Cheese- A strategy or tactic that relies entirely on surprise, niche, or matchup-specific weaknesses involving timings, lack of specific units, and or defenses. This reliance is so heavy a player will ignore any economic progress in exchange for an attack.
Cheese relies so much on this attack that if the player fails to inflict crippling damage or win the game outright, they will be economically behind to a large degree. At this point, the rest of the game should become an exercise in futility.
We must keep in mind Starcraft 2 has many variables and win conditions. We cannot ignore a player's ability/rank while attempting to apply this definition. There will be some cases in which a player will come back to win the game regardless of their failed attempt at cheesing. However, this will become less likely as a player's skill and rank improve but will never become absolute.
Cheese is usually characterized by its strategic fragility. The cheesing player is making a wager that the defending player will have some sort of oversight in his build. Examples are: Lack of detection against Dark Templars or Banshees, lack of anti-air against certain Stargate builds, or economic greediness against a two raxing Terran. However, the cheesing player is usually putting himself at a large deficit in the long run by attempting to expose this weakness. In all cases (see: BitByBit) the cheesing player will put 100% of his economy on the line in attempting an early win.
The most common cheese openings, like Bunker, Cannon, 6/7/10pool and Proxy rushes, prey upon the economic style's need for workers in early defense. Others invest in early tech that must be countered in a specific way. Some are more nuanced: seven and five Roach into Speedling Zerg vs Protoss, and Hellion into two port Banshee pressure Terran vs Zerg, rely on forcing a certain enemy tech choice and then countering it decisively as it is deployed (In the first case the Roaches bust the wall and force Stalkers; in the second case the Hellions force speed/Spines/Queens/Roaches and delay Lair). Here, the cheesing player gains the advantage by assuming that his opponent will react to what he sees in a "smart" way. Like most cheese, these builds are much easier to stop once a player has gained a familiarity with them. This is why many cheese builds surge in popularity on the ladder, only to disappear weeks later.
Examples of Cheese
- Cannon Rush (Protoss) - Building a Forge before the Gateway while placing one or more Pylons in the enemy mineral line, followed by Photon Cannons to stop mining and eliminate any army resistance that may have been established. There are also variations which place a proxy pylon in other locations inside the other player's base.
- Proxy-Rax - If scouted in time can be defended, but extremely hard to defend against if it comes at a surprise. No longer possible in SC2. The new proxy is SBB, usually against Zerg for early marine pressure, and combined with bunkers.
- Proxy Gateway - building a gateway in the opponent's base and Chronoboosting out the first Zealot in order to kill the opposing workers.
- Worker Rush - A worker rush is when you take all of your starting SCVs/Probes/Drones to the enemy base and attack everything there.
- Proxy Hatchery - A ramp assault using queens and Spine Crawlers can occur in ZvP when a Zerg places a hatchery in his opponent's expansion, and uses the creep spread to plant the spinecrawlers. The hatchery also allows faster unit attack after production, as well as queen production.
- Players able to resist the All-In push of a cheese build can often expect an early victory or at least a strong advantage going into the mid-game as the opposing player must make significant sacrifices to early game economy to properly execute the cheese.
- During a broadcast game on September 16, 2009, OGN commentator Um Jae Kyung (엄재경) briefly discussed the difference between a bunker rush and a "cheese" rush. According to his explanation, the term "cheese" originated from the word "cheaters" (words in Korean are sometimes shortened by the middle syllables, so 치터즈 [chi tuh zu] would become 치즈 [chi zu]). Since then 'cheese' has been expanded to include almost every all-in strategy.
- The term "cheese strategy" was first used in 1993. Quote: "The cheese strategy is just to use straight up and down Roundhouse kicks, (...)" (Street Fighter 2, Usenet)