|
So, my PvT is messed up. If the terran has less than 150 supply...I dominate. I don't remember the last time I lose a PvT remotely near my rank to a timing push without massively derping.
But let terran have that 200/200 army and I started getting scared of D+ Terrans.
Here is a classic example: Lategame PvT. The terran, despite deciding around 150 supply that macroing is for the weak, is around C rank.
+ Show Spoiler [Analysis] +From my point of view the game went fine till about 13 minutes. I don't *think* I have any major problems until then, but I could be wrong. After that, the money starts skyrocketing because I'm maxed and don't like attacking into the natural. This happens often, and I guess I should make it a point to just start adding gates + taking every other base on the map quite a bit quicker than I do.
I wait around a bit, he comes out, I attack, engagement doesn't go super well, I back off before everything dies. Then we battle it out for a while near bot right, I don't know what I need to do better to clear this area out (too sloppy? wait for bigger army?).
After that I just can't stop him, which I think was in part because I got low on gas b/c I decided I didn't like assimilators or something. My upgrades were terrible which I'm sure didn't help, though it doesn't ever feel like it makes a huge difference like it does in PvZ/PvP
Bottom line is, I don't know what I'm doing wrong. This seems like the kind of situation one should never lose in, but I pull it off wonderfully. I don't really know why, but I suspect it's some combination of bad control, lack of tech, lack of upgrades but I'm not sure exactly what it is to be able to pinpoint and fix it.
Unrelated note: terrans like this that are a little more "campy" are tricky for me. Not really sure what to do because they never give you an unsieged, relatively mine free attack. It feels like you become forced to engage in horrible engagements where terran is already sieged/mined/spread.
|
Well gas is important that's for sure. If he moves out with 1-2 and you have no upgrades, you better hope you have storm or stasis or both. At 13 minutes you should be maxed for sure. If you are maxed and don't want to engage, expand more and add more gates.
|
Upgrades bro, you have even forge and you dont roll any upgrade. On 5 bases u can add second or three.
He had 3-2 and you 1-0-0
|
United Kingdom3685 Posts
Being maxed is no excuse to let your money go high. Take another main and start massing gateways there. If the terran is super defensive on three bases, there is little you can do besides wait for him to come out and when he does come out, you will probably lose the first engagement as you have experienced. This is when you re-max in 90 seconds from the 25 gateways you now have. Essentially once you get to late game, you need to pretend to be zerg.
As a slight addendum: the more gateways you have in late game, the more money you actually want to have in the bank (once you're maxed). If you have 25 gateways but only 800 minerals because you just made 5 nexuses, you won't be able to remax from those gates if an engagement goes badly.
|
United States10253 Posts
maxed and bank rolling =
take another main and build 10 gates, get triple forge, get 2-3 stargates for pure arbiter pumping, and recalls everywhere. recalls is the best way to break a stalemate
|
From a former low D+ player,
mass stasis is nice. :D
|
As a former Terran player, I didn't know what to do when Protoss was maxed, had decent upgrades, 5+ bases, lots of gates, and statis+storm... I guess you should head for those things I mentioned, because I don't know how Terran can win after that point in short of splitting the map by half and defending recalls like Flash for 15 more minutes until minerals run out 
Anyway, for terrans that don't give you openings in his push, just recall on him all day long and wait until his mental stamina gives in.
|
i skimmed over the replay. the two major mistakes i noticed were preventing you from winning were #1, you kept attacking right into the terran force while he was in the best position possible and not threatening u, and #2 your geysers got depleted and you didnt replace them, meaning you had zero spell casters for much of the 2nd half of the game.
