|
A couple comments from the 3 games I saw.
If I go a gasless FE, I try to wall off before the probe gets in. You cannot hold a protoss 1 base immortal all in if executed properly. The one game, you were obviously fast expanding and didn't scout the protoss all in. You did have your scv at his natural so you probably had some idea it was coming, but you didn't know what type of all in it was. I think you are a little bit too greedy.
You need to group your ghosts separately than your main bio/medivac army. The one engage I saw with the 2/3 ghosts, medivac, and bio was not good. You didn't use all of your emps. There are tons of micro trainers and unit testers out there you can practice on.
I also don't like the timing in your attack. I think you waste too much time teching to ghosts for your first attack rather than just punishing the weak early game protoss army. As someone said earlier, the low tech vs low tech fights are easier to handle. I generally try to hit a 10 min timing push with a large bio + 2 medivac force with stim/shields/conc shells off a fast expand. It generally will hit right as the protoss are teching to charge or colossus. If for some reason they have fast colossus, you just pull back and already have a starport. If they have a bunch of chargelots and sentries, you can punish them by constantly dropping without much threat of losing your medivacs since they wont have many stalkers.
|
High masters, TvP is by far my best matchup, won most Ps in EU at one point or another. Watching now. Until then, about your points -
On October 11 2011 09:45 dschneid wrote:
1. I am absolutely terrible at engaging. I have a hard time managing the army, because separate hotkeys for ghosts, vikings, and MM are too difficult for me, and emp takes priority over stim and that usually results in me fumbling around to get my stim off. I can't dodge storms, and have a hard time hitting HTs with my ghosts since they keep them in the back and I don't get cloak for awhile...
2. I have poor decision making, I'm not sure how to punish the P, even if I come out ahead, which is rare. Furthermore, I don't really have a great sense of timings, usually I look to 2/2 finishing, or just max 200, or maybe I scout some tech like storm or what not and I try to beat it.
1. Either practice -alot- with different control groups (even vs AI, just prac with moving around, EMP-ing, etc). or just use a single control group (even though that kinda sucks with vikings) and get used to tab-ing. In any case, practice with the control in environments that are comfy to you.
I assume you don't play too much, so you probably end up in that situation about 3-4 times / day - you're obviously gonna be nervous. Just practice - spreading in a concave, emp-ing, tab-stimming, moving bits of army around - 10-20 times until you're happy with how it goes.
About dodging Storms - it's kind of a thing about seeing a templar moving towards one side of your army, grabbing that chunk and pulling it back. Also, make sure to try and scan the observers above the place you're engaging.
2. 2/2, punishing greedy expanding, just before a tech switch etc - are all really strong moments to push. Key is to not go berserk mode - if you wanna push, see that you can't (ie, he has better concave, good ht spread etc), just pull back and continue macro-ing.
LE:
You said you're fine with your macro, so about the engagements:
G1 (shakuras): In order of badness -> No concave, 15 vikings (?! - at least preland them), no EMPs, 2-1 vs 2-2.
Also, 75 SCVs with the style you play is a big no-no. The only way you can pull off 75 SCVs in TvP is if you play a "constantly trade" style with mass expos and mass barracks (style which isn't particularly strong on shakuras but w/e). Something similar to zerg muta/ling, if you want.
With how you play it, when you get to maxed, you -HAVE- to build "macro OCs" and go free at least 20 SCV supply for more army. It's the only way to keep up with Protoss' insta-reinforcements via gateways.
G2 (abyssal caverns): You're 1-1 vs 1-3, but still, you should never lose that. You hade units by the xn tower, and out of the 4-5 available EMPs, you hit, well, zero. I assume you're extremely nervous when you engage, I think that's one of the main things that is pulling you back. Just relax, scan so you know where the HTs are, and take it easy from there.
G3 (abyss 2): You should never really lose that. A) pull your SCVs faster and surround bunker with'em. B) When he walked up the ramp that was lolgg for him. Close depot and win. It was just mismicro on your part.
G4 (antiga): Again, mismicro. Just stim and emp as soon as he comes down that ramp, and you win by a longshot. Good concave there. Also, land the vikies as soon as you get rid of the collossi.
--
Yeah. Bottomline. Work on the engagements. Against AI, against friends, just on ladder, your control is really kinda iffy. Take a breath, get comfy with it, play relaxed and you should do fine.
|
On October 11 2011 20:48 n0ise wrote:
I assume you don't play too much, so you probably end up in that situation about 3-4 times / day - you're obviously gonna be nervous. Just practice - spreading in a concave, emp-ing, tab-stimming, moving bits of army around - 10-20 times until you're happy with how it goes.
About dodging Storms - it's kind of a thing about seeing a templar moving towards one side of your army, grabbing that chunk and pulling it back. Also, make sure to try and scan the observers above the place you're engaging.
Yeah. Bottomline. Work on the engagements. Against AI, against friends, just on ladder, your control is really kinda iffy. Take a breath, get comfy with it, play relaxed and you should do fine.
This! You don't need to do mega drop damage to beat toss, thats a myth. What you need to do is know how to micro the battle.
