On August 12 2009 13:44 tossinYoSalad wrote: Yeah, so I'm just a d+ protoss, but I can see your problem... Your build is all wrong, you need to start out with the stove.
On August 12 2009 11:32 Failsafe wrote: There are three gaping defects in this PvT.
Two Kwark has already pointed out. Your initial strategy is counter productive. You create the illusion of some kind of hidden tech by delaying goon range and then after you've convinced your opponent to play defensively you opt for 2 gate reaver instead of expanding aggressively.
The second is that you built an arbiter tech but only made one Stargate despite having ample gas for more. You need at least 2, and this is a huge leak in your play.
The third problem is an even larger leak, and probably explains most if not all of your difficulty in PvT. You chose to get arbiters but you don't have an arbiter until the 15 minute mark (despite the fact that you haven't been harassed at all). Your arbiter tech was roughly on time but you didn't DO ANYTHING with it. You had a Stargate and an Arbiter Tribunal just sitting around doing nothing.
Arbiter timing is ABSOLUTELY crucial to modern PvT because you HAVE to push the pace of the game if the terran (like this one) likes to turtle.
On the use of Arbiters in PvT: Upgrade the energy upgrade as soon as you have the tribunal. Wait a couple moments and start an arbiter (preferably two). This ensures that your first arbiters all benefit from the energy upgrade upon being built (and start with 63 energy).
Your PvT is too passive. You put all the control of the game into the opponents' hands as you have no way to efficiently attack him at any point.
To correct your play: Create a consistent early game that forces your opponent to do what you want, aside from that you play fine until midgame. After that you must time your arbiters properly so that recall becomes an asset as early as possible. If you opt for Arbiters it's absolutely essential that you have at least 2 before initially maxing out, and in most situations you should have at least another 2 in production. As soon as you max out several things should happen. You should lay down more gateways, though never more than perhaps 20 even on macro maps. You should begin probing intently as to your opponents' weakest points (and there must inevitably be weak points as you will max out ahead of him and have recall). And finally, once you are maxed out, have 15-20 gateways, and roughly 2,000 minerals you must commit to an attack that will do as much damage as possible and discombobulate the Terran.
With these simple rules in mind your PvT will improve drastically.
I guess I'll have to reply since some of you put a lot of thoughts into your replies.
I still don't quite agree with your first point. Even though it was counterproductive, it turned out just how I expected it to, it was a very safe opening that only set me slightly behind. I believe I accepted this and consciously turned it into a macro game. The late arbiter was a relatively small mistake that turned out to have worse consequences than I expected. It was really bad, and I usually don't miss things like that but hey, everyone makes mistakes. The passive play is something I agree with and I have a lot of problems because of this. I believe Ver addressed this flaw very well. In PvZ and PvP, I try to gather intelligence, send out scouts, corsairs, observers to see exactly what he is up to. In PvT however, it's as if I'm playing in the dark, and not just in this game. I don't prioritize to get an observer into his base and my observer usage is generally very low, which makes me unable to see if the terran has any weaknesses to exploit. I'm usually attacking blindly, so even if I understand the matchup well, I don't have anything to base my choices on. Because of this, I can deal with aggressive terrans a lot better than the passive ones (especially the ones who camp very well). But to my defense, it is at least to me a lot more difficult to scout high level terrans. With well placed turrets and early scan+goliaths you don't have any option to gather intelligence and you can't get a good look into his defense setup to see whether you should attack or not. It is even more impossible after he gets a vessel out and kill all of your observers. In the other matchups, you can always retreat after judging his army size without losing too many units, but if you attack a terran with a lot of siege tanks and make the judgement you won't be able to break him, you can't retreat until you've lost at least half your army which usually makes you lose the game right there. This is the main reason why I usually play a passive lategame in PvT...
You're right about the difficulty of using observers and gathering intelligence against high level Terran, and especially this style of camping Terran who makes it literally impossible to enter his base with an observer. On the other hand, I think that concerns about intelligence gathering and observer usage have been overstated in general.
The main priority of Protoss intelligence is to anticipate the Terran's attack by keeping track of the Terran army, and you did this well. It's great to know the Terran's factory count as well and all the other minute details of his base, but it's certainly not absolutely necessary at this level. All you really need to know is whether the Terran is playing aggressively or passively (and this guy certainly wasn't capable of mixing it up and tricking you).
Once it was clear that the Terran had opted to turtle you opted for 3 bases and an Arbiter tech. It's really worth emphasizing the point that this is a viable PvT midgame but you have to follow through with immediate arbiter usage and recall aggression. The immediacy of the Arbiters and recall can't be reiterated enough because the entire strategy depends upon it. Once you opt for a lower econ game (by not taking a fourth or even fifth expansion midgame) you have to play it aggressively. If you're not comfortable as Ver said "suiciding" armies with recall, then you can't really play a 3 base Arbiter midgame and should opt for a more macro-oriented approach.
On August 13 2009 01:59 Failsafe wrote: I'd say both games are between B- and A- level and feature better Terran than the one in the OP
No.
Gretorp and... Game?