when examining how best to play PvT lategame, we need to look at the maxed armies of terran and protoss comparatively and their strengths/weaknesses. Terran max is pretty much impossible to beat in their strongest position, mined/sieged spread like you said. You can waste multiple maxed protoss armies into a fortified terran position and lose. Protoss max on the other hand, has a lot more flexibility and maneuverability (carriers, recalls, dragoons and zealots being faster than tanks and not needing to siege/mine in order to be at maximum strength). So this means that in order for you to play efficiently, you need to play guerilla warfare. Hit their bases, not their position, and then when they move to defend you can engage or run away. This is obviously hard to do when the terran is turtled on 3bases, but when the terran tries to take bases that are farther away, they spread themselves thin and thats when you can go on the offensive. Before that point, your objective will be to take as many bases as you can before they roll out, so that in order to take new bases the terran has to commit some of his force to remove you from them, weakening his overall position to your solidified strength. In order to win with a weaker force you engage with your many against his few (sun tzus art of war fuck yeah)
OK to put that back into context in your game, when he moved out with his maxed army, you did well to attack before he was in a solidified position. Then you remaxed off your many gateways, engaged again and again and again into his force while he slowly crept towards your bottom bases and took them out. What you should have done instead was counter his 3rd base, his main, his 4th base, anywhere where the majority of his maxed army isnt. Then you force him to unsiege and come to you, leaving his army vulnerable, so you have the option to engage with greater relative strength. If he doesnt defend you can simply trade bases and win because you have like twice as many as him. As the game went on, you shouldve been recalling multiple times, possibly teching to carriers (you had more than enough minerals), wearing down his force with storms, not just stasis. His 4th base was vulnerable to attack and was his only mining base for like 10minutes, if you shut that down you wouldve won easily.
As a last note i also saw that you had a shitload of probes. Like too many probes. You had idle probes in your main/natural for most of the game that you didnt need, and were limiting the size of your maxed army so that can help you win fights as well.
|
Mind control an scv and build a 400 supply army. Put tanks in key locations on the map (high ground preferably). Build dark archons and mind control his key units away from him (especially expensive units like battlecruisers and science vessels). Don't be modest with flower fields. Put cannons everywhere and then put some more. Remember, overwhelming a terran is one of the perks to out-macroing him.
|
Usually I build lots of gateways, then use recall to hit their bases and keep remaxing.
|
iNfeRnaL
Germany1908 Posts
Mass upgrade + mass gateways + arbiter OR
carriers + mass gateways + decent upgrades (at least on carriers.)
That's pretty much all you can do in lategame PvT ^^
|
try the 5 gate 3 stargate carrier transition off your 3rd base
ignore all these arbiter hooligans
|
You were doing fine way into 20 minutes.
Your macro is solid your micro is good enough.
Your upgrades lost you the game.
Consider a +1 zealot vs a tank:
zealot: 2x(8+1) dmg, tank: 4 armor
= 10 damage, that is really tickling the tank
And upgrades matter even more in goliath fights - with your upgrade difference goliaths are way stronger.
So double forge for you and keep up with upgrades.
Other than that you're doing pretty good. Though you may try to abuse recall more when your terran is spread across the map - in this specific game recall to main -> kill factories and later on recall to 4th on top of tanks.
Also try not to lose patience in longer game, because clearly I could see that your forces where not synchronized later on and your flank was not attacking at the same time.
And don't listen to the carrierists above posting. Carriers are only good at some maps and fighting spirit is not really one of them.
|
iNfeRnaL
Germany1908 Posts
Actually I think Terran armies can reach a point where you cannot beat them anymore without Carriers.
Even if you first have to get a few arbiters, yes the transition is hard but if you look at the "former pro scene" Flash's lategame was only beaten by Stork or Jangbi going Carriers.
Whenever Bisu or Best won, they did by gaining a superb advantage earlier and never letting Flash get as big as he likes to be. 
|
|
|
Solution to campy terrans: Carriers.
|
Thanks everyone for the responses, especially DragOn for the very detailed response. I'll work on making sure upgrades are out crisply, as well as your thoughts on mobility and where I should be attacking and come back for more if I'm still struggling after that.
@DragOn - when you say too many probes, what do you see as an ideal army size. (i.e, if I have an army of barely 4 control groups without 12 arbiters or something it's obviously to many)?
