Well... as promised, here's a replay. This is my latest loss, and isn't exactly exemplary of how I normally play, but I figure it's a good one to look at all the same, seeing as these kinds of losses are bound to happen again.
As I'm sure you can tell, I'm still in bronze. I was going to go for my usual 4gate, but I think that when I saw that he had roaches, I freaked out and deviated from my build to start pulling in some cannons and other stuff. I got supply blocked a ton, stopped producing units and workers, and also apparently just have no clue what to use to defend.
Feedback would be much appreciated; I know I've got a long way to go. Thanks!
I didn't watch the whole replay, but I think I got enough info out of it. Seems like you should work on getting a working understanding of each matchup as well as solid build orders. I say this due to a combination of your suboptimal 4gate/forge defense off scouting solely a hatchery, and your comment that you have no clue what to use to defend.
The way I did this was by watching pro streams, although day9/tournaments are also very good. Assuming you know what each unit does, you start to learn which units are good against other units, and how units can interact with each other with synergy (ex: sentry and protoss "deathball" units). It takes a while to build up your game knowledge and it's a bit hard to set goals for yourself, so I would also work on one precise build per matchup.
After I got the game this summer, the first thing I did was make a word doc with precise build orders written out, then I would practice the builds against AI until I executed them reasonably well. To measure my progress I looked up some pro replays to see how many units they had at what point, what buildings, etc - ex, 4 gates, 20 probes, warpgate done at 5:35 for a 4 gate. It probably took me 10 games before I ever finished warpgate with 20 probes and 4 gates by 5:50, and then another 10 before I could do that relatively consistently, then another 10 before I got it down to 5:40. I don't want to go into the details of builds, you can find them here http://www.teamliquid.net/forum/viewmessage.php?topic_id=266786.
After I got my 4 gate timing relatively down pat, I would use it on ladder, and push out with 1z1s1probe like I saw pros do so many times and warp in 4 stalkers. After I went up the ramp, I'd forget to make units for the next minute and lose with 1k in the bank. This kept happening but every time I got a little better at making new units while not losing all of my attacking units, until I got pretty good at 4 gating. Along the way I developed many basic mechanical skills, like hotkeying units, adding units to hotkeys, making probes constantly (or, up to a certain point for a 4 gate), moving my screen around between different key areas of the map, and generally getting quicker and more precise with my mouse movements and keyboard shortcuts.
I guess what I'm trying to say is that getting specific builds has three benefits: 1) You do the work in finding the build and reading about it - increases your game knowledge 2) Executing the build, just making probes and buildings, develops lots of basic macro skills by itself 3) Playing games with the build will develop other essential skills like very basic multitasking and elementary unit control
There are so many other things to worry about in sc2 but I think everyone needs to get the very basics down first, and I believe that learning a crisp build order and playing with it against people is the best way to do that.
On October 26 2011 18:58 whistle wrote: I didn't watch the whole replay, but I think I got enough info out of it. Seems like you should work on getting a working understanding of each matchup as well as solid build orders. I say this due to a combination of your suboptimal 4gate/forge defense off scouting solely a hatchery, and your comment that you have no clue what to use to defend.
I questioned posting this one because I figured that that was going to be one of my largest problems with this particular replay. I deviated from my 4gate, which I can actually execute in a pretty clean 6/6:30, because I freaked out. I was reading how everyone was saying that Zerg had a massive advantage on Metalopolis, and automatically freaked out thinking "crap crap crap, must defend against cheese! Get forges up! GET EVERYTHING UP!!!" And from there it was just a downward slide.
When I post my next loss it'll be me losing to one of my "clean" builds, for something a little more... representative?... of my play. This was me trying to branch out (while freaking out) and trying to do something new that, obviously, didn't do much except for make me look like a hot mess, haha.
That being said though, I really appreciate the advice! I might start up that Word document you mentioned.
If you're 4gate'ing there's really no cheese a Zerg can pull off since 4gate will kill off the stupid stuff like expo to the gold, one base muta rush. Maybe 7rr is effective but I don't really know since the econ loss is about the same for both sides. I guess a proxy hatch might work too.
