PvZ MacSed’s 5gate

General Overview

This build is a fast expand build that opens with a Mothership Core for defense, and then follows-up with a very quick 5gate timing with Zealots and Sentries to crush your opponent before they can even enter their midgame tech.

Build Order

  • 9 – Pylon
  • 13 – Gateway
  • 15 – 2 Assimilators
  • 16 – Pylon
  • 18 – @100% Gateway, start Cybernetics Core and put 2 Probes in each Assimilator
    • You should have 4 Probes in gas
  • 19 – Zealot (*)
  • 23 – @100% Cybernetics Core (@95% Zealot), cancel Zealot (*) and start Nexus
  • 21 – Mothership Core
  • 23 – Warpgate Research
  • Benchmark: When your Mothership Core finishes (@4:10ish), you should have 21 Probes, a Mothership Core, and Warpgate around 20 seconds into the research.
  • ———— Above is NaNiwa’s Gate Expand Opener
  • 24 – @100 Gas, start 1st Sentry, and put 1 more Probe in each Assimilator
    • You should have 6 Probes in gas
  • 26 – Pylon (at Natural)
  • @100% 1nd Sentry, start 2nd Sentry and 2 more Gateways
  • @100% 2rd Sentry, start 3rd Sentry and 2 more Gateways
  • @100% 2nd and 3rd Gateways, start a Zealot and Sentry
  • Push out when Warpgate finishes (@6:30ish) with 4 Sentries, a Zealot, a Mothership Core, and 30 Probes mining at home.
    • Halt Probe production and constantly warp-in as many Zealots and Sentries as you can afford.

(*) This Zealot is started for defense against early Zerglings.  If you do not see any Zerglings going to your base when your Probe is about to build your natural Nexus, then cancel the Zealot.  If you do see Zerglings, then finish the Zealot, stay on one base, and get out the Mothership Core instantly for fast defense.

Note: If you are below Diamond league, I would suggest NOT making and then canceling this Zealot.  It is much better to be on time with your Nexus than to accidentally make an unnecessary Zealot that slows down your entire build.


For all early game scouting (before 5 minutes), look under the Scouting section of IMBAbuilds’ write-up on NaNiwa’s Gate Expand Opener.  After the early game, scouting is much more minimal than you might expect.  As long as you can confirm your opponent is getting a third Hatchery (and sometimes even if they are not getting one), you should be able to go through with this build.

With your attack, you will be able to scout what units he is using to defend.  If you are fighting mass Speedling, then constantly warp-in Zealots and Sentries.  If you are fighting Roaches, then constantly warp-in Stalkers.  Your opponent should not be able to get out either Hydras or Mutalisks in time to defend this attack.


If it appears that your opponent has way over-committed to defending your attack with units or Spine Crawlers, you can always retreat and go for the macro game.  Either force field your opponent back and walk home or mass recall your army to your natural Nexus.  A common follow-up to this build is starting a Robotics Facility and going for Colossi to deal with your opponent’s Roaches.  Alternatively, you can go for a Stargate and make Phoenix in order to prepare for your opponent’s Mutalisks.  Be sure to use Hallucinations often to check-up on what midgame tech your opponent is going for.  Take a third base when you feel safe.

Pros and Cons

This build opens with a Gateway expand, so your 5gate attack will hit very soon (before 7 minutes).  This build usually hits your opponent before they have any midgame tech.

However, this build relies on you deflecting your opponent’s scouting information in order for it to be maximally effective.  If your opponent sees your Gateway count before you move out, then it is unlikely you will be able to out-produce a mass Roach defense.

Favorable Maps

This build is strongest on maps with a narrow choke at the natural wall-off so that you can defend early on against mass Speedling builds as well as use forcefields to their maximal effectiveness during the attack.

  • King Sejong Station
  • Overgrowth


hellokitty beating Suppy with this build at ShoutCraft America

MacSed’s 5gate Tutorial Replay vs a Very Easy AI