|
Hey everyone. Here is a new unorthodox ZvP strategy that I've been testing out. It's a slightly modified 5 roach rush where you block your opponent's natural with creep instead of fully committing to the rush. Here is the build order:
+ Show Spoiler +10 Overlord
11 Pool
12 Extractor / 2 drones
14 Zergling
15 Queen
18 Overlord
18 Queen
18 Roach Warren
20 Overlord / Pull gas drones
20 Make 5 roaches
32 Hatchery
Edit: Here is an alternative build order that has a faster expo hatch but with less droning and two less roaches. I've found this to be far more effective for mid-game transition.
+ Show Spoiler +9 Overlord
11 Pool
15 Overlord
15 Queen
17 Zergling
18 Extractor / 3 drones
18 Roach Warren
17 Overlord
17 Queen
19 3 Roaches / Pull gas drones
28 Hatchery
27 Do hatch cancel trick
Send your first queen to your opponent's natural right after the first inject, along with the 2 zerglings to make sure your queen doesn't get scouted on the way. Halfway across the map, send a drone to the queen. At around 5:20, you should have 300 minerals and be a slight distance from their natural. Do the hatch cancel trick, place two creep tumors, send your queen and drone back, and THEN start your natural hatch. Spread the creep to block their build spot (you should only need one more creep tumor to reach it. Your five roaches should arrive shortly after to apply pressure to their ramp and defend the building creep tumors if need be. Save your roaches to defend against an early push.
Cost
The cost of the creep block is 75 minerals for the canceled hatch and a 3 second delay in spawning larva. Your first queen will arrive at your natural before the hatch finishes. This will also delay ling speed upgrade, which you may need to deny your opponent from expoing somewhere else.
Implications
What this does is limits toss to one base play, leaving you with less mystery and letting you adequately prepare. You will know that he will go for a build that counters the 5rr. You will know he went 3 gate robo if he clears the creep relatively early. Still, it takes up to 2 minutes for the creep to completely disappear. If he goes 4 gate, your creep can show you his army composition and the exact time he moves out. His expo will be delayed even more since he can't expand as easily while pushing. Success of this strategy depends on making sure he doesn't expand somewhere else.
Replays
Here is a replay of ZvP against a 2000 diamond player. He went blink stalkers, which was questionable since it cost him the game
http://www.sc2replayed.com/replays/142666-1v1-protoss-zerg-xelnaga-caverns
And here is a ZvT against a gold player. Pretty funny how the creep blocks his natural RIGHT before his CC lands. I got far ahead of him economically and had ultras when he did his first push. I had a huge window of opportunity to win after that but I made some terrible decisions.
http://www.sc2replayed.com/replays/142665-1v1-terran-zerg-xelnaga-caverns
Let me know what you guys think about the integrity of this strategy. I don't have the time or experience to develop this build fully, and I don't know if it provides enough benefit. However, I've had some great successes with it and have had people comment on its cleverness. It doesn't win games, but it does change the early game dynamics for your favor.
|
So what if they just 4 gate and dont even bother expoing? Your expo wont be finished in time and he will be warping in units all over you, you will have no spine crawlers, how can you possibly defend?
|
lol sounds like a fun strategy, but I wouldn't try it in any competitive setting. you need those injects! =P
|
Lol kinda of crazy thing to be doing...
Seriously, Queens are so slow. I chrono a Stalker and kill her in the middle of the Xelnanga Cavern. Also, you lose larva inject. I'd say it puts you behind.
Funny thing to do to throw him off but I doubt it'd work on higher leves.
|
this sounds far too showy to be practical. you waste a lot of minerals with the hatch cancel, and early on in the game that could be devastating, especially if it turns out he's going 4gate. on top of that, you either have to spend another 150 minerals early, which again is severely detrimental, or throw injections completely out the window. the drone/unit production lost during that time is RIDICULOUSLY detrimental, especially if you're not being aggressive with the roaches, which you suggested you didnt need to be.
also, why wouldnt you get an obs as toss anyway?
tl;dr this hurts your economy more than it's intended to hurt the protoss's. all it does is prevent FE
|
Ah I had this (or similar) done to me in a zvp recently.. It turned into mass spines actually encroaching into my base before I pulled a clutch void for victory. I'll attach a replay when I'm at pc.
|
somehow, it seems better to just proxy hatch, but that takes a longer time. it does sound fun to pull off, though.