No offense to Gretorp but why did you post a replay against Game? This feels more like a silly bragging post than anything else. And what do you mean by 'between B- and A-'? Why not just say between D- and Olympic while you're at it?
I still watched the reps though. Game didn't play this style at all, and while Gretorp tried to, he was far behind and even so went back to defend with all but two tanks against the recall. I don't intend to question the skill of you and your opponents, I'm sure it's at least B- games like you said. Yet I don't see the relevance to this thread. It's as if you try to tell me 'this is how you recall with arbiters'.
I dislike to generalize between korean and foreigners, but there is no non korean terran who can play this kind of camping style well. I never had this kind of problems against top non korean terrans on either of these two maps. They usually overextend themselves, panic against recalls, doesn't EMP well and doesn't scan army and arbiter movement well.
While I do think I am pretty good at PvT, I genuinely thought the replays were good examples of a meaningful difference in strategy. On the one hand, your late game was very passive and illustrated how an intelligent turtling Terran could effectively prevent Protoss from making a sufficient dent unless the Protoss had a cohesive, aggressive late game strategy. On the other hand I thought the replays offered some indication of what I thought that strategy to be, not that I executed it perfectly nor that my opponents were capable of the same level defense as Koreans. You can understand my surprise, though, to see a strong Protoss player losing a game with two armies worth of resources in the bank and no clear indication of any strategy for what to do with them.
A very nice read indeed. I, like many others, am also struggling with mid-to-late game PvT and this thread has greatly increased my understanding on the subject. Thanks to all who've contributed, Kwark, Failsafe and Shauni in particular
Also, this should be a featured thread of the strategy forum - a very relevant PvT discussion.
Maybe I was too harsh, you didn't have to remove them. There could be others who benefit from them. I was just saying I didn't and that I have a lot of similiar recall games.
On August 13 2009 06:21 Failsafe wrote: Alright, my bad. I'll remove them.
While I do think I am pretty good at PvT, I genuinely thought the replays were good examples of a meaningful difference in strategy. On the one hand, your late game was very passive and illustrated how an intelligent turtling Terran could effectively prevent Protoss from making a sufficient dent unless the Protoss had a cohesive, aggressive late game strategy. On the other hand I thought the replays offered some indication of what I thought that strategy to be, not that I executed it perfectly nor that my opponents were capable of the same level defense as Koreans. You can understand my surprise, though, to see a strong Protoss player losing a game with two armies worth of resources in the bank and no clear indication of any strategy for what to do with them.
I saw the replays and am curious about why you so rarely used storm - in the one vs game, I think, you never made templar despite having an archives and his tanks being clumped much of the time, and in both you rarely used it to engage the push. Just a stylistic prefrence or is there something deeper than that?
On a related note, Shauni, I saw you use shuttle storms effectively in that rep. How do you micro that exactly? I find if I leave the shuttle stationary I don't have the clicking accuracy to select ht and if I drop them on the move I have to either drop them one at a time and storm before returning to the shuttle and dropping the next or risk losing them all to vultures. In early game battles I can use them fairly well but as the scale of the battle goes up the amount of micro involved increases too (for the Protoss at least).
On August 13 2009 10:19 Kwark wrote: On a related note, Shauni, I saw you use shuttle storms effectively in that rep. How do you micro that exactly? I find if I leave the shuttle stationary I don't have the clicking accuracy to select ht and if I drop them on the move I have to either drop them one at a time and storm before returning to the shuttle and dropping the next or risk losing them all to vultures. In early game battles I can use them fairly well but as the scale of the battle goes up the amount of micro involved increases too (for the Protoss at least).
I used to have the same problem with the lack of clicking accuracy instead of improving my accuracy I actually hotkey each templar before putting them in the shuttle so when they unload you 5t6t i use 5 and 6 because they are so close to t it just seems logical
On August 13 2009 10:19 Kwark wrote: On a related note, Shauni, I saw you use shuttle storms effectively in that rep. How do you micro that exactly? I find if I leave the shuttle stationary I don't have the clicking accuracy to select ht and if I drop them on the move I have to either drop them one at a time and storm before returning to the shuttle and dropping the next or risk losing them all to vultures. In early game battles I can use them fairly well but as the scale of the battle goes up the amount of micro involved increases too (for the Protoss at least).
Yeah, I've also noticed that Shauni has very good shuttle micro (particularly for storm drops) and have been wondering if he (or anyone else) has any tips.
I just watched a stream of Shauni playing a PvP on Outsider and the shuttle control was exemplary.
On August 13 2009 10:19 Kwark wrote: On a related note, Shauni, I saw you use shuttle storms effectively in that rep. How do you micro that exactly? I find if I leave the shuttle stationary I don't have the clicking accuracy to select ht and if I drop them on the move I have to either drop them one at a time and storm before returning to the shuttle and dropping the next or risk losing them all to vultures. In early game battles I can use them fairly well but as the scale of the battle goes up the amount of micro involved increases too (for the Protoss at least).
Press U and click on your shuttle after you told it to move somewhere, it drops the temps while moving. As far as what failsafe said about factory count. It is incredibly important for probe cutting and when you start massing. When terran is movin up past 2 facs, you should be very alert to what he is trying to accomplish, whether it be 4 fac 3rd expo, or eventually a 5-6 fac rush.
