|
Two things were clear 1-2 months ago:
1. Archon toilet became a standard in late game PvZ
2. Sooner or later zergs will learn to spread their broodlords. And then P will be in trouble.
If we look back at the time when archon toilet was transitioning from a gimmick to standard, there was a great post by NrGmonk: Lessons learned from Brown vs Losira
#6 of his list was "Mothership tech should be the norm versus broodlord tech"
Clearly, Brown was timing all his tech so that he would have the mothership and archon/templar at the same time. He was aiming for archon toliet+storm as his counter to mass broodlord, which I believe is the absolute safest and solid counter. Any ground based army just simply will not work. Voidrays are countered too easily by infestors + corruptors, and 1 or 2 carriers will not make enough of a difference in a fight as one single mothership.
The question that wasn't asked at that time was "What if zergs spread their broodlords?" I do understand, that BL is a slow moving unit, and group of broodlords tends to clump up... well, kinda like templars. But in PvT protosses learned the hard way that they desperatly need to spread their templars to not get hit by just one EMP. Terrans responded by building ghosts in bulk, but if we go back to PvZ, protoss simply cannot spam vortexes! All he can do, is to wait (quite a lot) for mothership to build, and then wait even more for 200 energy.
I wrote about Dimaga vs JYP game recently, but people were quick to dimiss it: If you think turtling with 30 spines in an armpit expansion is a solution to combat mothership, life is indeed very hard as a Zerg.. Fact is, Dimaga did pretty bad job splitting his broodlords in that series overall. But today we had incredibly frustrating (well, for me, as a protoss players and Grubby fan) series between Grubby and Stephano at ShoutCraft Invitational #4. Game 1 showed how good Stephano is at spreading his army versus vortexes. It also showed how hard for protoss is to flank late game zerg army on Metalopolis.
So, what is the next step in developing late game PvZ? Many people on forums and reddit, including NrGmonk, pointed out that protoss should switch to carriers late game.
Near the end of the game, Brown begins adding 3 stargates for carriers to add to his huge deathball. In fact, he even sacs probes to do so. The fact is that while a maxed mothership/colossi/templar/archon/stalker deathball is near unbeatable, that deathball with carriers IS unbeatable. The only counter to carriers zerg has is corruptors, which are completely negated by vortex and storm. Yes, carriers are hard to get, but if you mothership/colossi/templar/archon tech, I’m sure you can afford carriers. Why not voidrays you might ask? The fact is that voidrays are much easier to deal with than carriers in that deathball, because while corruptors don’t do as well versus voidrays, fungal and neural do very well, and infestors are much scarier than corruptors
In game 3 Grubby had a very good advantage coming into late game, so he decided to transition into that unbeatable mothership/archon/colossus/stalker/carrier army. He hidn't show his carriers too early, so when the big engagement happened, Stephano didn't have too many corruptors. So what went wrong?
Again, when the final battle happened, Stephano new about mothership threat and spread out his forces. He also had a high amount of infestors, and used them well. So after the dust has settled... not much left of protoss deathball, and Stephano was quick to build a dozen of corruptors to shut down carriers, and Grubby's fate was sealed.
So, the phrase "corruptors are completely negated by vortex and storm" is not completely true. Actually, carriers have very bad synergy with vortex and storm, since interceptors are easy to get stuck in a vortex and damaged in storm.
And therefore the question remains, what should be a new development in late game PvZ? Is it more active harassment? Using mothership for recalls instead of vortex? The thing is, in a famous Kiwikaki vs Stephano game, Kiwikaki was very good at sniping expos and using recalls, but in the end he still needed to battle zerg's "deathball", and he managed to get the win with 2 very good vortexes (in fact I think that was one of the games that showed to me and many others, that archon toilet is not just a Kiwikaki's "thing", but a solid way of countering late game zerg force)
Another concern of mine is what will happen in HotS. Protoss will not have mothership anymore, but supposedly they will use Tempest for big air-to-air area of effect damage. Obviously, the intention of Blizzard is to make them a counter to mass muta, but mass broodlords and corruptors should also eat a lot of these big balls of aoe doom, right? Well, there are 3 things that make me not so optimistic about tempest vs corruptor/broodlord:
1. Zergs are already taught to spread their air units
2. Tempest will be a massive unit. Corruptors will probably still have their strong anti-massive damage to counter colossi
3. Infestors will still be hard to deal with
Perhaps the focus should not be "how the protoss deathball should deal with zerg deathball", but more "how to prevent zerg from getting the control of the whole map instead of turtling to 200/200". For example, Hero's style incorporates a lot of warp prism harassment to keep zerg busy. The thing is, warp prisms get figured out as well, and zergs are more actively building static defense at their expansions and denying drops (again, today's Grubby vs Stephano series showed that very well)
 
|
|
|
It's very interesting how pros still cant spread their units for some reason. it's such a menial, basic task that would save them the game. I really hope tempest has some ability/abilities otherwise it's clear they're not even trying to stop it from being a 1a unit. I also wonder whether the new thor will have the same role for terran as it did for protoss. Maybe a late-game mech deathball will have it for some purpose. I forget the abilities they said it has, aoe stun I think.
