0
First I want to give a shoutout to the best competitive gaming community around: vVv Gaming. Check us out over at http://www.vvv-gaming.com/
So anyone familiar with the Protoss vs Protoss matchup knows that the four gate rules the scene. If you aren't familiar, four gate is an early-game strategy that relies on you rushing four gateways (tier 1 production buildings) and warp gate tech so you can mass units right at your opponents base with a proxy pylon, or a pylon built in close proximity to your opponent's base. The strength of the build is that it maximizes production capabilities on a single base as efficiently as possible, allowing nonstop production of units without wasting time with production facilities sitting idle or pooling resources. It is considered an all-in build because there is no smooth way to transition out of it: once you've decided to four gate you are committed. Often players will wait to expand until their main base is completely mined out, sometimes even opting to long distance mine then rather than risk putting up an expansion.
So how do you counter such a strong early game build? Why by four gating of course. Needless to say this leads to a very stale and repetitive matchup. Occasionally players will venture out into the pool of forbidden strategies and try silly things like 10 2gate, cannon rush, proxy 2gate, dt rush, blink stalkers, 2 or 3 gate robo, 1 base colossus, 1 gate fast expansion, or early phoenixes (sorry to those of you who don't play SC2, but there are too many to adequately explain here). Almost invariably this unwarranted and unwanted creativity is rewarded with a loss, excepting the very rare occasion when both players decide not to four gate. The four gate is just too strong to lose to such shenanigans without significant mistakes made by the four gating player, even if the player defending the four gate has perfect control.
So my goal for a while has been to find a build that breaks the four gate. Why bother, you ask? Why not just keep using the strategy that's known to be strongest and hope that I can do it better than my opponents? Well there are a couple of reasons:
1. Personal pride. Many people disdain the use of the four gate build, because it is so easy to pull off and requires almost zero ability to macro effectively. This turns the matchup into a micro battle, with the win going to the player who can execute the build order better or micro his units to gain an advantage.
2. Strategic advantage. In a game where most matchups have an option to play a long game where controlling resources is the key to victory, it seems that finding a way to beat the four gate with a more macro-oriented build would provide a significant advantage. Four gating diminishes the defender's advantage by providing instant reinforcements to the aggressive player. finding and implementing a build that can stop a four gate and set up an expansion would be a more sure way of securing a victory than hoping you can execute and micro the build better than your opponent.
3. Intellectual satisfaction. It's a puzzle waiting to be solved. Can we figure out how to out-fox the fox before Blizzard nerfs the fox and turns it into a rabbit? That's what I intend to find out.
Okay, so we now know why we want to defeat the four gate, so let's move on to how. Since I've been looking into this for a while, I've found two progressively better ways of dealing with four gates. Neither seems sufficient and I'll explain why.
The first method I hoped would work is a 1 gate fast expansion. It's a promising build, and against a player who does not quite execute the four gate smoothly it does a fairly decent job of holding up. The basic idea is to produce units from a single gateway while chronoboosting that gateway instead of probes or warp gates, allowing you to keep up in army size and build an expansion before placing another three gateways to hold off the inevitable attack. This build does an admirable job of holding off the four gate, but against a skilled player the timings are just such that you won't be able to hold off the attack. The build is still valuable and worth learning, since it is very strong in the Protoss vs Terran matchup, but is not viable against another Protoss.
The second build I looked to was a two stalker rush. I found this build to be even more effective than the 1 gate expand build. The idea is to build 2 stalkers as fast as possible while cutting probe production as little as possible. Then you send those two stalkers to harass the opponent and keep producing out of your two gateways to reinforce as long as you still have a strategic advantage. This build does very well for a number of reasons, but ultimately fails due to a few glaring flaws:
1. The build is very versatile, allowing a number of transitions after the initial pressure, including an expansion. Unfortunately, you almost always have to followup with a four gate transition unless you do significant damage. You can usually do enough damage to win with that four gate transition, but we are looking for a way to gain a strategic advantage, not a tactical one.
2. Most players aren't familiar with it, so when they scout two gateways they assume a zealot rush is coming and don't adequately prepare. Typically they will opt for a zealot and a sentry, which are easily defeated by the two stalkers with appropriate micro. This is negated if you are facing a skilled player who knows how to scout the build. With certain wallins, a well placed force field can sandwich one or both of your stalkers between his buildings and a zealot, which means you will probably lose the stalker and end up at a significant disadvantage. Unfortunately, it is very easy to stop the build from doing any damage by simply walling in your front with an extra pylon. By the time the two stalkers + reinforcement break down the pylon, the defending player will likely have four gateways warped in or nearly finished, allowing him the production he needs to prevent your pressure from doing any damage.
