Cheaters are Out of Control Today. Is Lichess anti cheating team on vacation today?

@mrbasso said in #50:
> Of course all this automatic cheating detections fail if the suspected player only cheats occasionally and/or possibly makes some belivable mistakes on purpose.

No, occasional cheating will merely lengthen the time required for a compelling case against the player to be made. And making believable mistakes on purpose does nothing to mask the genuine cheating when that occurs.

@Brian-E said in #51:
> No, occasional cheating will merely lengthen the time required for a compelling case against the player to be made. And making believable mistakes on purpose does nothing to mask the genuine cheating when that occurs.

Did u try out?

Perhaps it would help if people speculating here about what lichess uses or does not use read e.g.

lichess.org/@/Cedur216/blog/cheating-how-lichess-works-and-what-users-should-think-about-it/D3KJ43n2

and followed some of the resources linked there. The author admits that he's not an insider and has only collected various incomplete sources of information but I'm quite sure even that is much more research than did most participants of this discussion, especially those making cathegoric claims which are known to be completely wrong.

yesss daddy i believe you

@EvilPyrokar said in #52:
> Did u try out?

No, but I've seen it tried.
In this fairly recent example Black, following several perfectly played games and now once again with a won position and 28 minutes left on the clock, apparently decided that in this game they had better let their opponent have a draw. The player was busted for cheating shortly afterwards.

I could show you dozens of counterexamples where obvious cheaters were not banned and others were banned although they didn't cheat IF I'm allowed to post games but I'm not...

@mrbasso said in #56:
> I could show you dozens of counterexamples where obvious cheaters were not banned and others were banned although they didn't cheat IF I'm allowed to post games but I'm not...

It would be of doubtful value because for non-specialists it's impossible to be certain whether someone is cheating or not.

The first category you mention, which you must indeed not post, would require everyone to judge whether it is "obvious cheating". That is something non-specialists simply cannot do.

The second category would tell us nothing whether we are able to judge the play or not. Someone who did not cheat in the game you post could still have cheated in other games.

I got cheated means I lost.

I honestly don't think this open forum is the best place to talk seriously about cheat detection. You need a place where the noise of the trolls and paranoids can be filtered, as well as those on the other side who believe there is no problem, and where an open technical discussion can take place. Lichess' cheat detection system is not good at the moment (see the link below), but at least most of it is open sourced. Anyone who worked on security knows that secrecy based security systems are not great. Expecting that adversities will not find out the flaws of your system does not tend to work. It is better to let the community address those flaws. Chess.com cheat detection, at the moment, is solely based on secrecy, and thus, we have no way to know whether it is really working.

lichess.org/forum/general-chess-discussion/analysis-of-lichess-cheating-detection-with-machine-learning-ml-a-mis-use-of-ml--doesnt-work

@delorenflie if you refer to trolls and paranoids, then why did you thumb #1 up and thumb me down?

also, if you refer to that thread, then you should also refer to lichess.org/forum/general-chess-discussion/analysis-of-lichess-cheating-detection-with-machine-learning-ml-a-mis-use-of-ml--doesnt-work?page=2#15