Also, as the game goes on do most of you continue to escort HT in shuttle, or later on do you just walk them around with the army?
|
Ive watched your replay.
Your build order was great and you are very good player who understands the style he was going for.
I think the things you needed to do in order to win this game were:
1. Add two more forges as +1 attack finishes/2nd stargate starts (and upgrade at all 3).
This is such an easy thing for you to add to your game and will help alot. Dont underestimate.
2. Attack into the terran army with more cohesion and a bigger army.
You maxed at 13 minutes and were in a very good position to win the game Everything had gone very well so far. (Didnt add forges though but thats an easy fix as mentioned already).
Fight 1
@16 minute mark; Terran was a little bit slower than you but eventually moves out. You decide to attack his army as he moves out and it worked very well. You actually would have won the game if you had pressed that attack imo but you retreat. However, vs stronger opponents that is the correct move so all is good.
Fight 2
@18 minute mark; Terran wins this fight.
Fight 3
@20 minute mark; Terran wins this one aswell. You left 4 arbiters behind.
Fight 4
@21 minute mark; Protoss wins
@22 minute mark; you donate 9 goons and then 9 zealots soon after.
Fight 5
@27 minute mark; Key battle! Zealots all die before reaching the tank line. 3-3 vs 1-0-0
Why did you lose some of the fights? You had too many probes meaning not enough army, and also not quite enough coordination in your attacks.
3. Expansion pattern could be changed. You could have taken the 12 o clock as your 4th and then the 11 o clock main as your 5th.
This thread is a rare example where Ive read a post by FyRe_DragOn and not fully agreed with it.
So this means that in order for you to play efficiently, you need to play guerilla warfare. Hit their bases, not their position, and then when they move to defend you can engage or run away.
I believe the best tactic in general is to wait as long as possible before directly engaging the terran army with stasis just before they start to damage your nexus. The first attack will be zeal goon stasis and the second attack will have 4 ht too. Its ok to poke and threaten but always be aiming to back off before losing any units or getting emped arbiters.
Before that point, your objective will be to take as many bases as you can before they roll out,
Yes I agree that you should take as many bases as possible, but the way you word it leads me to believe that you are suggesting any number of bases up to 8 perhaps. I believe the number is lower. This is assuming two pros are playing each other but what I want to make clear is that even if you are way better than the enemy, and have therefore got a nexus and cannons at every single base, that doesnt necessarily mean you should be mining from them all.
What you should have done instead was counter his 3rd base, his main, his 4th base, anywhere where the majority of his maxed army isnt.
At one stage L_Master does actually attack the 4th but anyway..This is extremely risky and verging on impossible on FS. Its also a style all of its own that needs different preparation. You anticipate losing your new main and nat and so you get less gates there and start making a nexus at the other corner sooner. Personally I think the generally optimal strategy is to kill the army as the main priority. Backstab styles exist ofcourse though. L_Master seems to be a player who prefers management though and therefore killing the terran army is the best. But only when it actually gets to a position on the map where you are forced to engage.
As the game went on, you shouldve been recalling multiple times
The way this is worded leads me to believe you mean recall on mining bases or the main rather than on the enemy army. I think that if a Protoss is going to start using recall in a backstab tactic after having engaged the Terran army several times with stasis first then its normally because things didnt go perfectly to plan. Teleporting to an enemy base is always risky and this risk is even greater if you start the game with a stasis management style.
So, these are things I disagree with.
However, I do agree that L_Master made too many probes and his army was small as a result. This is a Major issue.
The three things I believe lost you the game L_Master are:
1. Expansion pattern.
2. Upgrades
3. Army size.
1. Expansion Pattern. You spawned cross position on a rotational map. Terran will always be able to take 3 bases. The threat of recall means he has to stop there. When terran moves out with 2-1 he will attack a natural. Soon after he will take his 4th. You should expand so that the terran army is either protecting his 4th base OR attacking your base. However, in this game terran is attacking you and is simultaneously covering his 4th because you chose to expand in that direction.