3. The early stalker pressure is an almost guaranteed win against any non-standard play. He will not be able to match your production if he went for a dt rush, or if he decided to go one-base colossus. Even if he does hold, you have scouted his build order and can now counter, while he has lost map control and does not know what you have. Unfortunately there is one particular build order that is an auto loss for 2 stalker rush: gate robo gate with a fast immortal.
What caught my attention was that a gate robo gate immortal rush build is capable of stopping the two stalker rush. Many players are optimistic about 2gate robo with that type of build order being able to stop a four gate. It makes sense when you think about it: immortals simply destroy stalkers, and zealots can be kited almost indefinitely as long as you have a few stalkers. Sentries could throw a wrench in the works, but it takes a long time to build up the gas needed to mass enough sentries to effectively block in a force. Additionally, the defenders force will be growing considerably during that time, negating the advantage of the force fields by simply having enough firepower to blast through the smaller army and weaker sentries of the attacker.
Other builds can be dismissed outright. Stargate requires too much gas to open phoenix before the four-gate hits. DT rush does not get the tech up in time to save you from the attack. One base colossus also fails in that regard. 14 nexus automatically loses to cannon rush or 2 gate pressure. FFE loses by virtue of your opponent getting a free expansion without the investment into either forge or cannons. Blink stalkers or chargelots lose to a dt rush, although the chargelots might be worth considering with the addition of an early forge. Such a build would possibly make stargate openings viable, which would hard counter the build.
Ideally I want to have two gateways and a robo up, then producing at least 1 zealot, 2 stalkers, 2 sentries, and an immortal by the 5:30 mark (earliest time for a four gate to hit). After the robo finishes chrono boost will be used almost exclusively on the robotics facility. I am certain I will have to cut probes at some point to achieve that, but I don't want to stop producing them for too long or I will simply be out-macroed and lose anyway.
As I started working on this build tonight I found that gas was not an issue early on, so I'm thinking the opening build order should be 9 pylon, scout, 14 gate (for max econ), 15 pylon, 16 gas, then cybernetics core when gateway finishes, start warpgate research, robo when cyber finishes.
I am not sure when to place the second assimilator. I had a shortage of around 100-150 minerals for my first immortal when placing the second gas after the cyber core. Using math, that's about 30 trips, at around 8 seconds per trip. You lose 3 harvesters to a geyser, so that works out to 10 trips times 8 seconds means I either need to cut probes or place my second geyser 30 seconds after my robo starts (80 seconds later than I was starting it tonight). I don't know if that will provide enough gas, since I will be on one geyser until 5 seconds before my robotics facility finishes. This may be the time where I need to cut a probe or two to make things line up right, I guess we will see.
So anyone familiar with the Protoss vs Protoss matchup knows that the four gate rules the scene. If you aren't familiar, four gate is an early-game strategy that relies on you rushing four gateways (tier 1 production buildings) and warp gate tech so you can mass units right at your opponents base with a proxy pylon, or a pylon built in close proximity to your opponent's base. The strength of the build is that it maximizes production capabilities on a single base as efficiently as possible, allowing nonstop production of units without wasting time with production facilities sitting idle or pooling resources. It is considered an all-in build because there is no smooth way to transition out of it: once you've decided to four gate you are committed. Often players will wait to expand until their main base is completely mined out, sometimes even opting to long distance mine then rather than risk putting up an expansion.
So how do you counter such a strong early game build? Why by four gating of course. Needless to say this leads to a very stale and repetitive matchup. Occasionally players will venture out into the pool of forbidden strategies and try silly things like 10 2gate, cannon rush, proxy 2gate, dt rush, blink stalkers, 2 or 3 gate robo, 1 base colossus, 1 gate fast expansion, or early phoenixes (sorry to those of you who don't play SC2, but there are too many to adequately explain here). Almost invariably this unwarranted and unwanted creativity is rewarded with a loss, excepting the very rare occasion when both players decide not to four gate. The four gate is just too strong to lose to such shenanigans without significant mistakes made by the four gating player, even if the player defending the four gate has perfect control.
So my goal for a while has been to find a build that breaks the four gate. Why bother, you ask? Why not just keep using the strategy that's known to be strongest and hope that I can do it better than my opponents? Well there are a couple of reasons:
1. Personal pride. Many people disdain the use of the four gate build, because it is so easy to pull off and requires almost zero ability to macro effectively. This turns the matchup into a micro battle, with the win going to the player who can execute the build order better or micro his units to gain an advantage.
2. Strategic advantage. In a game where most matchups have an option to play a long game where controlling resources is the key to victory, it seems that finding a way to beat the four gate with a more macro-oriented build would provide a significant advantage. Four gating diminishes the defender's advantage by providing instant reinforcements to the aggressive player. finding and implementing a build that can stop a four gate and set up an expansion would be a more sure way of securing a victory than hoping you can execute and micro the build better than your opponent.
3. Intellectual satisfaction. It's a puzzle waiting to be solved. Can we figure out how to out-fox the fox before Blizzard nerfs the fox and turns it into a rabbit? That's what I intend to find out.