Maybe they should make mutalisks turn into corruptors, like they could turn into devourers in bw. At least that would slow down anti-massive production, but then again it would be hard to deal with colossi. Then again mutas are proving to be a good unit in zvp.
|
YAAAA i was quoted. Life completed. I think the next step for Protoss late game is mobility play. Just like mass medivac drop as Terran when BLs come out, Protoss can easily emulate that with recall, warp prism, and blink. Any mobile army can easily out maneuver BLs by hitting left and right, that's why it's so hard to spread them out. If you need to constantly move the slow air units in many directions, you either ball up more or never get there in time. I don't agree that you can say phase prisms are 'figured out', that's like saying Zerg mobility play is figured out, which can never be true because it's a technique integrated as a play style.
|
Excellent write-up. The future looks grim for PvZ, I agree with your perspective on a solution. I myself am trying to prevent these ultra late game scenarios where the map is split in half which has not been a great problem on ladder. But in the pro games it seems that the MC-esque style of approaching the match-up by 2-base timings and rapid transitions is the most effective way to prevent the infestor/broodlord doom ball.
|
United States8477 Posts
I only watched the last 10 minutes of game 3 of this series, but Grubby's biggest mistake in the late game was forgoing templars completely. His colossi count was also quite too, I believe 2-3. In addition, he didn't abuse the range of his carriers and instead pretty much just a moved his army into the broodlord army. Because of all these reasons, a combination of fungal and mass infested terrans were able to clean up a lot of the carriers, something that shouldn't happen normally. If you look at Grubby's resource count, he was easily able to afford templar, as his resource count was something like 4k/2k even after mass carriers and double air ups.
Also, Dimaga vs JYP showed one situation where carriers are a necessity.
|
I remember the thread you're referring to and I actually asked the question, "What happens if zerg spreads his units".
The problem is that on some maps its just not feasible to split up because your infestors can't cover what they need to cover. On some maps you can and the BL/Infestor deathball is incredibly potent, just like the protoss deathball is.
Another question I asked is, "why are protoss players entitled to a 100 energy spell which can kill infinite units when used with archons". Blizzard answered that one for us, the mothership's days are limited.
The fact is, You can't declare one race superior in the late game over the other because of the myriad of other things that can happen before that. The unit choices, the unit losses, the upgrades, the tech patterns, the economy, the mis-micros, the missed macro, all these things add up.
Mobility play is really the counter to brood lords. I recently veto'd shakuras plateau because I like infestor/brood in ZvP but blink stalkers are almost impossible to defend against on that map. The architecture of it means that literally EVERY forward wall and cliff, is in danger from both air and ground attacks. Terrans just start dropping marines everywhere once BL's come out, Protoss should drop DTs and blink stalkers around. This is how it should be; that's how you counter the standard protoss deathball or terran mechball.
|
You can negate the majority of the damage from Archon toilets with Infested Terrans spread around the Vortex. I don't understand why people don't look for solutions, even if they are unorthodox, instead of dismissing it as impossible to defend.
|
On January 16 2012 01:50 Sated wrote:
Grubby had a massive advantage in that last game on Shakuras and failed to push it. With Colossus/Blink Stalker he could've busted through the Spine Crawler wall and dealt more damage, forcing Stephano to commit more resources to his defence. Alternatively, he could've done a lot more with Blink Stalkers up and down cliffs, since he had Colossus available to give high-ground vision. He instead choose to sit back and paid for it, because he gave Stephano enough time to get a "300-food" death-ball (in that Stephano could instantly re-max his army multiple times) that rolled him over.
Well his desicion making was far from ideal, but one thing was shown pretty clearly in that game once more: you shouldn't build carriers, ever.