2. Upgrades. Your build order was really excellent up to the point you maxed out at 13 minutes. You need to add a little bit on to your build order at that point. 2 more forges! and make it a priority to check all 3. As previously mentioned by others you were +1 attack only for the whole game and this really showed in the fight at the 27 minute mark.
3. Army size. This is extremely complicated meaning I dont have the perfect answer yet (the reason being the switch to starcraft 2 has cut off the flow of games at the top level). Your game is the perfect example of what the professional Protoss players experienced around 2011. Having twice the number of bases compared to the Terran is nothing if you cant do damage. The best style was to get arbiters and stasis to limit the dps of the terran army and then attack it at the last moment before he starts killing your nexus. This still holds true ofcourse but a few things have changed. Free and Shuttle were the masters of the style at the time and were very good at this Absorb and Management style of play. However, Terran slowly developed their attacks so that they quickly killed the natural and the third, and ignore the main. They would then fairly rapidly attack the other natural and leave only some army behind forming a weak contain that would be as cost effecient as possible.
What happened is Protoss found they were able to re max once from their 20 gates but in the mean time they had lost 4 mining bases and crucially all the probes too. Terran only had to survive a couple of waves to win.
Protoss have developed understanding of a few things during late 2011 and 2012 in my opinion.
1. The risk/reward intricacies between many bases with low saturation or powering on 3 for slightly longer.
2. The intricacies of stasis/DPS ratio.
3. The true value of geographic distance between bases and the true value of empty mains depending on start locations.
4. Recalling reinforcements both in a defensive manner and also in an offensive manner but directly on the enemy army rather than backstabbing recalls.
5. General intricacies of safe transitions, the exact limits of expansion timings, limits of necessary defense, and details of attack windows.
The more recent PvTs generally feature:
Longer period of time on 3 bases.
Later 4th and 5th nexus, and even later still to actually start mining there.
Later to take a 6th nexus and content to sit on 6 bases maximum.
Faster 3 forges.
4 HT after the first stasis attack and 4 ht in every new army after.
Offensive recalls on the terran army in order to deliver 2 waves in short period of time.
This thread needs links to some examples or pro games. I cant find the perfect one yet.
Im afraid this is rather rambling but its taken long enough so Im going to stop there for now and come back later maybe.
Please come forward and post which bits you think I am wrong about even if you cant be bothered/ dont have the time to explain why. I would like to see opinions.
|
let me clarify some things in response to L_Master and Miwyfe.
@DragOn - when you say too many probes, what do you see as an ideal army size. (i.e, if I have an army of barely 4 control groups without 12 arbiters or something it's obviously to many)?
You should have approximately 5 full control groups of units when maxed - 3-4 arbs, 4templars + shuttle, about 20goons and 30zlots. This army is intended to crush a maxed terran army that is mostly comprised of tanks. When engaging, wait for the terran to roll out into a vulnerable position and then move in with arbs for stasis first before the rest of your army, both to nullify science vessels which can similarly rape your army, and prevent all of his tanks being able to fire their first shots on your army. (stasis vessel and tank clumps)
At one stage L_Master does actually attack the 4th but anyway..This is extremely risky and verging on impossible on FS. Its also a style all of its own that needs different preparation. You anticipate losing your new main and nat and so you get less gates there and start making a nexus at the other corner sooner. Personally I think the generally optimal strategy is to kill the army as the main priority. Backstab styles exist ofcourse though. L_Master seems to be a player who prefers management though and therefore killing the terran army is the best. But only when it actually gets to a position on the map where you are forced to engage.
