Several changes in chess

Hello,

In recent days, we have made several changes to chess in vint.ee:

1. We made Computer Level 0 significantly easier
2. At the end of the game, we now calculate the Average Centipawn Loss for both players. This helps us identify users using external help.
3. We moved the game board design button to a more visible place (right next to the table)
4. We now also save the last selected computer difficulty level!
5. We also fixed the Fischer chess bug, where sometimes when playing with the computer, the computer's time would run to zero.

If you have any other ideas on how to improve chess playing on vint.ee, let us know!

Could you please explain what exactly this "Average Centipawn Loss" means and what those numbers in the ACL indicate; before the game, during the game and after the game!?

If you asked in the vint.ee chat room

@vint.ee what is "Average Centipawn Loss"?

Then you would get the answer.

"Average Centipawn Loss" is a chess term used to evaluate the quality of a game played by a player or a computer. It measures how much a player's moves, on average, deviate from the best moves suggested by the computer. One centipawn is one hundredth of the value of a pawn, so the average centipawn loss indicates how much value a player loses on average with each move.
For example, if a player's average centipawn loss is 20, this means that their moves are on average 0.2 pawn values worse than the computer's best moves. The lower this figure, the more accurate the player is playing. Top players often have very low average centipawn losses, which indicates that they make very few mistakes.

Is such a theme possible in different types of checkers?

And regarding this ACL, I would ask another question: is it possible that the player whose number is higher during the game will win?

There are no programs for checkers like Stockfish for chess. So the answer today is no.

Yes, I think it's possible that the player with the higher ACL wins the game. The opponent made a particularly bad move at the end and lost.

Probably, at least in Russian checkers, the possibility of detecting cheating is several times smaller, based on such an analysis, because there are often only 2 options to make a "good" or what the computer thinks is a good move, and after the middlegame, there are often only so-called 1-move variants :) . It is very difficult to distinguish there whether it is cheating or not....