A measure of chess ability that doesn’t depend on human players is average centipawn loss: how much worse the player’s moves are than the engine’s moves when evaluated. (This measure depends on the engine used, of course.)
Thanks. This looks like a measure based on “perfect play” as a reference, with the strongest engine you can find serving as a proxy for it. I expect it works well in the typical range of human play, but I think it’ll be a poor metric at the extreme ends of the range, near random and at close-to (and definitely at better-than) engine level.
A measure of chess ability that doesn’t depend on human players is average centipawn loss: how much worse the player’s moves are than the engine’s moves when evaluated. (This measure depends on the engine used, of course.)
Thanks. This looks like a measure based on “perfect play” as a reference, with the strongest engine you can find serving as a proxy for it. I expect it works well in the typical range of human play, but I think it’ll be a poor metric at the extreme ends of the range, near random and at close-to (and definitely at better-than) engine level.