|
On February 22 2011 10:43 Deathmare wrote:
this sounds far too showy to be practical. you waste a lot of minerals with the hatch cancel, and early on in the game that could be devastating, especially if it turns out he's going 4gate. on top of that, you either have to spend another 150 minerals early, which again is severely detrimental, or throw injections completely out the window. the drone/unit production lost during that time is RIDICULOUSLY detrimental, especially if you're not being aggressive with the roaches, which you suggested you didnt need to be.
also, why wouldnt you get an obs as toss anyway?
tl;dr this hurts your economy more than it's intended to hurt the protoss's. all it does is prevent FE
I wouldn't say it's that detrimental to your economy. You're merely foregoing early speedlings to hurt their mid game economy. Instead of spending 100/100 on ling speed upgrade and lings, you spend 150 on a second queen and 75 minerals for the cancelled hatch. You don't miss any injects or production, unless 3 seconds will cost you the game.
|
I've optimized the build order:
9 Overlord
11 Pool
15 Overlord
15 Queen
17 Zergling
18 Extractor / 3 drones
18 Roach Warren
17 Overlord
17 Queen
19 5 Roaches / Pull gas drones
32 Hatchery
You will have 3 roaches and 2 zerglings with your queen by 5:20. That way your queen will be protected from assassination.
|
On February 24 2011 07:30 JaqMs wrote:
I've optimized the build order:
9 Overlord
11 Pool
15 Overlord
15 Queen
17 Zergling
18 Extractor / 3 drones
18 Roach Warren
17 Overlord
17 Queen
19 5 Roaches / Pull gas drones
32 Hatchery
You will have 3 roaches and 2 zerglings with your queen by 5:20. That way your queen will be protected from assassination.
So that would put her arrival time at the protoss base at around 17:38?
|
On February 24 2011 07:55 thebigdonkey wrote:
So that would put her arrival time at the protoss base at around 17:38?
The queen and 3 roaches arrive within two creep tumor's distance from the opponent's natural at 5:20, early enough to block the fastest of FE's. Well not forge fast expand, but then your roaches will do major damage. That timing is in xel naga caverns, so it would be even earlier in steppes or such.
|
if he 4-gates you, don't you auto-lose, since you've invested a queen, roaches and lings into blocking an expo he would never take?
And why does the creep allow you to see his army composition if he proxy'd a pylon that you never check for because you are too busy defending the migrating queen?
I think the concept is creative and could be very useful, but I am highly skeptical of this execution vs a traditional 4-gate, which most protoss under master seem to opt for anyway.
|
this is funny. I like to sentry expand against zerg and if I went to do that and found creep, I would be heartily disappointed. I'd need to rush an observer in order to take my expo, and that is just annoying. It seems easy enough to scout that your opponent is not 4 gating, and since many toss players will sit back and mass up sentries for the first few minutes, you can probably get the queen to his nat in time to block his expo. A nice reversal on the protoss blocking the zergs natural. I'd love to see this work
|
Would this get creep to the opponent's natural in time to deny a 16 nexus opening, or a forge -> nexus variant that would be a response to a pool that early? If I saw creep encroaching and my nexus wasn't going to be placed soon enough, it's fine to cut a few probes, because I know you put yourself behind to get that creep out there.
Also, if the protoss scouts this, I think even a sentry expo can snipe building creep tumors to keep space clear for a nexus.
Finally, I don't think it's possible to keep the Protoss from scouting this with only 2 Zerglings, because if they scout no queen in your main and a hatchery that's been injected, they know the queen has to be somewhere on the map.
|
This intricate and well thought out cheese is what owns me in ladder. I sincerely hope, for every protoss' sake that no good zerg reads this xD
Also, can I hire you to think of some toss cheese for me? lol
|
Yeah, what if he just turns it into a 4gate once you block the nat? What exactly are you going to do to stop it? Drop 5 Spines lol?
EDIT: I mean, if he does turn it into a 4gate, that Queen is going to die before it will get back, and your roach army/ling army is going to get overwhelmed. Without another queen, you just aren't going to have the larva to stop a 1gas 4gate
|
The biggest problem I see is that you have to pool 300 resources for a longer period of time than normal when expanding. You're pooling 300 at 5:20 and effectively 375 at ~5:30-5:35 when you actually drop that hatch, and the hatch is 10-15seconds later than it would otherwise be.