First -> scan like crazy, you need to get your ghosts into position where you can blanket emp the toss army and you do this with cloak or just by getting the better postion. Then as soon as you blanket emp him, stim and engage, try to have a big concave if possible but its not always that easy. Obviously most important targets are hts, but getting the archons down to 0hps is pretty key as well if he has a few in his army. Try to snipe collosus with viking before engaging, this is not that hard once you have like 10vikings but its not essential since if u emp the shit outa the collosus then they should fall quickly to vikings.
The next trick is in the unit micro, and this is SUPER important.
You don't want to be performing the "standard stutterstep" where you just move your whole bio ball back shoot back shoot. This is what causes you to die a horrible death in TvP. Instead you want to be moving back small groups of infantry making your concave bigger and bigger preferably out of the most damaging areas (storms, where the collosus are firing). Its kind of similar to splitting vs banelings mixed with stutter step I call it spread stutter step as you basically just do one stutter step for each small group trying to make as big a concave as possible. If you don't do this you will be storm/collosus food and just die horribly in the late game, but with it, emps and at least even upgrades, your army can go toe to toe with protoss. With some practice you will just be doing this rapidly during battle and when a storm goes down you will naturally choose that as the group that needs to move back next.
As for timing, I like to go a bit after I take my 3rd around 160food cuz typcially he cant have a strong army of collosus and HT by then he can only have a few of each or a bunch of one.
|
my tvp is also weak... I think most of it boils down to unit control. I had a hard time with both sides of the matchup (when i used to random) and i still have a rough time late game against protoss. I fail miserably at confronting a deathball of protoss... so I try to be aggresive and trade units with the protoss all game long until they get out collosi to keep the unit counts lower (which makes the control easier)
|
Thanks all for the advice, I've got a new BO to try as well as a few key points.
I think the main takeaways are:
1. Make really sure I'm doing well with upgrades, I think I do a decent job but could be better
2. Try to implement some more dropping, particularly being annoying with the third to keep them somewhat gas starved. I tried this recently in a game and noticed it went really well, but then macroing behind this was tough and I forgot to really get my upgrades...ultimately losing, but the drop play was nice lol
3. Work on more favorable engagements (this is the really tough one).
It will probably be a lot of work and I'll likely get demoted in the process but I guess that's the way it is.
|
On October 11 2011 11:37 dschneid wrote:
I agree with most of this but have a few questions/comments
2. I definitely think my drop play needs improvement, you will see in other games if you watched those replays I do some drops and some of them are successful and some of them aren't. The thing is, I feel that in general these tend to be a wash, sometimes u lose all your units, sometimes you snipe some nice tech, it all depends on how prepared the toss is. Furthermore, even when you do do a "successful drop" you probably kill a few probes, etc, but I never feel like anyone attributes their win to the drop that happened at 10 min etc. This is not only from my own play but from watching many many streams of pros.
Aren't you supposed to lose nothing with drop play unless he has blink?? Imho if you see the stalkers you can get out there in time. You're right that sniping a pylon or a couple probes doesn't seem like a big deal, but firstly it buys you time to get up the right tech(wether it be ghosts or vikings). Plus it creates an economical advantage for you.
Obviously it's not the same, but imagine getting supply depots and scv's sniped all game long.
|
Blazinghand
United States25558 Posts
On October 12 2011 02:44 dschneid wrote:
2. Try to implement some more dropping, particularly being annoying with the third to keep them somewhat gas starved. I tried this recently in a game and noticed it went really well, but then macroing behind this was tough and I forgot to really get my upgrades...ultimately losing, but the drop play was nice lol.
Bear in mind that drop play can be a success without actually killing pylons or denying bases-- as long as he knows you have dropships in his airspace, he can't move out, and feels a lot of pressure. You can assert map control and cover for your third base using a couple of dropships if you don't lose them and the units inside them, and punish him if he tries to move his units out of his base.
|
sure, I guess the point I was making was that even though you're not SUPPOSED to, let's face it, we all do at one time or another, I see it all the time on pro streams.
Also, I felt like in my match ups the battles were so overwhelmingly one sided that I found it hard to believe it was the result of lack of/poor dropping that made the difference, there are even some replays there where i have no drop play and snipe a robo bay and still lose, but I'm willing to make a better effort at drop play given the strong support I've seen from it in this thread...
|
On October 12 2011 03:04 dschneid wrote:
sure, I guess the point I was making was that even though you're not SUPPOSED to, let's face it, we all do at one time or another, I see it all the time on pro streams.
Also, I felt like in my match ups the battles were so overwhelmingly one sided that I found it hard to believe it was the result of lack of/poor dropping that made the difference, there are even some replays there where i have no drop play and snipe a robo bay and still lose, but I'm willing to make a better effort at drop play given the strong support I've seen from it in this thread...
The thing is that continuous dropping hurts a Protoss. He can't freely take expansions, can't move out with his full army, and most importantly has to keep running Probes, putting 3 back on each gas, etc. This causes a lot of macro mistakes and thrown off timings. If he's trying to micro against a drop/run Probes, he might, say, miss a chrono on his Forge, or forget to warp-in for ~10 seconds, or gets supply blocked and can't get another Colossus out for 30 seconds. This ultimately makes it easier for you to out-macro him.
