Zerg is underpowered right now especially and also Protoss is very overpowered currently, especially vs Zerg but even vs Terran.
You’d have to be ignorant to think Protoss doesn’t have the lowest EPM.
They can win games with 50-80apm vs my 250-300apm
The crazy thing is that I am not aware of a single performance metric that protoss lead in. There isn’t a single variable that correlates with skill and which also over-estimates protoss’ performance.
Terran and Zerg have a mixture of variables that over-estimate their performance. Terrans have ridiculously high spending skill, for example, and zergs have ridiculous screen movements per minute as they dart back and forth to spread creep, inject, deal with harass, etc.
Protoss assert that Protoss in GM are just as skilled but then the obvious question is in what way. Skill has a measurable impact on performance and variables that predict performance. If they are more skilled, there would be variables that we can measure that would show it. Their refusal to admit that Protoss has a skill issue while also refusing to provide a cognizeable way that their skill can be measured is superstitious and paranormal reasoning. It really is. Ghosts are real! Can we measure them? Nope, but they are real none the less! Protoss skill is real! Can we measure it? Nope, but they are more skilled none the less!
This is how you know starcraft isn’t a strategy game because it’s very obvious from a 5 minute conversation with a typical GM that they can’t think logically and scientifically. High levels of sc2 is dominated by people who click fast and think later, which means thinking isn’t their strong suit. They memorize build orders which they then repeat for 10’s of 1000’s of games.
So on Reddit a guy says “wow, the game is so balanced because maxpax/serral/clem are so closely matched.” This data supports the opposite conclusion because the probability that outliers cluster so tightly is low. I point this out with a chart that shows a gumbel analysis which tells us how big the skill differences should be if the game is balanced. These three players have a 10% chance of clustering this tightly if the game is balanced, which means it is 9x more likely that the game isn’t balanced than that it is balanced.
So anyway I share this and immediately have commenters saying I am biased. Yes, the one using data and assigning probabilities to observations is the biased one. Out of all the people they could’ve called biased, they call that guy the biased one. They are so bad at thinking logically that they don’t realize the only unbiased person in that thread, and arguably in the entire sc2 competitive scene, was the guy they are currently attacking as biased. Their world view is a complete inversion of reality. That’s how you know starcraft isn’t a strategy game. If sc2 required strategy, these guys would hate sc2 with a burning passion because they can’t think strategically. They think irrationally and mystically.
So the conversation goes on a bit and the entire thing is just a crap show. They say “you assumed the win-rates in aligulac are independent” without realizing a dependent distribution would narrow the variance and increase the confidence protoss is overpowered. But they are adamant I made I mistake and that this mistake proves not only that I am wrong but that I don’t know what I am talking about. Meanwhile they don’t have entry level understanding of simple bernoulli variables.
I could rant all day but the point is that these people are dumb as rocks so expecting them to figure out how to win with protoss is totally insane. They will never figure it out. The game should never be balanced around people who think mystically. They should nerf protoss and send these guys back to bronze league where they belong. They never deserved GM in the first place and they certainly don’t deserve to beat Serral and Clem. After that, rely on volatility injected by crazy maps to mix up the skill structure of the pro scene.
If protoss were truly underpowered, we would see exceptional game play where the protoss loses despite playing better. We would be able to quantify their skill somehow in a measurable variable like APM. We’d be able to find examples in the game where Protoss use their superior skill to gain an advantage, and that this advantage still wasn’t enough to win. None of this exists for Protoss. They abuse f2 which means they haven’t even bothered to learn how to hotkey their army. They make ludicrous positioning and micro mistakes by rallying random robo and stargate units into enemy territory. They hit their own army with their own disruptors. They forward blink into meatgrinders. They mass ground units and f2 amove from a single direction against lurkers. They refuse to multiprong. Then they go onto reddit and cry about balance, and accuse YOU of being biased if you post charts showing, actually, no, protoss is favored.
To be clear, this isn’t a judgement about the protoss players themselves because skill level can change. You can improve or degrade. It could be that in the future Serral and Clem start to slip as a Protoss star practices really hard. This is merely an analysis of the game state as it is currently. Game design should not be used to patch skill issues, period, and people who are obviously biased, refuse to learn game basics, etc, should be ignored, period.
No worries mate. I’ve read many of your posts about the gumbel equation and other posts. Even without your statistics, I already can tell what you’re saying is true, the stats just make it easier for me to “see/believe/know”.
Like look at the greatest people in sports or at games like guitar hero and stuff, there are very very few others who are at the TOP TOP levels, some people are just way better than others.
Serral and Clem are the best players in my opinion, I think Serral is the best and it’s ridiculous that others are even close, when you watch him play.
He actually has strategy as well as speed/skill. — not saying no one else does, but he has them at levels others do not really match, — maybe there are faster players but I don’t know, he seems to have really good EPM. He has the best strategy + speed combo in SC2 imo
Partially the reason why protoss APM and EPM is so low is the result of how protoss production works - protoss warps in bunch of gateway units every 21 seconds plus they build 1-2 beefy unit from robo - immortal or collossus. Terran constantly builds ton of units from FOUR production facilities at the same time and zerg constantly checks for larva available and then queues dozens of units all at once - usually zergs increase repeat rate for they keyboards which further artificially inflates their APM.
To compensate for this - i think gateways should no longer automatically transform into warp gates and select all warpgates button should be completely removed - this would increase protoss APM by some amount.
Strategically both Serral and Clem are awful. They have no strategy at all - they play same style over and over but they play 2x faster than other progamers - that’s why they dominate pro tournaments - they execute commands much more efficiently than any other pro gamer + Clem has insane unit control with his bio.
you want make game harder for easier race not easier for hard races? but why that’s dumb should make the game as flawless as possible to make your units do what you want them to with as little as possible effort. You should never from a game design standpoint try and make anything more difficult mechanically that’s the opposite of your goal. Better is making it so lower apm totals can result in more optimal play make so you don’t need even 200 now that’s a good idea. I agree with your other statement and its bad it should be a strategy game nerf the skill ceiling so it can be don’t buff it to even the scales what a terrible idea.
This is exactly how i would love to see SC2 from the start - remove that stupid stutter-step/kite/split micro idiocy from terran bio - making it mostly “a-movey” and compensate less micro with nerfed splash for all 3 races - tanks/banes/storms etc. so that bio is still strong but does not evaporate to few tanks shots/banelings detonations.
SC2 should be as strategical as possible and creativity and tactics should be rewarded rather than unit control with insane APM/multitasking which only few human beings on the planet can pull off.
For me terran micro is easy its the fact their macro cycle has to many moving parts where as with zerg I memorize one button spam combination and i can inject all my hatcheries in seconds i can be focused on micro with toss forget about macro then warp in full armies for me the hard part about terran is having to tab to the base to build with scvs tab aue units set rallies its just a lot of different things where as with the other races you cankind fo fixate n whatever your doing and catch right back up if I miss 2 inject cycles my spam click still hits them all and i catch up if i forget to qur terran units for 1 minute game over i dont know how you fix that the core mechanics of the other races are just more forgiving. Like basic stuff like one probe shift queing bases vs needing to pull and shift click 6 scvs makes a big difference. I dont know hwo you fix it but its certainly in the make terran macro more streamlined somehow department imo. If i rnadom i know im worse with terran but its not that i cant micro marines thats easy its that i cant macro half as fast as its not as repetitive and simple and requires tabing to que while fighitng where as with toss and zerg you can mess that up and no real concern