Even a 6pool shouldn't be too scary since you should be able to wall off.
EDIT: Also I don't play Protoss so I can't be sure about this, but doesn't a clean 4gate still warp in the first round before 6 minutes? I remember that the "perfect" 4gat used to be 5:35 or something like that but the warpgate nerfs can't have delayed the build that much.
I don't play Protoss so I can't be sure about this, but doesn't a clean 4gate still warp in the first round before 6 minutes? I remember that the "perfect" 4gat used to be 5:35 or something like that
Hmmm. SC2Builds says that that's about right. So unless something's changed, I need to work on cleaning things up even further! I've never gotten below 6 minutes and I thought my build was pretty clean... though you sure wouldn't know it from this replay. -.-
I can't watch the replay on my current computer but you should definitly listen to people here. I can do a 4gate everytime around 5:45 is when my attack hits.
You need to focus on one buildorder per a matchup and practice each build till you know for a fact you have it down clean. Your 6-6:30 4gate won't cut it.
when you 4gate u can use "w" to auto select warpgates. If you can do 1 build very clean you will move up. I placed into bronze so i know its possible.
Yaaay, a replay to critique. I've been doing a lot of this lately. It is fun. It also gives me something to do while I wait for my green tea to cool off. For reference, I'm a high Diamond Toss. This was the series of thoughts that popped into my head whilst I observed your replay:
1) Basic opening build stuff that, judging based on what you've said above, you're probably already aware of (but just requires more repetition): Clean timing on the first pylon, gate and chronoboost. Gas was late being built and filled. Cybercore is supposed to come before the first zealot, not after.
2) Against zerg, watchtowers are a bad place to send your scouting probe because it's usually the first place they send their lings. Try to hide it somewhere outside their base, then sneak back into their base just after 5:00 to check A) that they haven't canceled their hatchery to do some 1 base cheese and B) whether they're still mining gas (which is an important tell, but what specifically it means you can worry about later, just get used to noticing it).
3) If your probe gets chased by a ling that does not have speed, it can actually make it home alive. Just select the probe and click onto one of your main mineral patches, and it should make it back fine.
4) If you do decide to deviate from your 4 gate and make cannons, you might as well throw down a pretty quick expansion first, and then build the cannons out front to defend it. Just more efficient.
5) Definitely need to start making sentries! Sooo important against zerg, you should get solid foundation of at least 6 as soon as possible, and protecting them from being sniped as long as possible is an important dynamic in PvZ. Making a bunch of sentries also would have let you build your expansion sooner, since they're super cheap on the minerals and you had plenty of gas (the basis of sentry fast expand builds, probably the next thing you'll want to get into for PvZ once you're past the 4 gate).
6) You scouted the spire with the observer, so you had some time to prepare for the mutas. The standard response is to put 2 cannons behind each mineral line and spread out your stalkers while you tech to blink. But, since you had 2 stargates, you could have also started chronoboosting phoenix and gained air control that way. That said, he didn't really use them, so uh... just something to keep in the back of your mind for future reference I guess.
7) With all that tech and plenty of minerals, you definitely want to take your 3rd and 4th gas at the natural.
8) You put 2 cannons in the middle of your base. Not sure why. If it was for mutas, behind your minerals is where you want to put those.
9) You had 2 sentries in your army during the first engagement. Remember to put up guardian shield and throw down a couple forcefields!
10) You cruuushed his army. This would be the perfect time to take a 3rd. Anytime you win a battle decisively means that they can't pressure you for a while because they just lost their whole army, which means you have a timing window of safety where you can get even more ahead.
11) When you counterattacked and lost all of your army except your voids but he had no anti-air, your best bet probably would have been to take advantage of this by targeting down his 3rd with those voids, killing that off, and then backing off with them.