Okay, so we now know why we want to defeat the four gate, so let's move on to how. Since I've been looking into this for a while, I've found two progressively better ways of dealing with four gates. Neither seems sufficient and I'll explain why.
The first method I hoped would work is a 1 gate fast expansion. It's a promising build, and against a player who does not quite execute the four gate smoothly it does a fairly decent job of holding up. The basic idea is to produce units from a single gateway while chronoboosting that gateway instead of probes or warp gates, allowing you to keep up in army size and build an expansion before placing another three gateways to hold off the inevitable attack. This build does an admirable job of holding off the four gate, but against a skilled player the timings are just such that you won't be able to hold off the attack. The build is still valuable and worth learning, since it is very strong in the Protoss vs Terran matchup, but is not viable against another Protoss.
The second build I looked to was a two stalker rush. I found this build to be even more effective than the 1 gate expand build. The idea is to build 2 stalkers as fast as possible while cutting probe production as little as possible. Then you send those two stalkers to harass the opponent and keep producing out of your two gateways to reinforce as long as you still have a strategic advantage. This build does very well for a number of reasons, but ultimately fails due to a few glaring flaws:
1. The build is very versatile, allowing a number of transitions after the initial pressure, including an expansion. Unfortunately, you almost always have to followup with a four gate transition unless you do significant damage. You can usually do enough damage to win with that four gate transition, but we are looking for a way to gain a strategic advantage, not a tactical one.
2. Most players aren't familiar with it, so when they scout two gateways they assume a zealot rush is coming and don't adequately prepare. Typically they will opt for a zealot and a sentry, which are easily defeated by the two stalkers with appropriate micro. This is negated if you are facing a skilled player who knows how to scout the build. With certain wallins, a well placed force field can sandwich one or both of your stalkers between his buildings and a zealot, which means you will probably lose the stalker and end up at a significant disadvantage. Unfortunately, it is very easy to stop the build from doing any damage by simply walling in your front with an extra pylon. By the time the two stalkers + reinforcement break down the pylon, the defending player will likely have four gateways warped in or nearly finished, allowing him the production he needs to prevent your pressure from doing any damage.
3. The early stalker pressure is an almost guaranteed win against any non-standard play. He will not be able to match your production if he went for a dt rush, or if he decided to go one-base colossus. Even if he does hold, you have scouted his build order and can now counter, while he has lost map control and does not know what you have. Unfortunately there is one particular build order that is an auto loss for 2 stalker rush: gate robo gate with a fast immortal.
What caught my attention was that a gate robo gate immortal rush build is capable of stopping the two stalker rush. Many players are optimistic about 2gate robo with that type of build order being able to stop a four gate. It makes sense when you think about it: immortals simply destroy stalkers, and zealots can be kited almost indefinitely as long as you have a few stalkers. Sentries could throw a wrench in the works, but it takes a long time to build up the gas needed to mass enough sentries to effectively block in a force. Additionally, the defenders force will be growing considerably during that time, negating the advantage of the force fields by simply having enough firepower to blast through the smaller army and weaker sentries of the attacker.
Other builds can be dismissed outright. Stargate requires too much gas to open phoenix before the four-gate hits. DT rush does not get the tech up in time to save you from the attack. One base colossus also fails in that regard. 14 nexus automatically loses to cannon rush or 2 gate pressure. FFE loses by virtue of your opponent getting a free expansion without the investment into either forge or cannons. Blink stalkers or chargelots lose to a dt rush, although the chargelots might be worth considering with the addition of an early forge. Such a build would possibly make stargate openings viable, which would hard counter the build.
Ideally I want to have two gateways and a robo up, then producing at least 1 zealot, 2 stalkers, 2 sentries, and an immortal by the 5:30 mark (earliest time for a four gate to hit). After the robo finishes chrono boost will be used almost exclusively on the robotics facility. I am certain I will have to cut probes at some point to achieve that, but I don't want to stop producing them for too long or I will simply be out-macroed and lose anyway.
As I started working on this build tonight I found that gas was not an issue early on, so I'm thinking the opening build order should be 9 pylon, scout, 14 gate (for max econ), 15 pylon, 16 gas, then cybernetics core when gateway finishes, start warpgate research, robo when cyber finishes.
I am not sure when to place the second assimilator. I had a shortage of around 100-150 minerals for my first immortal when placing the second gas after the cyber core. Using math, that's about 30 trips, at around 8 seconds per trip. You lose 3 harvesters to a geyser, so that works out to 10 trips times 8 seconds means I either need to cut probes or place my second geyser 30 seconds after my robo starts (80 seconds later than I was starting it tonight). I don't know if that will provide enough gas, since I will be on one geyser until 5 seconds before my robotics facility finishes. This may be the time where I need to cut a probe or two to make things line up right, I guess we will see.
