[D][G] ZvP - 3 base into fast Ultra/Hydra/Viper

Useable against most passive Protoss FE openings. The idea is to bait him in building a WoL compo with lots of colossi, immortals or archons and getting him as far away from the air tech tree as possible while getting a 16 minute ultra, hydra, viper, ling max army that can eat any ground based toss army in a matter of seconds.

It begins with the standard 3 bases zerg opener. I won't come back to the details of it as anyone should know them by now. I'm not going to detail what to do against non FE openings either as it's not the point here. Anyway:

14 - pool
15 - hatch (or 15p15h if map is large enough or 4hatches before pool if you're called CatZ)
15 - 2 pairs of lings to scout and deny pylons.
17 - ovi
18 - 1st queen
21 - 3rd hatch

Now the twist

@5:10 - 2 extractors
@100 gas : speed (around 7 min)
@100 gas : lair + 2 more extractors + 1 evo chamber
@150 gas : +1 armor
@100% lair (8:00-ish) : hydra den -> range then speed

Due to the earlier gas investment you can spend your first 25 energy from your second and third queens as a creep tumors without floating money. Just don't forget to keep your hatch rallies up to date (or better, manually assign drones to specific bases).
In theory without being harassed, at around 9:00 to 9:30, you should have an optimal saturation (16+2x3) on all 3 bases (minus 2 gas) and be producing a ling/hydra army. You do not need a macro hatch for this build. In fact skipping it (and the roach warren) is what allows you to get basically the same drone supply as with the WoL opening. You'll have to hit good injects if you're encountering a 2bases all in, though.
A good marker is to achieve 62 or more supply at 8:00 granted you haven't been harassed at all.

If the Protoss is going for a 2 bases all-in just produce units, dump gas into hydras and complete with lings. You should handle pretty much any standard build even if he get a colossus or 2 as long as you get a good angle and don't suicide your stuff. Don't forget that you can kite with your hydras (you know like you do with roaches against zealots). Otherwise if he is simply poking/pushing while expo-ing take the infestation pit, another evo chamber and 2 more gas ASAP. And take a forth as soon as he takes his own third. I can't stress this enough, the sooner you take your infestation pit, the sooner you'll get to T3 which is essential with this build. Even when under fire you should be able to find resources to do so.

Anyway get Hive and 2 armor / 1 melee attack (and continue upgrading) as soon as you can. When Hive is done, which should be at around 13min if the enemy stayed relatively passive, get the Ultralisk Cavern, build 4-5 Vipers and start stockpiling enough supply and resources for the Ultralisks. Get the carapace upgrade when the ultra cavern is done and start production.

At 15-16 (still granted he stayed more or less passive otherwise it will be a bit later) you should have 5+ ultras, 4+ vipers, 20-25 hydra and a bunch of lings. Attack, take another base (for gas, again) behind it. and keep your drone count at around 75.

Resupply mostly with ultra, try to keep your hydra count at around 20 and replace dead vipers. Lings, while being critical in absorbing damage in your first 200 vs 200 army, will become more and more useless in your main army as the time goes on. However, keeping a dozen in a separate control group to deny ninja expo and/or drops can be very handy.

      Notes about the mid-game

It's not always necessary depending on who you're pitted against but you can get away being extremely aggressive with this build as you'll be much faster than the toss (which is soo much funnier than the WoL turtle style imo). It can be fairly easy to deny or kill his third and delay his push by a few minutes by picking off units especially after a failed immo/sentry all in (or any non colossus based kind of all in). Don't forget that any mid game ground army without AoE that is not mass speedlots can melt to hydra ling with non abyssimal control. Even mixed air/ground armies don't bode much better.

      Downside

There's a timing window, before Ultra and vipers are out -especially if he didn't pushed while taking his third- where you will be sitting with an army that can get ripped to shreds by storms, Archons or enough colossi. As soon as your vipers are out and that you're not wasting them you should hold pretty easily until your ultras are ready. Loosing your 4th is kinda expected. Loosing your third may not be that much of a big deal as long as you keep the drones alive, just take the expo(s) back as soon as your army is ready. Granted you'll be gas starved and have probably several useless K of minerals, it could be a good idea to invest in a lot of spines to delay the enemy. You can also use this excess mineral for ling run by which can arguably be way more effective.

      Pros

• Strong against FFE and Nexus first
  
• Large maps and maps that offer multiple attack angles are a plus
  
• Very strong against air based openers and ground/air mixed armies
  
• Can be switched as a (hard to scout) hydra/ling all in if the protoss takes an early third without access to decent AoE damage
  
• Hard counters to this BO are relatively easy to scout

      Cons

• Weaker on smaller maps
  
• It can't be followed blindly, you need to scout well so you know if the protoss is hard countering your build or not
  
• In the late game, when your T3 army is out, a good toss can still negate your blinding clouds with feedback.
  
• doesn't work in the late game against air toss (still good opener to kill it before its ready, though)
  
• build order loss against several easy to scout all ins (2 base colo)

The basics of this build is against a standard FFE where the protoss doesn't chrono his warpgate research. Against a gate/expo, a gate/robo expand or a delayed 4 gate, you will need to take an early gas and drop a roach warren to avoid being overrun by zealots.