12) Mined out in main but still no 3rd. Need to send the idle probes from main to natural. It might be a good idea to set certain benchmarks for yourself to make sure you take extra expansions - ie, "make sure I take a 3rd by 16:00" or something like that, and gradually make the time earlier and earlier.
13) Don't forget upgrades! They're important. You had that forge super early, might as well be chronoboosting some ups.
14) Zealot harassment with the pylon at the third - you could have actually warped in 6 (or waited for 2 rounds, 12) zealots and just targeted them right on the hatchery, and they would be able to kill it before the roaches could kill them. This is actually a very, very effective tactic that is used a lot in high level play, simply warping in a group of zealots and suiciding them onto a hatchery to deny the zerg expansions. Zealots have super high DPS, so with like 6-8 they can almost always take out hatcheries before the zerg can respond, unless they're already like right beside it. Much more useful than trying to take out those spines that were building and ultimately doing no damage.
15) Reeeally need to take a 3rd. Preferably you should take a 3rd well before your main becomes mined out, but that will take some practice, so if it does reach that point, that should definitely be a trigger in your mind to just take a 3rd immediately.
16) You kept making zealots for your main army. Against roaches, zealots are actually really terrible. Pure stalkers and immortals is the way to go against that.
17) Taking the gold was good, best possible decision given the circumstances.
18) You attacked and lost your voids and immortal. When you're behind, it's best to stay defensive until you've re-established an economy and turtle up until you have a giant army of death to come back with.
19) Just noticed you hotkey your nexuses (nexii?) separately. Are you aware that you can put both onto a single hotkey? Some people prefer them separate though, it's a stylistic thing. Also probably want to set up a hotkey for your robo.
20) Basically your best bet against what this particular zerg was doing was to take a 3rd much sooner (nothing you could really do once you were starved out), have an army of purely blink stalkers and immortals (zealots are sooo useless against roaches, only good for the proxy harass mentioned earlier, and the voids aren't useful enough against that composition to justify their cost), and also kept up with upgrades (he was 2/1 while you were still 0/0/0).
So, yeah. Way too much to remember or implement anytime soon, but, you know. Just telling you to 4 gate more would be boring, so there's some stuff to keep somewhere in the back of your mind for later.
On October 26 2011 18:58 whistle wrote: I didn't watch the whole replay, but I think I got enough info out of it. Seems like you should work on getting a working understanding of each matchup as well as solid build orders. I say this due to a combination of your suboptimal 4gate/forge defense off scouting solely a hatchery, and your comment that you have no clue what to use to defend.
I questioned posting this one because I figured that that was going to be one of my largest problems with this particular replay. I deviated from my 4gate, which I can actually execute in a pretty clean 6/6:30, because I freaked out. I was reading how everyone was saying that Zerg had a massive advantage on Metalopolis, and automatically freaked out thinking "crap crap crap, must defend against cheese! Get forges up! GET EVERYTHING UP!!!" And from there it was just a downward slide.
When I post my next loss it'll be me losing to one of my "clean" builds, for something a little more... representative?... of my play. This was me trying to branch out (while freaking out) and trying to do something new that, obviously, didn't do much except for make me look like a hot mess, haha.
That being said though, I really appreciate the advice! I might start up that Word document you mentioned.
One of the best things you can do for your play is to stop listening to what the mob says about X race having X advantage on X map. It's so easy to psyche yourself out before the game is even close to decided.
A good way to keep this from happening is via scouting. Even if its not optimal, get that observer out and keep an eye on his base - by knowing everything that he (or she, sorry ladies) is doing, your stress level stays nice and low and your mind stays clear and focused.
Many players (myself included) know what it feels like when the panic switch flips and all you can do is watch yourself turtle up - but by scouting/observing as much as possible, even though this deviates from your 4gate, you'll find your play become more confident, and that's the first step to improvement.
First I will say that I think it's important to start with the biggest problems in your play, fix those, and move onto more subtle fixes. This will give you the fastest improvement rate. Your play is actually not that bad, but you need a better build order, and there are some obvious macro and control issues, but like I said, start with the biggest problems and work from there.
