Black Is OK

@Toscani said in #28:
> A Standard deviation diagram should be able to show how much the WDL is changing as our rating increases or as a player is given more time than the other to play. Engines don't try to flag or berserk their opponent, but humans do.
Stop right away. You are changing the actual point of debate.
This talk isn't about 'rating' at all. I don't need those stats. Irrespective of rating, do you have any strong evidence?

> Rating gains is a good way to look at the similar player results and differences. Some below make it look like it's white that has a small advantage but one does not. Maybe one studied black openings more than white openings.
> When the color no longer matters, and both rating gains are positive, what do you conclude there? White still won a bit more.

Rating rain is different concept that what my discussion aims to achieve for.
Plus, everyone seems to be an engine as they give preference to White for First Move Advantage for no reason. SF evaluation at start of game maybe +0.2 or whatever, but it literally doesn't matter. That maybe called bias or whatever! Is chess solved? Not yet. How would you explain SF giving slight advantage to White at start of game? (in case you can)
For the least, can you try to show some data where SF plays itself many times and display the figures like win% and all.

> ------------------------------------------------------------------------
> Please, don't copy quote my forum posts. Thanks.
> ------------------------------------------------------------------------
I don't need to as there was lot of junk that wasn't required for discussion as it deviates from actual motive. No need to thank me either!

@BeresOliverHU said in #1:
> What advantage would white have anyway?
bro is literally destroying racism

I AGREE BLACK IS OK

BUT IN DARK ONLY
CUZ IT'S NOT VISIBLE ONLY

ok this is just a joke . pls do not take seriously . i am also not tubelight

You guys are sooo racist

I think people are looking at this the wrong way.

Suppose two identical people are playing their first game of chess. Neither has any idea what an opening is.
White is more likely to win, by virtue of being able to make black react.

This is backed up by white retaining its slight edge on lichess at lower levels, I’d assume.

Unless I’m mistaken, white wins more at all levels of competitive chess

At some point, you’ve gotta think that an apparent advantage that manifests itself in data is a real advantage.

@clousems said in #34:
> Suppose two identical people are playing their first game of chess. Neither has any idea what an opening is.
> White is more likely to win, by virtue of being able to make black react.
> This is backed up by white retaining its slight edge on lichess at lower levels, I’d assume.
> Unless I’m mistaken, white wins more at all levels of competitive chess
> At some point, you’ve gotta think that an apparent advantage that manifests itself in data is a real advantage.
'likely', 'assume', 'unless', 'apparent' are all weak terms while dealing with such cases. Do you have any stats for debut matches of players (at least at Lichess) to support your 'thinking'?

Yes, yes I do.
In very low level games (note: debuts was an example, not proof), white wins approximately 4% more often than black. There are enough games to make me think that is indicative of an imbalance (feel free to run a significance test on the numbers if you like, but I don’t care enough to do so).

As for my word choice: “likely” is used in a mathematical sense; “apparent” is in no way a “weak” term; “unless” is a standard word used to introduce possible alternatives.

(PS: “Thinking” should not have been placed in quotations—unless it was used ironically, in which case that probably was not a great idea. Always make sure you’re right before you’re snippy)

@Akbar2thegreat said in #29:
> Pretty amusing!
>
>
> As per you, you provided info but these are mere words that cannot prove something.
> And I am not talking about lichess games rather all in history and I am not talking about first move advantage only rather about how do you guarantee of White winning more (in classical) than Back?

Lichess has a database of historical games up to 1952, it is the Master's database I was referring too. But if for some reason you don't like this, you can check Millionbase, a database containing around 2 million chess games, including historical chess games. White wins ~56% of the time (see www.sumsar.net/blog/2015/06/big-data-and-chess/). You can also check the statistics by ELO, and note that White wins more often at a higher level (chess.stackexchange.com/a/1507).

>If you really want to provided so called 'evidence', then play 100 Games vs Carlsen where you will be White in all games (I will manage and fund the event) and I will see the result of stats later. If you cannot do so, then your explanation has no value and hence nullified.

This is an ignoratio elenchi fallacy: Me playing 100 games against Carlsen with white (and losing the large majority of them) does not prove that white has no first move advantage. Instead, this is fully explained by the fact that Carlsen is a much stronger player than me. As explicitly said in my last comment, I am making a statement about *all games*, not about the specifically constructed situation by you in which a much stronger player plays with black against a much weaker player with white (by specifically correlating the colors with playing strength, you are purposefully introducing the confounders about which I was talking; in a typical match against Carlsen I would be about 50% times white and about 50% times black).

But to provide insult to inury, you can check easily using extended search here on lichess that Carlsen's win rate with white in online chess across all time controls is about 72%, while it is only about 65% for black. So even across the subgroup of only Carlsen's online games, white has a statistical advantage, and so if a generic opponent of Carlsen were to play with black 100 games against Carlsen, he would be expected to do worse than if he were to play 100 games with white.

TLDR: If a generic opponent of Carlsen who is much weaker than him were to play with white against Carlsen, they are expected to do bad, but they are expected to do even worse with black!

@Roadto2_1k said in #37:
> Lichess has a database of historical games up to 1952, it is the Master's database I was referring too. But if for some reason you don't like this, you can check Millionbase, a database containing around 2 million chess games, including historical chess games. White wins ~56% of the time (see www.sumsar.net/blog/2015/06/big-data-and-chess/). You can also check the statistics by ELO, and note that White wins more often at a higher level (chess.stackexchange.com/a/1507).
Still all that doesn't answer my question. They are all like boundary/peripherals that you gave but not the actual inside one. It lacks the quite a lot of things.
Still going by that last link (stackexchange), draw% has increased as level increases. So, advantage for White (if any) decreases. It rather keeps my point to some extent though not fully.

> This is an ignoratio elenchi fallacy: Me playing 100 games against Carlsen with white (and losing the large majority of them) does not prove that white has no first move advantage. Instead, this is fully explained by the fact that Carlsen is a much stronger player than me. As explicitly said in my last comment, I am making a statement about *all games*, not about the specifically constructed situation by you in which a much stronger player plays with black against a much weaker player with white (by specifically correlating the colors with playing strength, you are purposefully introducing the confounders about which I was talking; in a typical match against Carlsen I would be about 50% times white and about 50% times black).
So your theory fails. If your theory can't handle situational problems like these, you ought not to force your thinking over others by saying White has more chances of winning. You can't escape the discussion by saying that I deliberately paired two lot different players based on strength. (I meant 50 White 50 Black actually vs Carlsen for the case not 100 White!)
So, up for challenge?

@clousems said in #36:
> Yes, yes I do.
> In very low level games (note: debuts was an example, not proof), white wins approximately 4% more often than black. There are enough games to make me think that is indicative of an imbalance (feel free to run a significance test on the numbers if you like, but I don’t care enough to do so).
I need proper strong proof/evidence supported by verified citations by experts and not any random guy's made up theory.

> As for my word choice: “likely” is used in a mathematical sense; “apparent” is in no way a “weak” term; “unless” is a standard word used to introduce possible alternatives.
With terms like those, you can never prove anything about particular subject. Ever heard of a known proof containing those words?

> (PS: “Thinking” should not have been placed in quotations—unless it was used ironically, in which case that probably was not a great idea. Always make sure you’re right before you’re snippy)
Yes, thinking was put in quotes to take a dig towards you for being too subjective over a problem that has a proper solution (which I am looking for).