When it comes to cheese (and by cheese, I mean an attack that starts before 1st base Saturation), there are many different types. Protoss has zealot rushing and Photon cannons, zerg have Ling Rushes and Proxy spines while Terran have Planetary rushes and Proxy Maruders.
But at the pro level only 2 sorts of cheeses seem to be viable, Proxy Hatch and Proxy Rax. Protoss don’t seem to have a viable pre-Saturation Cheese. Photon cannons come close but they are used in tournaments once in a year rather than at least once every tournament.
So the question is, should we let 2 races get quick wins and 1 not. This is an issue that affects winrates, so if Protoss is underperforming, shouldn’t this issue be address by either making Proxy Hatch and Rax worse or Photon Cannon rushes better.
No matter how you feel about these strategies, if you hate them or think they are good. This is a worst case scenario where some races have an option to try and win a game in 2 minutes and some don’t.
Personally i would love to see ANY cheese gone from this game for good. All it does is promoting easy, effortless cancerous wins without any improvement and hard work.
And of course idiots are arguing that “cheese is essential part of the game because otherwise game would be boring”, forgetting that ladder is literally infested with cheesers. About pro level don’t know, but cannon rushing in the hands of Parting was really strong. The problem is pure cannons are very easy to defend for all 3 races, protoss usually needs to transition into some sort of gateway, robo or stargate pressure.
Power level. Cannon rush can be totally unstoppable if unscouted & isn’t even allin. Proxy hatch is very easy to scout, is totally allin, and has a much lower power level. Proxy rax has a middle ground power level. Whether it’s defended or not depends on the skill of the players. Whether you can transition or not depends on the skill of the players. The only way to blind counter a cannon rush is if you blind 14 hatch, 14 pool, and bank 3 larva. You can hard counter a cannon rusher but you are behind economically vs a protoss who just aborts the rush and chronoboosts probes. So it only works vs the low iq cannon rushers. Only way to beat it is an extremely fast drone pull. The moment a probe goes behind the mineral lines you have to pull all drones but 4.
No need to even mention that cannon rush is banned from tournaments but they don’t announce it publicly and do it once in like 2 years for people to not suspect it’s banned, but it is. It has no counterplay.
Cannon rushing isn’t banned from tournaments. Pros just don’t do it because it’s bad.
Maps need specific features for cannon rushing to be “viable” at a high level, and they generally only work if scouted very late, and must be in very easily scouted locations.
From a win % perspective, ling all ins/12 pools/proxy hatch spine rushes, bunker rushes, and proxy rax or factory builds are way more viable at the pro level than cannon rushes, which is why you see them a lot more frequently in pro play.
It’s not bad. The pro protoss just suck. Parting had huge success with cannon rushes and was only narrowly beaten by serral. One tried a cannon rush vs serral in masters collosuem and looked like a gold league player because he’s bad. There just aren’t any S teir protoss right now.
Watching recent pvx games, they make very basic mistakes like not knowing how to wall their cannons, or accidentally rallying their first colossus to its death. Or they will a-click their carriers, no pull back micro, and lose several when it was totally unnecessary. Or they lose 4 oracles for free in a single game… Professional protoss are so bad it’s frankly insane they qualify for tournaments to be frank. You cannot judge the powerlevel of strategies by the pro scene because frankly tge top protoss are terrible.
I can’t count the number of times I’ve seen professional protoss forget to reset their robo rally after building an obs and just rally a robo unit to its death. They just can’t get the very basics of the game correct, and that goes right down to where to rally their production facilities. Maybe it’s nerves on tournament day or something else, but there is no denying the play is very bad.
That kills 1 type of cannon rush. Another type is where the Protoss cannon rushes the zerg nat, lets them see it and then grabs a Nexus behind it as the zerg player scrambles to stop 2 cannons from completing.
Orrrr you can do a proxy 3 gate cannon rush but that’s pretty all-in, no real way to macro out of that if it fails.
This is a disingenuous point. This sort of thing happens to high level Zerg/Terran players as well. Dark killed his own hatchery at a mining base with consume recently, for example. You’re likely just falling into a confirmation bias problem and missing the similar/equivalent missteps from T/Z players/failing to account for why certain mistakes might be more frequent for Protoss players.
Stray rallies might be a bit less common for Z/T (I’m not sure whether or not that’s the case; it does happen even in very high level play), but keep in mind that gateway production doesn’t use rally points at all, and the robo has long production cycles, so rally points come up less than they do for Z/T, so it wouldn’t be surprising if mis-rallies were more a bit more common for Protoss.
Though stuff like medivacs that were being used to harass flying over the enemy army after an F2, or vipers/infestors running into the enemy army happens reasonably often even at the top level.
Oracles flying to their death in particular is something that’s very easy to happen because they’re very fast (5.6 speed on faster), short ranged (or unable to attack), and are extremely fragile.