So many people (including some smart people I might add) said that protoss should try using carriers in the late game PvZ, I myself was thinking about how I should add carriers in the late game PvZ. Well here we have a situation where protoss had a distinct economical advantage before the battle, he seemingly didn't do any obvious blunders going into it, had a decent number of carriers, but still lost the engagement pretty bad. So yeah, those carriers, strike them out, done.
Carriers are also often called as a unit that is good for finishing a game after you've got a good advantage (like in the game I referenced in the begginning of my post) -- but really, if you almost won the game anyway, what composition would not be good?
Edit:
On January 16 2012 02:23 NrGmonk wrote:
I only watched the last 10 minutes of game 3 of this series, but Grubby's biggest mistake in the late game was forgoing templars completely. His colossi count was also quite too, I believe 2-3. In addition, he didn't abuse the range of his carriers and instead pretty much just a moved his army into the broodlord army. Because of all these reasons, a combination of fungal and mass infested terrans were able to clean up a lot of the carriers, something that shouldn't happen normally. If you look at Grubby's resource count, he was easily able to afford templar, as his resource count was something like 4k/2k even after mass carriers and double air ups.
Also, Dimaga vs JYP showed one situation where carriers are a necessity.
Hmm, interesting points, thanks for the reply. I would post most of this in your thread, but then I thought that some people might be upset about the spoiler.
Edit2:
On January 16 2012 06:40 darkscream wrote:
The fact is, You can't declare one race superior in the late game over the other because of the myriad of other things that can happen before that. The unit choices, the unit losses, the upgrades, the tech patterns, the economy, the mis-micros, the missed macro, all these things add up.
You're right. However I should point out that
1. If the player gained too much of the advantage in the midgame, so that the outcome is pretty much decided, this discussion really doesn't matter. We mostly discuss a situation where players did some minor damage to each other going into the late game (though in that last game Grubby was at an advantage but lost it)
2. In the end, if the game is not decided in the early/mid-game, the players *will* go into late game on more or less equal footing.
3. Protoss are designed in such a way, that is much harder for them to do anything but timing attacks or turtling to 200/200 on 3 bases.
4. Last but by no means least, I search solutions for myself as well  I am a plat toss, and currently I am trying to work on my macro. Which means, I am turtling in a lot of my games (since when I focus on harassment too much, it's pretty hard with keeping up my macro) Incidentally, zergs make me rage a lot, since PvZ is my worst matchup
On January 16 2012 06:40 darkscream wrote:Mobility play is really the counter to brood lords. I recently veto'd shakuras plateau because I like infestor/brood in ZvP but blink stalkers are almost impossible to defend against on that map. The architecture of it means that literally EVERY forward wall and cliff, is in danger from both air and ground attacks. Terrans just start dropping marines everywhere once BL's come out, Protoss should drop DTs and blink stalkers around. This is how it should be; that's how you counter the standard protoss deathball or terran mechball.
You're right, however going for mobile composion means having less scary army. Blink stalkers are good for sniping expos, but if they get hit by infestor broodlord, they will die. And in the end protoss army has to clash with zerg army at some point. In that IPL3 game I referenced (Kiwikaki vs Stephano) protoss did a lot of dropping and harassing and denying expos. But in the end the game was decided only with 2 good vortexes
And about your remark on Shakuras... well look at Metalopolis for the opposite: BLs on center can easily control 2 main attack routes, also protoss has absolutely no way of defending gold from BLs
|
United States8477 Posts
Just watched that game again. Grubby actually only had ~94 army supply in army in the "big" fight. His army was 3 colossi, 5 carriers, mothership, ~17 stalkers, and 2 sentries. The rest of his supply was in probes, building carriers, and zealots/dts harassing. I suspect stephano's actual army supply was much bigger. In addition, grubby pretty much completely missed his vortex and lost his motership for nothing. Combined that with my previous comment on lack of templar/archon and the fight becomes a lot more understandable.
|
Out of interest, what strats does MC employ recent that are so good?
Does he always do some variation of a 2 base timing? If anyone knows his goto builds, hola.
|
so i spread my brood lords and he has enough energy saved up for 2 vortexes
and even if the toss doesnt pull off the archon toilet, vortex is still a really powerful spell
a key thing to note is that zvp still seems very unexplored compared to other match ups
so even if the mothership has balance problems there may be other issues that may need to be addressed in other aspects of the game that would even out the mothership
|
|
|
|
|
|
EPT
