 14 cc or 1 rax cc? 14 CC? how in the hell do you stop a gas first banshee with 14 cc????
EPT
iirc Kas used it against MVP on Entombed Valley in IEM group stages, however he did a weird transition and blew his lead. 