Protoss has a second wave advantage with Warpgates similar to Zerg's larva stockpile in the late-game, so be sure to make way too many Barracks once you're maxed and on 3+ bases, so you can have a reinforcement wave just as strong (even though it finishes ~30 seconds later).
Big engagements are won by splitting the Protoss up similar to how Zerg needs to fight Terran. You're faster but weaker in this match-up, you can't bulldoze Protoss lategame.
|
Blazinghand
United States25558 Posts
Oh yeah there's nothing quite as fun as getting like 10+ probe kills or sniping a tech structure with a drop <3 and that's a huge success-- I'm just saying that you can be successful without it.
The drop doesn't really change a one-sided fight into an even fight or a victory, either-- if you were gonna be crushed by a deathball, you were going to be crushed by a deathball. The purpose of a drop is usually something like "I want to be able to take my third, and I want his army to cower in his base for a couple of minutes" rather than "this is going to reduce his army's food count in a meaningfully direct fashion" because a drop is just a 10-food harassing unit.
Rather, a drop is more about affecting macro play and map control.
To think about how drop play works, think about mutalisks in ZvT-- zerg will use mutalisks to try to keep your marines in your base and force either units to stay at home or for you to make turrets instead of additional marines, thereby letting him secure his bases. The mutalisks focus not on dealing damage (though that's a goal) but primarily on not dying so they can be a part of the zerg army later. Bio drops in TvP should act the same way. You're trying to assert map control and trap him in his base for a bit. Losing a couple marines etc is fine, but ideally your drops stay as alive as possible so they can rejoin your army for fighting later. If nothing else, keeping the medivac alive is important since that way you don't have to cut a viking to rebuild it later.
|
Go 3 rax. 1 techlab, research stim asap. Marauders/marines/5-6 scvs to melee. Kills pretty much most FE builds. If they're on 1 base you're also a bit more safe. That's 60-70% win rate right there.
|
Blazinghand
United States25558 Posts
On October 12 2011 05:20 vnlegend wrote:
Go 3 rax. 1 techlab, research stim asap. Marauders/marines/5-6 scvs to melee. Kills pretty much most FE builds. If they're on 1 base you're also a bit more safe. That's 60-70% win rate right there.
I believe you're suggesting a 3 rax stim all-in off of 1 base, although if those latter 2 rax lack addons, this could still get an expansion in a reasonable time. If you're suggesting a 3 rax stim allin, I must disagree-- it is eminently holdable by a protoss with good FF micro, especially given the stim time nerf. If you're trying to do a pressure build like this, I'd suggest a reactor on the 2nd rax rather than a 3rd rax, since that will let you bank up to expand but give you similar production capacity as well as a still fairly powerful poke with marines and marauders.
|
On October 12 2011 05:31 Blazinghand wrote:Show nested quote +On October 12 2011 05:20 vnlegend wrote:
Go 3 rax. 1 techlab, research stim asap. Marauders/marines/5-6 scvs to melee. Kills pretty much most FE builds. If they're on 1 base you're also a bit more safe. That's 60-70% win rate right there. I believe you're suggesting a 3 rax stim all-in off of 1 base, although if those latter 2 rax lack addons, this could still get an expansion in a reasonable time. If you're suggesting a 3 rax stim allin, I must disagree-- it is eminently holdable by a protoss with good FF micro, especially given the stim time nerf. If you're trying to do a pressure build like this, I'd suggest a reactor on the 2nd rax rather than a 3rd rax, since that will let you bank up to expand but give you similar production capacity as well as a still fairly powerful poke with marines and marauders.
No, the 3-rax stim/conc timing is still extremely powerful. I suggested it earlier in the thread, and even posted some replays. The reason it works is because protoss are expecting a maruder expand after they see the first barracks with tech lab, and often get to greedy. I have about a 60% winrate with the 3-rax stim, and I'm a 1500 pt master. The key is to stim up and get right up close to the sentries and stalkers if they have no zealots, and if they do have zealots, make sure not to kite too close to the sentries so you can't get forcefielded out. If you micro well enough, no amount of good force fields can kill it, if they are too greedy. Its also good against the 4-gate and 3-gate robo, and has a 50-50 chance against void ray all-ins and DTs.
|
I was going to make a thread just like this myself, except I'm Diamond. The weird thing is I don't think I really have problems with the things you say you do, that is I feel pretty good controlling my army and all of that. But still, TvP is giving me hell (TvZ is too). There just doesn't seem to be a way to get an advantage and even if you do it doesn't matter. I've played games where I was 2-3 expo, 20-30 harvesters, and 70 total food ahead and have still lost. I've tried dropping, and have ahd some success with the drops, but the outcome is the same. Two big armies go up against each other, we trade or I lose, and even though I have a much bigger econ and twice as much production, I just can't come back. It's driving me nuts.
|
|
|
|
|
|
EPT
