Whoever made skytoss a thing should be ashamed

[Adventurer-16000863]

@Eliwan-1354

I have reread that and a few posts up - I stand by it. Nothing I said there is lying.

There you have it, ladies and gents: Kelthar/abs style trolling. Brazen lies that can be easily fact checked with 2 minutes of reading:

Eliwan makes three claims: 1) a player can have PvP performance different than PvZ sufficient to explain a GSL match, 2) I am completely discounting the possibility of #1, and 3) there is no reason nor evidence for my argument. Those are the three claims.

In the post Eliwan is responding to I point out the correlation between matchup rankings for a singular person is very high, I estimate it to be around 0.97 but it’s closer to 0.9, which definitively falsifies #1, definitively falsifies #2, and definitively falsifies #3. If the correlation between PvP and PvZ is 0.91, then Stats’ PvZ could be +/- 6200*(1.0-0.91) which is 6200 +/- 558 mmr¹. Remember, Stats beat a guy 800 mmr higher. Variability in matchup performance cannot explain that result, and I pointed that out. Source: https://i.imgur.com/lxxutME.png

I also link to two other posts (1, 2) which provide further analysis. In one, I calculate balance metrics via Grandmaster win-rates which showed protoss is strongest. If Protoss have a balance advantage, that’s a strong reason to suspect balance impacted Stats v Dark. Again, this definitively falsifies Eliwan’s first claim, definitively falsifies second claim, and definitively falsifies the third.

In the second post, I list a secondary way to calculate the same metric and found identical results. Each post had data and sources. Eliwan has now rebuffed that lie again. This is exactly how Kelthar and Abs troll the forums. They will brazenly lie to your face over something that can easily be fact checked. Eliwan=kelthar=abs=goba. Thus it is proved.

As always, sources, sources, sources. I consider the possibility that players performance varies between matches, and absolutely dismantle the argument here:

@Adventurer-16000863

The variability at the pro level can’t explain an effect this large. At the pro level, matchup performance correlates with a player’s skill in the ballpark of 0.97. That means any variability in performance, between matchups, is within the standard mean error. The standard mean error represents the inherent volatility of an experiment. A player’s performance per matchup varies so little it’s indistinguishable from background noise.

If a player is winning at a 6900 mmr level in PvZ and losing at a 6600 mmr level in PvP, it means their true skill level is <= 6600 and the asymmetries of PvZ gives a >=300 mmr advantage. Amazingly, that perfectly corresponds with numbers generated using 3 separate calculation methods on the GM population (here, here).

If it walks like a duck, quacks like a duck, looks like a duck, it’s a duck. This isn’t variability in matchup skill. It’s Protoss being advantaged

Eliwan immediately responds:

@Eliwan-1354

A player can be good at PvZ and bad at PvP. You’re completely discounting this as a possibility, with no reason or evidence.

Like most of your posts, you’re saying something that’s complete nonsense and pointless

Three brazen lies: 1) PvZ can be sufficiently different from PvP to explain Stats’ performance in the gsl, 2) I have not considered the possibility of #1, and 3) I have no evidence or reason.

As you can see, these are brazen lies with reckless disregard for the truth, and Eliwan has doubled-down on them 3 months later. This is kelthar style trolling to the T. #3 in particular is a rather egregious lie. No evidence? Really? You consider THIS no evidence:

Continuing the discussion from Fix MMR range for Toss and Zerg:

@Adventurer-16000863

GM is very homogeneous in age according to research done by a computational biologist on github. The same is true for practice level. The way it bypasses alts is that win-rates don’t count duplicate accounts. Whether a GM protoss wins on this account or that account doesn’t matter – it’s all counted the same. 100 games on 1 account with a 60% win-rate is the same as 50 games on 1 account, 30 games on another, and 20 games on another. It makes no difference. GM level ZvP is within 2 points of Masters 2 PvZ. Flipping wild. By contrast, GM level PvZ is 64%. There are two types of people on Earth: those who know protoss is busted, and those who need to take an intro class on statistics.

The only caveat to gm winrates is that the ladder demotes/promotes to equalize the win-rates. What that means is that, if P is advantaged over Z, Z gms will face easier T/Z opponents as a result of having reduced mmr, and that creates a counter-pressure that fights against the insane PvZ win-rates. The same is true for Protoss who are pushed upward by PvZ (they face tougher P/T opponents). Basically you have two matchups pushing zergs up and one pushing them down, two matchups pushing Protoss down and one pushing protoss up, and PvZ is still so busted to that the 1 matchup is putting almost 2 protoss into GM for every 1 zerg.

Due to how the ladder behaves, the PvZ win-rates equalize as a balance between the advantage of Protoss and the counter-pressure that other matchups create. Ergo Zerg performance is ZvZ + TvZ + PvZ = 1 which if rearranged creates -(ZvZ + TvZ) + 1 = PvZ.

It’s a little more complicated than even this because you have to account for how common each race is on the ladder. The previous equation becomes:

ZvZ*iZ + TvZ*iT + PvZ*iP = 1
ZvZ*iZ + TvZ*iT= 1 - PvZ*iP
-(ZvZ*iZ + TvZ*iT) + 1 = PvZ*iP
(-(ZvZ*iZ + TvZ*iT) + 1)/iP = PvZ

The incidence of each race in GM is 0.25/0.32/0.4, and TvZ is balanced and so is ZvZ, which makes it:

(-(0.5*0.25 + 0.5*0.33) + 1)/0.4 = 0.42
(-(0.125 + 0.165) + 1)/0.4 = PvZ
1.775 = PvZ

Double checking:

ZvZ*iZ + TvZ*iT + PvZ*iP = 1
0.25*0.5 + 0.5*0.33+ 1.775*0.4 = 1

So PvZ win-rates are under-estimated by a factor of ~1.8x. That means PvZ in GM, which has a difference of (63-56=7) is a 14% win-rate difference after demotion is accounted for. That’s a 69/31% win-rate split. This is an approximation, because to have the exact number you’d have to build a full probabilistic model and make sure it fits the ladder data etc. But this will be pretty close. We can do a sanity check on this figure using the Elo algorithm:

r0 - r1 = -ln(1.0 / Wr - 1) * 400
r0 - r1 = -ln(1.0 / 0.69 - 1) * 400
r0 - r1 = 320

We’re talking about ~300 mmr advantage for Protoss. That seems reasonable

This is what “no evidence” looks like, ladies and gents. Lmao.

Goba, it’s hilarious that you convinced bourne that you were a chick. But that would never work on me. You know I am too smart for that, man.

Subnotes:

1. This is an approximation. The true value requires the mean and standard deviation of PvP and PvZ, which are o=214,u=220 and o=195,u=249 with stats mean performance is 746. So for Stat’s performance range, we have a 68% chance his PvZ is 6200 +/- 212 mmr via the formulae 6200*(1.0-((746-214)/220*195+249)/746) which converts the correlation into a confidence interval. There is a 95% chance it’s within the range of 6200 +/- 424. To put this plainly, there is a 0.3% chance that his PvP varies from his PvZ by >= 636 mmr, and we need 800 mmr to get to Dark’s rank. There is a >99% chance balance impacted the Stats vs Dark match in the GSL.