What it is like being a Zerg GM during ProtossCraft II

We all know the super tournament is not the same as a straight up GSL Title. If anything, the protoss race would be better off if those events were never put on in the first place.

Referencing the one with the elite 8 consisting of 1 Z and 7 P. Everyone knows how hard terran whined. Hell, even the douchebag protoss players on the pylon show were saying it was disgusting. (So glad that show has died since then)

After that tournament, which of course the one lone zerg player predictably won over 7 protoss, protoss was heavily, heavily nerfed in many areas. You can go back and look at patch after patch. The severe overreaction that blizz always does to protoss.

Simply put, if that tournament was 7 terran and 1 zerg, there is no way the zerg wins that. And there is nowhere near the same level of whine. I just am not seeing enough said about all these cancerous TvT GSL finals, which i proudly do not watch anymore.

Sadly they killed GSL by putting it on Youtube, and I have no interest to watch terrans win another every 3 months.

1 Like

TLDR

“Global warming is false because it’s cold in Antarctica” is not a valid argument; it’s cherry picking.

“Stop cherry picking, the temperature in Antarctica doesn’t invalidate the greater trend, and it’s fallacious to say otherwise” is a valid argument and it certainly isn’t cherry picking.

When your argument relies on inverting the definitions to words, you need a new argument.

Nice cherry picking of the fake GSL title. May I repeat;

I hope you don’t think you are honest, lol, most of your forum actions are dishonest, smurfs, hiding data, cherripicking, creating arguments from things wasn’t said…

Supertournaments usually are considered as GSL titles, it is not a regular GSL but counts as it, at least people have been considering it as GSL since wings of liberty.

1 Like

I just witnessed Reynor begging to play Maru himself instead of Showtime because Reynor didn’t want to play lategame protoss LOL

2 Likes

super tournament is old.

You missed the point…again.
My point is that if the primary aim of AI was to win than just using the inherent advantage of AI (immense APM and multitasking) that would be easy.
Nowhere i said that AlphaStar should have 100 000APM.

Ok this is back where GOM TvT came from.

Proof? Alphastar was famously bad at building walls. Alphastar would stomp cannon rushes just by out macroing/building the Protoss player. Alphastar had amazing Zerg play… I would love to see some actual numbers.

2 Likes

Terrible at building walls, yet got 6.3k mmr with protoss LOL.

Proof?

20 characters.

How about, uh, AlphaStar’s actual MMR? It had waaaaay more MMR as Protoss than Zerg.

These figures you’ve given aren’t entirely fair; the odds of any distribution is small. Exactly 33% Terran, exactly 33% Zerg, and exactly 33% Protoss is (if we assume these distributions are assigned by a die) as likely as 1% Terran, 1% Zerg, and 98% Protoss. What you need to look at is the “ranges” (idk how to word it I’m not a mathematician.

Like, nobody would blink at 30% Terran; 35% Zerg; 35% Protoss, or 35% Terran; 35% Zerg; 30% Protoss, because there is some “wiggle room” from the ideal 33% each. However, something like 45% Protoss is completely outside of that range, and we must also consider that 45% Protoss is much less likely than 33% Protoss, because these figures are not assigned randomly by a die.

But still, the 1/1,000,000 doesn’t exactly hold up, because you need to add on 269/600, 271/600, etc.

Where are these actual numbers?

https://www.google.com/search?q=alphastar+mmr first result (in case you can’t read, AS-P was just shy of 500 MMR above AS-Z)

1 Like

Figure 2.a on page 352 of the AlphaStar paper.

Zerg had 6,500 symmetrical Elo and 5,643 asymmetrical Elo.
Protoss had 6,196 symmetrical Elo and 6,297 asymmetrical Elo.

P>T>Z exactly like the pro scene data & exactly like the ladder data.

PvP < PvT/PvZ exactly like the pro/ladder data.

ZvZ > ZvP/ZvT exactly like the pro/ladder data.

2 Likes

The average mmr they list is a bit misleading about balance because the average included symmetrical matchups (TvT, PvP, ZvZ) which say nothing about how AlphaStar performed in asymmetrical matchups (aka matchups affected by balance).

If you compare PvZ to ZvP, the difference is 654 MMR. This was before the void ray & shield battery buffs and before the most recent map pools.

2 Likes

You cannot make an argument over what “percentage” of a race “should” be in any particular league without taking the total racial distribution into account.

In practice people have many of their own biases (what they consider normal, favored playstyles, etc) that affect the total racial distribution on the ladder; and this shows when you look at those results.

2 Likes

LOL, Terranicll is not aware that what is paraded here as JSTLF is nothing but a BatZ alt.
It’s funny.

1 Like

I am aware, and do not respect the OP’s opinion. GM is too small of a sample size for that kind of argument anyway.

I was just nitpicking:
If someone was to seriously use a ladder distribution argument, then they could not just use 33% as the expected value for each race. They would have to base their expected values on the ladder’s overall racial distribution.

2 Likes

I mean obviously you have to consider that. However, one should exclude (or reduce the impact on statistics) of metal league players from these figures, as new players overwhelmingly pick terran.

2 Likes