At this point you should have expanded to 5bases, working on a 6th with about 24gates. When the terran moves out, you stop expanding if you are engaging the terran army, because youll need the money to reinforce units instead of building more bases. If you dont engage the terran army because hes not really attacking, hes just setting up a position, you can continue expanding and then proceed with the guerilla warfare I advised earlier. Use that at any point where the terran has a solid position that you cant directly break. If you can break it though, because hes being aggressive which means moving towards you and unsieging/moving out of his minefields, then absolutely go and crush him. In the replay you gave, you were unsuccessful with your initial attempts to do this, so guerilla warfare is necessary at that point because he had an established position that you needed to remove him from by dictating his movements, not wasting armies into it.
I believe the best tactic in general is to wait as long as possible before directly engaging the terran army with stasis just before they start to damage your nexus. The first attack will be zeal goon stasis and the second attack will have 4 ht too. Its ok to poke and threaten but always be aiming to back off before losing any units or getting emped arbiters.
imo this is a bad strategy because by letting the terran that close to your base before engaging, when you lose the first battle which you inevitably will, there is no time to gather reinforcements and you automatically lose a base and get contained. I also disagree with building templars only on the 2nd round. If you built them that late, you still have to wait 30s to have 1 storm each, at if the terran is already at your base hes already set up a contain and spread tanks by that time so storm is not nearly as effective as it could be on the open field.
Yes I agree that you should take as many bases as possible, but the way you word it leads me to believe that you are suggesting any number of bases up to 8 perhaps. I believe the number is lower. This is assuming two pros are playing each other but what I want to make clear is that even if you are way better than the enemy, and have therefore got a nexus and cannons at every single base, that doesnt necessarily mean you should be mining from them all.
yeah i wasnt really clear on that. 6 is all the bases you can afford before the terran is maxed, after that priority is on pumping units from all gateways and not dying if the terran attacks. If hes not attacking you can expand to more bases right away. You should have max mining from only 4bases at a time, more than that is too many probes. The way it usually works out though is that your main is depleted soon after taking the 5th base, and the natural depleted to saturate the 6th base, so only 4bases will be mining at capacity at once.
The way this is worded leads me to believe you mean recall on mining bases or the main rather than on the enemy army. I think that if a Protoss is going to start using recall in a backstab tactic after having engaged the Terran army several times with stasis first then its normally because things didnt go perfectly to plan. Teleporting to an enemy base is always risky and this risk is even greater if you start the game with a stasis management style.
yes i do advocate recalling on mining bases, specially new ones that are not defended with 1000 turrets and mines. In a perfect game for protoss, he would win the 1st/2nd maxed engagements easily and stop a 4th base from going up just through superior engagements and macro. However this doesnt always happen, because terran max is stronger than protoss max in a straight fight. I do not agree with recalling directly on a maxed terran army, because of science vessels. any good terran will EMP the shit out of your arbiter by the time it gets directly over the tanks. Stasis is effective because its range is equal to EMP range, so you can stasis a clump of units just as your arbiter gets EMPd. definitely a good trade for protoss. Terran can also do the same to clumped arbiters/units so its best to hide the arbiters and then move in to stasis before you engage with your army. Recalling an enemy base is much less risky and more rewarding than recalling on top of a powerful terran army. In the pro games where protosses do this, its BEFORE science vessels are present, in an offensive capacity, or when science vessels dont happen to be around for some reason.
|
I agree with everything in the above post by FyRe_DragOn. Absolutely quality advice for modern PvT (2012 Korean pro style).
30 zealots+20 goons+4 arbs+4ht and shuttle=126 population
Therefore 74 total probes.
Assuming 9,7,8 patches at main, nat, and 3rd and three on each gas:
It equates to 2 per patch (18+14+16) plus 9 for gas with 17 left over for your 4th base.
18+ 14 + 16 + (9 for gas)=57
74-57=17
Therefore 17 probes at your 4th base. Obviously in a real game you would be mining with slightly more than 2 per patch while still on 3 base for a while and you might be mining with less than 2 per patch when on 5 base etc. But thats a great start point for us.
Edit forgot observers:
Add 6 observers to army size so 132 population.
Therefore 68 probes total.
|
|
|
|
|
|
EPT