Abduct and Viking attacks have only 1 less range than the vision range of an Oracle, so even if they’re not moving, that leaves you with less than 0.2 seconds to react before the Oracle is in range to get attacked/dragged in, less if the enemy army is moving towards the oracle, and oracles are what Protoss use to secure vision in the late game; it’s their “safest” option. Even if casting from out of vision, Revelation can’t hit something more than 15 range from the oracle (range 9, radius 6).
Terran has Scan, and Zerg has changelings and creep - these are no risk/negligible opportunity cost options for getting vision/checking the position of the enemy army in the late game, as they have practically unlimited range. If Protoss didn’t lose oracles/obs to get vision more frequently than their T/Z opponents lose units in such a manner when on a similar skill level, that’d actually be shocking.
Typical oracle gets 15 to 20 seconds of reaction time before dying. The first indication is the sound played when the oracle is finished building. The second indication is the travel time on the minimap. Then you get an indication that your forces have engaged the enemy. At this point, the game gives you the option to hit “space bar” and move your camera to the oracle, so you don’t even need to know where it is at to react to the event. Then the oracle takes damage until it dies. Protoss fail all these checks and lose 4 oracles in a single game.
You’ll never convince anyone outside of the platinum protoss patrol circle of players that it’s asking too much to expect an S-tier protoss to right-click a building a single time. You are setting the bar so incredibly low it’s mind blowing.
These mistakes aren’t comperable and thinking they are is an indication of terrible game understanding. Losing the first oracle is equivalent to losing the first 100 gas spent. That’s equivalent to losing stim or losing metabolic boost. It’s a colossal and totally unnecessary mistake. Consume is an ability that has risk to use because it subtracts hp, there is no risk to using an oracle. You don’t lose hp on an oracle just because you made an oracle. These are flat out unforced errors.
What’s particularly embarassing about them, for these Protoss, is it shows how little the Protoss watch the minimap. They basically don’t watch the minimap at all. I saw one game in the GSL and it was about 30 seconds that a lone colossus sat on the terran’s side of the map. 1 glance at the minimap would’ve fixed that issue.
They don’t use as many hotkeys. They don’t rally their units to the proper locations. They don’t watch the minimap. Professional protoss are terrible at SC2 and protoss is what allows terrible players to punch deep into tournaments. Professional protoss have their performance capped by skill, not by game design. If a protoss was in the same ballpark in skill as Serral or Maru, they would have a 100% win-rate; they would literally never lose another game of SC2.
Imagine if it were common-place for Zergs to punch deep into tournaments while losing metabolic boost 4x in a single game. The very thought of it is a joke. It would never happen. Zerg can’t function without his first 100 gas investment being successful.
The points about Oracles were more about how easy they are to get killed while controlling them, rather than losing them because of a bad rally (which is the same for most things, not just Oracles).
The risks to Oracles that Asamu pointed out are using Oracles to scout and keeping Oracles in the army as part of a single group, the latter of which you should not do because the Oracles will just charge to their deaths. Unlike Overseers and Observers, Oracles are marked as “always a threat”, so if they get ahead of an army, everything that can shoot air will attack them first. In contrast, Overseers and Observers will normally be targeted last unless the opponent gives a manual order.
Using Oracles to scout is much riskier than Scan, Creep, or Changelings; since you are scouting with an expensive unit, with only a 1 sight range margin against Vikings and Vipers. It is easy to take damage or to get caught and lose the Oracle.
Absolutely nobody makes a protoss fly his oracles over spores. Before stats went to the military, he was famous for only scouting and defending with his oracles (with rare exceptions). Protoss dive-bomb their oracles to their death because the protoss is ahead by default and losing the first 100 or 200 or 300 gas doesn’t meaningfully hurt their mid or late game play. Remember, to get value out of the oracles you have to have the mechanical skill to be flying them around all midgame, which they can’t do and that means oracles are useless outside of dive-bombing them into spores.
The value of a single well-placed stasis trap can be that of winning or losing a game. But protoss has been balanced around players who have zero unit retainment skills.
Zero risk to cast revelation from a distance.
F2 problem / skill issue.
An overseer can’t lock your entire army in a stasis trap. The power level difference here is enormous.
Any money not spent on drones/injects is guaranteed damage to the zerg’s economy. If a zerg didn’t need creep to drone then zerg expansions would be drastically faster. The economic damage caused by having to creep spread is drastically higher than any risk an oracle poses to the protoss’ investments, and that’s guaranteed damage unlike the risk of losing an oracle with bad micro.
Every zerg unit is a dead drone. If the zerg scouts with 1 overlord it’s -3 drones if you combine the mineral and larva cost. Protoss scouting is cheaper, more reliable, with less risk, and that’s just a fact of reality. Casting revelation from a distance is nowhere near the guaranteed damage of an overlord sacrifice, nor the guaranteed damage of extra queen production. Every time you make 2 queens more than bases is equivalent to a dead zerg base. You could’ve taken another base instead of making those 2 queens. Protoss will tell you that scouting is too “risky” when risk by definition doesn’t even guarantee a negative outcome, but zerg scouting (creep) guarantees a negative outcome, and a very drastic one at that.