So, rather than pulling the drone and sending it out to build a hatch in such a way that it drops it down right at 300, you're pulling the drone out at the same time, but then forgetting about it for 15seconds AND not spending that money.
I guess this could be mitigated somewhat by sending 2 drones out (one to your nat, one to the middle), but you're still dropping your hatch at 375mins instead of 300.
On top of that, you're dropping your creep tumor at the time that WarpGates are warping in if they are in fact 4gating you.
Z really needs to be on top of their early game to hold off a 4gate, either with strong one-base play or a early enough expo that they can produce out of 2 hatches when the pressure starts. 32-hatch with 5 roaches and 2 zerglings and 1 queen (either you engage in the middle to defend the queen in the middle, or the queen in the middle dies b/c you're pulling back)...that's just not going to cut it. The hatch isn't up in time to be productive and you're short something between 12 zerglings and 4 roaches.
If it can't hold off a 4-gate, then its not really viable in this metagame.
TLDR: This forces your opponent to 1-base play, and you're weak to 1-base play. It would only work against an opponent intent on expanding and not taking the win when presented to him.
|
Alright. I've created yet another build order to address the problem of 4gate counters:
+ Show Spoiler +9 Overlord
11 Pool
15 Overlord
15 Queen
17 Zergling
18 Extractor / 3 drones
18 Roach Warren
17 Overlord
17 Queen
19 3 Roaches / Pull gas drones
28 Hatchery
27 Do hatch cancel trick
The advantage of this build is that your second hatch starts 30 seconds earlier. The disadvantage is that the returning queen will miss about 15 seconds of inject on your second hatch.
I decided to cut the roaches from 5 to 3 since you are not going for the kill anyway. What you mainly want to do is make the toss use a force field or contain any stalkers that try to move out to stop the creep spread.
Since you're almost guaranteed to face an early push, I was thinking 2-4 spine crawlers should be built in your main and moved to your natural upon completion. This, along with lings and/or roaches, should be enough to defend.
|
|
|
I wouldn't know to laugh or cry if this ever happened to me  /
|
This is one of the most creative builds I have ever read about! I'm gonna try this out in some customs. Well done sir!
On February 24 2011 16:58 Klesky wrote:
I saw Morrow do a similar build 80+ times in a row on his stream a while ago. He had moderate success in high diamond on EU. It was significantly more all-in than this build; he built spinecrawlers on the creep to encroach on the protoss main.
He only really won with it because of the opponent was spinning around trying to figure out what he was doing.
I assume this was before Master's existed? Was Morrow playing against any other pro level players?
|
this is clever. probably not viable but still clever :D I don't think you can prevent the queen from getting scouted because probes are almost as fast as slow lings - if the toss is actually looking for a queen in the middle of the map :D
I might look if / on what maps you can get the queen there in time after a standard 14/14 opener. If you can you would lose almost no injects and your hatch would not be delayed, that would be pretty cool
|
I'm bumping this because I remembered seeing this thread today and tried the creep block idea on ladder. On e.g. xel naga the latest you can pool is 11 in order to have the queen there in time, HOWEVER in close positions on metal and slag you can do this with the queen from a standard speedling expand (I haven't tried temple yet). I think this has a lot of potential because it makes dealing with close positions a lot easier; you delay their expo by a LOT, no drone/units dilemma because there's no point in droning as long as he's on 1 base, and you force them to go robo off 1 base which is basically the worst opener possible in pvz. All this is easily worth the 1-2 injects you'll miss imo.
Plus you will feel compelled to watch replays from their perspective and giggle trying to imagine their face when they move out and see creep at their nat :D
|
I think this has the potential to be useful, but not in combination with an early roach push. Instead, I'd like to see the tumor(s) being placed more towards the center of the map, and then spreading creep in both directions, towards his base and mine. This could allow for a complete creep highway by the time you've reached a nice number of hydras or banes or something, and before the opponent has reached colossus/other higher tech.
|
Not quite the same thing, but I've been messing around with a 15 proxy hatch, 14 pool, no queen in the main for a while, so you get a queen started at the opponent's natural asap for a creep tumor along with 2-3 spines. With constant queen production at the natural, spines and lings to knock in the front door, and overlords for spotting, it applies similar pressure and gets creep down quickly. I like it as an alternative to speedling expand, since it punishes blind 3 WG expo (1 zealot and a bunch of sentries are bad against 3 encroaching spines with an overlord or two).
|
|
|
|
|
|
EPT