Speedlot based openers are a pain, they are thankfully rare but if my build stays, it may become a problem. The issue with speedlots isn't really the unit itself, because with ling/hydra, granted you micro your hydras in a corner and wreak the pathing with your lings, you should still be on top. However while you micro hard and the toss basically A move, he has time to spend the bazillion gas he is saving while you may not. Dropping even a late roach warren (it's not that long of a build time anyway) will help. I don't think you need to change your upgrade focus (melee/armor) but such opener yell "scout the protoss tech" immediately as he is obviously saving gas for something.

If you detect early enough that the toss is going for air mainly (2 starports, way too many cannons, or speedlot opener) you may want to take your 2nd evo chamber earlier and go for range + armor upgrades. In this case, the best course of action, granted you're too far away in this build to switch to a more appropriate response, is to attack him continuously before he is able to deal with your hydra too easily. In this case i'd go range/armor + infestor energy and prevent him from keeping 3 bases active at the same time. You are kinda all in in this case, but given how much gas cost your hydra/infestor/ling army you should float enough mineral to expo and drone up too.

In the late game (achieving the T3 ultra/hydra/viper 200 army), do not get cocky, a good toss will drop you and DT focus your new expos. Keep your creep going, and do not forget to build several spores and spines in each base. I lost a few games because of that myself. Given how much mineral you're supposed to float with this build you have no excuse when you don't have at least 4 spines and 2 spores per base at the 20 minute mark. More to the point, in the late game against a player who harass you, keep like 1 ultra 4 hydras 1 overseer and a few lings in a separate control group to defend your bases.

It's obvious, but always take a few seconds before any engagement to put your ultras in front even if it means loosing a base. Avoid engaging on ramps. Vipers should have a separate control group or a shared control group (like whole army on 1 and vipers only on 2), but a good player will focus them down anyway, just don't forget to replace them and consume buildings with the new ones before joining the army.

In theory your main spell is the blinding cloud. First you need to know that enemy units will auto walk out of the cloud and you can use this to your advantage. The direction units go is not random, they will go toward the nearest non clouded spot if there's not enemy in attack range, and otherwise they'll go in melee range toward this unit. So, by dropping the center of the cloud a tiny bit behind enemy units you'll force them to go toward your ultralisks and the Protoss player will have to order them to retreat through the whole cloud again, making kitting kinda pointless.
Also, clouds stay active for a little while on the terrain making it sooooo much better than fungal for zoning the toss ball out of range. As such It's a very good retreat tool (smoke bomb?) when used on ramps.

However, if your opponent is skilled enough with HT(it's easy to notice, if your 5 initial vipers die to feedback before dropping any cloud, then he is), you'll have to favor abducts instead. In this case you will need to have your vipers in their own control group so you know where they put the targeted units. Priorities are as follow: colossus, then immortal, then anything air based that can target ground. Don't abduct too much stuff at the same time, especially don't abduct 10 archons in the middle of your hydra ball . Alternatively, you may want to show the viper last: engage with the army first to bait storms, then cloud them all. You'll loose much more resources but still less than your opponent.

If there's next to no HT in his army and if for some reason (like your viper getting killed before doing their magic) you miss your clouds when facing a sizable enemy force, you will have to disengage and gather more energy. Ultra are (very) good but not invincible either. Well, if you are floating enough money for a full remax, you may as well continue to push on.

When used decently this compo wreaks anything ground based the protoss has. Colossi and Immortals are probably the worst answer to it. They are too slow to get out of the clouds in time. And a robo based army, even fully upgraded won't trade well at all (i usually get like a 2:1 resource loss difference).

Damage wise, loss are a bit more even with archons + storms based armies than with any other robo/gateway army. You then may want to stop most of the ling resupply and build queens instead so you can heal your ultras and keep being cost efficient. However as long as you keep getting new bases and deny his owns there's no reason you should lose (except if you're on 2 completely different skill levels) with or without queens.

Those replays are from when I was stuck in plat in the beginning of HoTS. I am, atm, compiling a much better set against (low) master players and I will update this section as soon as I have collected enough of them.

My apologies for the quality of those replays. It's far away from what you'd expect from a proper guide. Both me and and my opponents make many mistakes, float money forget supply cap and so on more often than not. However we BOTH do mistakes It's not like I am smashing someone down 2 leagues with a much superior macro here, armies tend to have the same value when clashing.

- Fast 3rd into colo/stalker (probably where i have the most proper viper control)
- 2 base gateway voidray
- 7 Gate
- Sentry 3rd -> colo -> archons (longer macro game)
- Robo into 3rd

      Download Archive Here (zip file)

None of my opponent did a zealot / stalker / core harass when trying this build. I don't think it would change much apart from slowing both of us down by a few seconds but i may be wrong.

I don't have a decent Immortal all in to show sadly (the only one I got doing this build was badly executed and very late), people tend to immediately switch to colo or another tech when they scout the hydra den. I may have to hide it and build a roach warren just to test the build against it even if I am fairly confident it works pretty well.
(will update if i get one)