Revelation is the lowest risk and lowest cost scouting mechanic in the game. A mule costs 220 minerals for crying out loud. How many minerals does revelation cost again? This conversation is off the rails crazy. It’s like a tuned into a Hupsaiya stream right after he got wrecked by a zerg player.
If you watch stats vs shin (SPOILERS) you will see this in action. A proxy hatch isn’t scouted, kills 13+ workers, and the toss easily transitions to an advantaged mid game that is unwinnable for the zerg. Now imagine how bad a proxy hatch is if you just probe scout.
If the zerg doesn’t scout a cannon rush, the game is just over. Your lose your natural, you have no ability to expand, the protoss is free to expand and crank probes. The power level of cannon rushes is drastically higher than a proxy hatch.
No, they don’t, but there is a smaller margin of error for this kind of thing than there is for Overseers, Observers, and even Ravens. There is little risk with Scans, Creep, or Changelings. Hallucinated Phoenixes are also low risk, so maybe Protoss should be scouting with those more often, when they can afford it.
The Oracle itself has 10 vision range. Vikings and Vipers can potentially shoot or abduct at 9 range, giving Oracles a 1 range margin for error. There is always some risk unless you are casting Revelation blindly at max range. That is not to say that any of this is a problem, just that Oracles are more likely to be shot down, all other things being equal.
That is true, but it doesn’t change the fact that Oracles are more likely to get shot down, which (scouting) is the point of this particular discussion.
While it is true that Zerg players want to drone as much and as fast as possible, no faction can actually get away with that. Neither Terran nor Protoss nor Zerg can get away with spamming workers and their command-center equivalent without making other units for defense.
Creep spread is generally supported by extra Queens, on a separate production line, that you would need to make for defense anyway.
You must have both the larva & the supply for Drones in order to produce them. It is patently absurd to claim that losing one Overlord, which takes 100 minerals and 1 larva is equivalent to losing three drones.
An Overseer is an upgrade to an existing Overlord, and it is cheaper in total than one Oracle. Changelings, and Creep spread are also cheap, low-risk methods of scouting, and so are individual Zerglings when you don’t need detection.
A mule can mine 225 Minerals during its lifetime, but that is just an opportunity cost. The player can still mine those minerals later, just at a slower rate. Mules matter when a player is still trying to build-up, or otherwise scrapped for resources with available mineral patches to Mule. Mules matter much less when the player is near saturation, has a bank, and has decent production output already.
The benefits of scan often outweigh the opportunity cost of a Mule. Whether that is denying creep, figuring out enemy tech, emergency detection, or safely advancing without losing any actual units.
That’s straight up wrong. Revelation is the lowest risk scouting ability in the game. It’s like we are having a conversation about unicorns and faeries. In what universe is a 15 range ability that costs only energy somehow even in the risky category at all? It does doesn’t have any meaningful risk. It’s in the “guaranteed / no risk” category.
Protoss lose their oracles because they are bad. There is no other explanation. I wish there were. It wouldn’t be possible to play that bad and yet punch that deep into tournaments with zerg or terran. Losing your first 100 gas is equivalent to a zerg losing metabollic boost. It’s not unusual for protoss to lose multiple oracles and still win. Losing 2 oracles is -300 gas if you include the cost of stargate. If a zerg lost his first 300 gas, that would be like losing his lair, metabolic boost, and an infestation pit. That’s how massive these errors are, and it’s totally normal for a protoss to make these mistakes and yet play a competent mid game. They are advantaged, and if they can’t win a premier, it’s 100% a skill issue.
Are you blindly casting it, casting it on a known structure, or looking for fresh units to cast it on? Because all of that makes a considerable difference to how much risk is involved.
Attempting to track a moving army, with long ranged units (Vikings/Thors) or Vipers (assuming the Zerg player has vision from creep or some other source) can be risky. Casting Revelation on known structures or units is decidedly less risky (you can safely cast from further back without seeing the targets). Casting Revelation blindly is an option, but it can easily become wasteful.
M8 if the bar is so low for protoss that a 15 range ability is considered risky vs a 7 range attack unit that is limited to creep-only movements, then you are admitting that protoss have horrible micro. You’ve proven my point for me. There is ZERO risk to revealing a zerg’s base. Zero.
Protoss are losing hundreds of gas in early game investments because they have poor unit retainment skills. They misjudge situations and dive-bomb for drone kills, they have bad rallies, bad unit movements, bad micro, etc. It is 100% a skill issue.
I am going to say this as calmly as possible. There is no army when the protoss are losing their oracle. They are losing them to spores and queens, not corruptors and mutalisks and hydras and vipers. They have the evasion advantage, the mobility advantage, the range advantage, the risk advantage, the vision advantage, and they still can’t keep their units alive. No reasonable person looks at this set of facts and says oh yeah it’s totally a game design issue.
It is not reasonable to expect protoss to win a premier tournament when they can’t keep their first 100 gas investment alive. That’s not a reasonable position. Premier tournaments are supposed to be the absolute best play. They can’t get the basics of the game right.
