[sharedfrog]

For sure. On its own, it's no more than a raised eyebrow. The next few days will show if there's any solid evidence.

Btw. the 24% chance of "winning" against a 200 Elo higher rated opponent refers to "winning a point" - it includes draws as well.

[texasbigdata]

Rookie question but don’t draws occur more than 20-25% of the time at this level? Or is that just a unique feature of chess versus say StarCraft.

Restated, are sports specific ELO scores real? Alternatively phrased, does an X ELO score in one competitive endeavor represent the same relative strength as the same ELO in another subject?

[sharedfrog]

Draws are the most common outcome at the "supergrandmaster" level, yes, to the extent that calls for changes to tournament formats or the game itself are often called for. Chess960 is the most popular chess variant, in which the starting positions of the pieces are shuffled in some way, resulting in 960 possible starting positions. There's also been experiments done with no-castling chess or tweaks to how the pawns move, or the ability to capture your own pieces.

I'm not sure about your Elo question but I think a 200 points difference should meant the same thing regardless of sport.

[lesuorac]

> Restated, are sports specific ELO scores real? Alternatively phrased, does an X ELO score in one competitive endeavor represent the same relative strength as the same ELO in another subject?

Yes and No.

Yes but in the long term. ELO is a self-fulfilling prophecy so if your ELO states you should win say 30% of the games vs a different ELO but you win 50% your ELO rises until you have an ELO that states you should win 50% of the games vs that ELO (at which point you lose the same amount of points in a loss as your wins so it stops rising. Previously you gained more points for a win than a loss so you gradually increased in rating).

No because not every sport starts ELO at the same baseline and not every sport league (i.e. Chess website) starts the base ELO at the same (i.e. Lichess starts everybody with 1500, Chess.com starts everybody with 1000). It's also a relative metrics amoungst active players (who indirectly/directly play each other) so if a GM 500 years ago reached an ELO of 2100 that doesn't mean they're the same strength of a current player with an ELO of 2100.

[sharedfrog]

> No because not every sport starts ELO at the same baseline [...] (i.e. Lichess starts everybody with 1500, Chess.com starts everybody with 1000).

It's not just where they start - they actually use different rating systems, and neither of them is the actual Elo rating system. Chess.com uses Glicko and Lichess uses Glicko-2. This is also a big part of what makes comparing players from different player pools a bit meaningless.

What I think OP is asking is "If you used the same rating system in two different sports, would a 200 point difference mean the same thing?", for which I think the answer is "yes".

[lesuorac]

> What I think OP is asking is "If you used the same rating system in two different sports, would a 200 point difference mean the same thing?", for which I think the answer is "yes".

Well, I agree with your points but I'm not sure that is what OP is trying to do.