@Alecjan said in #5:
> Brilliant system! What can I say!
What can I say? Remove accuracy ...
that would be a great thing but then we would see post claiming less 7 cp for several games in row. so improvement would not be big as one would hope
Dont let these people gaslight you. Cheating on Lichess is absolutely rampant and they have the internal data to easily demonstrate this.
Here's an easy way to demonstrate this:
Go to puzzles or historical games that are from 2022 and before. Run analysis on a random sample of games from your same ELO and you will see across the board the accuracy and ACL are all WAY lower than the current set of players.
These hosting sites will do everything they can to make you be quiet about the scope and rate of the problem. They will say their fair play detection is top notch. They will not tell you the internal threshold they have for medium and hard to detect forms of cheating.
@petri999 said in #12:
> that would be a great thing but then we would see post claiming less 7 cp for several games in row. so improvement would not be big as one would hope
Yeah, that's exactly my concern: the same people would only find another misleading and misinterpreted metric to "detect cheaters".
Don't listen to yet another propagandistic troll. Better check this for reference https://lichess.org/forum/general-chess-discussion/years-of-reports-with-no-indication-that-any-reported-user-facing-any-consequences?page=3#23
> I sent a report during my game!!! with the cheater as I straight away felt that the opponent was playing suspiciosly. Expecting for the cheater to be banned during our game
Not even waiting for engine analysis, lmao talk about people's paranoia!
@mkubecek said in #14:
> Yeah, that's exactly my concern: the same people would only find another misleading and misinterpreted metric to "detect cheaters".
That's not much of a reason to keep it though. You remove one misleading metric and that's helpful on its own.
Several mods have repeated how useless and misleading it is in terms of cheat detection, and the mod reply from #15 just confirms how these many useless reports affect the moderation team's resources.
I remember exactly why the accuracy % was added to Lichess. People were begging for a copy of the game review from chesscom, which we know is bullshit, and tired of the spamming, Thibault in person coded this awful feature. Which, I mean, in terms of coding it's probably a good feature, but from a cheat detection pov it's a nightmare.
Except #15 is not a mod reply lol
@Cedur216 said in #18:
> Except #15 is not a mod reply lol
I think that referred to the linked post, which was made by "A Lichess Moderator". ;-)