First of all, It seems like your goal was a midgame play, including an expansion since you didn't cut probes, yet you built 4 gateways, a robo, and a forge before the expansion. This is a huge investment of minerals that could have been an earlier nexus since it's impossible to support all of this off of one base. My advice would be to three gate expand with a lowground forge. If you like the use of defensive cannons, this will be much better. 4gates should only be used if your plan is to end the game early with a 1 base timing. This is a fine strategy, but it's a rather boring way to approach the game in my opinion. Doing the three gate expand will give you an economic advantage, and allow you to make more units later. You can put the forge down after your nexus planted. Putting the robo before the nexus unless you have 1-2 gateways is a waste of resources because you can't fund it until the 2 base econ kicks in.
Second thing, you could have been a bit better about producing probes constantly. I'll admit, when I watched the replay and just stared at the nexus, you did pretty well with the probe production, better than I expected, but it is such an easy thing to perfect so do it. Also, you seemed to cut at 30 probes for quite a while. Not sure why you did that, since you had over 1k banked at that point. Use that to make your nexus, don't let it pool up like that.
Next, you saw his spire. You responded with void rays. Void rays are bad vs mutas, don't respond in this way unless you have a very good reason to believe that this spire is for corruptors (ie, he scouts you going colo, which he didnt in this case). You would have been better off going phoenix or blink stalkers. This is not why you lost the game, and is quite minor, but the game is so much built around "counters" that it's important to keep in mind as your macro improves and these things become more important.
Another thing I noticed is that you left your main way over saturated for a long time. Transfer those to your natural. You only need 16-20 probes on minerals per base, and your goal should be to get each base with AT LEAST 16 probes before you begin to put more on any other base. Get the gas geysers at the natural. Protoss needs gas, and you could have funded some better tech with it. You ended up having a very zealot heavy army because you didn't take the gas geysers, but zealots are really useless against roaches and mutalisks.
Another thing to note, you got a VERY early forge. This is okay, but it was not utilized. The preemptive cannons served no purpose and you didn't get any upgrades. It was a 450 mineral investment that really didn't do anything. If you want upgrades get a forge. If you see a reason for defensive cannons, get a forge. Don't make a forge so it can sit there not do anything though.
Get upgrades. You had a superior army composition and watching the final battle was almost confusing since it seemed like you should have won that convincingly. The problem however, is that he was ahead of you on upgrades due to your idle forge. If you had used the forge for getting upgrades, you would have easily won that battle, and ultimately the game.
The thing that lost you this game in all honesty, is your third timing. Your third should go down almost 12 minutes before it did. You were mined out in your main before this nexus was planted, and as a result you were unable to build enough stuff to defend his counter push.
Keep watching some Day9 Dailies and keep practicing a ton! When watching replays and observing others, Pay specific attention to the things that you are having trouble with. In your case, pay attention to how many and what buildings the player is making, and at what times. For example, if a player wants 6 gateways for the midgame, are they only getting 3 of those before their nexus, or maybe even 1? Keep making pylons and probes. Have a planned time that you want your third base to go down. Don't be afraid to put it down at that time! This was the HARDEST thing for me to learn when I first started playing. I was always scared to place the third nexus because I was scared it would die. Just put it down, and defend it if you have to. The exact parameters of how to do this will come with experience, but you won't learn how to defend the nexus if you never make one. You learn by failing, and you can't fail if you dont try
Anyways, you seem to have potential, and I didn't realize just how much bronze has changed since the early days of the game. The play exhibited in this replay looks like what gold league used to be like a year ago. This just goes to show that the community is getting better as a whole, and in order to get promotions you not only have to improve your play, but improve it faster than the average rate of the rest of the community!
I am only a platinum protoss player, but I could probably help out some if you are truely looking to improve your game. Message me or something on B.net if I'm ever online. Orestes.533 or Revelatus .723 (my ladder account) I've been busy as hell for the last few days with grad homework and stuff, but I'm usually on like all day xD
EPT