Yeah I didn't mean to downplay the importance of factory count. It's essential to know what's gonna happen after the first two factories, but once you've seen that it doesn't matter much. Since Shauni had already seen what was going on in the replay (5 factories) it doesn't matter so much whether the Terran is getting 8 or 10 after that. Was definitely misleading on my part though
I was just meaning to address some commentary I've seen on PvT lately where players insist that Protoss needs to know not only how many factories, but how many armories, and at what time they start spinning and also needs to observe the Terran's SCV cuts etc etc. And of course while all this is useful, it's not in my opinion absolutely essential except in the most refined situations.
On August 13 2009 06:21 Failsafe wrote: Alright, my bad. I'll remove them.
While I do think I am pretty good at PvT, I genuinely thought the replays were good examples of a meaningful difference in strategy. On the one hand, your late game was very passive and illustrated how an intelligent turtling Terran could effectively prevent Protoss from making a sufficient dent unless the Protoss had a cohesive, aggressive late game strategy. On the other hand I thought the replays offered some indication of what I thought that strategy to be, not that I executed it perfectly nor that my opponents were capable of the same level defense as Koreans. You can understand my surprise, though, to see a strong Protoss player losing a game with two armies worth of resources in the bank and no clear indication of any strategy for what to do with them.
I saw the replays and am curious about why you so rarely used storm - in the one vs game, I think, you never made templar despite having an archives and his tanks being clumped much of the time, and in both you rarely used it to engage the push. Just a stylistic prefrence or is there something deeper than that?
It was mostly that my plans relied heavily on Recall, and I needed to constantly pump Arbiters in order to constantly recall. I think I had double digit gas quite often in the game against Game and so there just wasn't any room for HTs in my army composition as Psi Storm was secondary in importance to Recall. I could have cut Dragoons of course, but I think that's a pretty big error, especially on a low resource map like Blue Storm where you can't afford to gobble a lot of mines with Zealots.
I'd say it's mainly a stylistic preference, but it's one that effects pervasive changes to your master plan. Actually, the difference in preference for HTs vs Arbiters was another reason that I posted the replays.
Against a camping and relatively passive Terran who is looking to stalemate maps such as Destination or Blue Storm (which I think are conceptually quite similar maps) and then win by virtue of his more efficient army, I prefer more Arbiters to a mix of Arbiters and HTs. Against a more aggressive Terran who likes to take stabs and who I can catch off-guard, I prefer HTs in conjunction with Arbiters.
It's just so difficult to play aggressively against a defensive Terran by attacking them headlong which is what HTs equip you to do. Even with HTs, it's too often gonna be the case that you simply can't get enough of an opening against a camping Terran. In such a case, I strongly prefer the added mobility of Arbiters with Recall to the added headlong offensive capacity from HTs. Against the aggressive Terrans or counter-attacking Terrans, I'd say HTs are as good if not better suited to the task.
An additional consideration is that HTs are simply more tricky to use and require more micromanagement than Arbiters (as Kwark pointed out). Moveover, other factors in favor of Arbiters include cloaking which creates more opportunities for Terran to make mistakes. HTs, on the other hand, are quite the opposite in that they have to be used from a Shuttle and kept safely inside a Shuttle at all times. HTs essentially create more opportunities for you to make mistakes and are a major multi-tasking concern. Arbiters are healthier units that fly and so are not prone to Vultures, plus they're roughly as quick as your army which itself is a big advantage over HTs. In addition to Recall for mobility Arbiters also have Stasis which is a good substitute for Storm should you push come to shove. While as Bisu demonstrated, Hallucination is not a particularly great substitute for Recall.
On August 13 2009 10:19 Kwark wrote: On a related note, Shauni, I saw you use shuttle storms effectively in that rep. How do you micro that exactly? I find if I leave the shuttle stationary I don't have the clicking accuracy to select ht and if I drop them on the move I have to either drop them one at a time and storm before returning to the shuttle and dropping the next or risk losing them all to vultures. In early game battles I can use them fairly well but as the scale of the battle goes up the amount of micro involved increases too (for the Protoss at least).
It's okay if you lose them as long as they get off one or two good storms. I don't really have any secret and I probably do not use the most effective way, but I tell the shuttle to move a few tiles away while I click the portraits (not u because I sometimes want to decide the drop order). Then I just select them manually one by one and storm. It's often good to have a zealot dropped first because I like to move the shuttle in closer range than the rest of my army and then the first unit dropped will be killed by tank fire that doesn't reach anything else. The 'secret' is just to have very good mouse accuracy so you can storm quickly and return them to the shuttle.
(Bisu vs Canata Fpvod) At 17:31 you can see Bisu's fpview doing it, it looks like he uses the method skeptical outlined above, while using the wireframe click to unload the first one, and then just uses timing/magic boxes for storming. Seems really complex to get the position right, but I think that using that method if you unload the first one manually, it will unload all the rest in a predetermined distance from it (due to the cooldown), so you can drop them in a line equidistant.