Cheat detection is not openly discussed. Try to cheat and get banned, because you won't know how to avoid detection.
//Sometimes my opponents take a very long time at a particular position where nothing is forced and there are many options, so long that I do wonder if they have put the position into an engine...but when they come back they play just as crap as before...and probably still beat me.//
This is what I was more or less alluding to. Assuming that the engine assistance is sporadic and is only utilized in these critical moments, it would seem to be the case that it would be very difficult, if not impossible, to detect.
As a general rule, especially in shorter time controls, most people have no problem blitzing out the first 10-20 moves (develop, castle, preventive pawn moves, etc.). But then there is usually that "critical point" where simply shuffling pieces is no longer an option and either a useful waiting move must be played, or an active move in pursuit of a concrete plan. Reconstructing the position and consulting an engine can easily be disguised as "taking the time to think" and, of course, a sub-optimal (but sufficiently advantageous) move can be played which is very difficult to meet with a correct response (without using engine assistance).
Can these moves be "detected"? I'm not not sure that it would be possible. I suspect many have done this, and continue to do this. Sure, the game analysis may show 4 inaccuracies, 3 mistakes, and 1 blunder, but these statistics will never show the one right move, made the one right moment, that turned the tide against an honest, human player.
This is what I was more or less alluding to. Assuming that the engine assistance is sporadic and is only utilized in these critical moments, it would seem to be the case that it would be very difficult, if not impossible, to detect.
As a general rule, especially in shorter time controls, most people have no problem blitzing out the first 10-20 moves (develop, castle, preventive pawn moves, etc.). But then there is usually that "critical point" where simply shuffling pieces is no longer an option and either a useful waiting move must be played, or an active move in pursuit of a concrete plan. Reconstructing the position and consulting an engine can easily be disguised as "taking the time to think" and, of course, a sub-optimal (but sufficiently advantageous) move can be played which is very difficult to meet with a correct response (without using engine assistance).
Can these moves be "detected"? I'm not not sure that it would be possible. I suspect many have done this, and continue to do this. Sure, the game analysis may show 4 inaccuracies, 3 mistakes, and 1 blunder, but these statistics will never show the one right move, made the one right moment, that turned the tide against an honest, human player.
IMHO, chess is far more about avoiding the one bad move that ruins the whole game, than finding the one right move.
I personally disagree with your last point.
That Jacob, first game we ever played with each other, he played 1.c5 on me and before I could move, I thought he was cheating. (Crushed me that game)
@DavyKOTWF lol
Cheat "Detection" where you know it's been done with even 80%+ accuracy is impossible. And the only reason that you don't challenge it is because if people didn't believe in the system they wouldn't play. This is well known. It's why sites like Chesscube publicly announce when they "Get a cheat". The only way I know to put cheat detection up to 90%+ is when a player cheats with the engine on his own comp and the site uses algorithm detection software. ICC has this built into their interfaces. I think only Blitzin and Dasher. I can't speak for Lichess. I don't know. I know ICC does it because I use to run my engine on FICS as a stand alone at the same time as running my ICC account and they detected that. Both accounts were legal. But because they "Detected" the algorithm the site automatically banned my account without question. That was back in like 1998.
You can't "detect" in the sense that people sometimes think.. You can only assume. Some assumption have very very high accuracy. If you called it evidentiary detection as lets say similar to how crimes are "Solved", then you would have to pull more than one "Evidence". Again.. It's considered skilled assumptions. And the accuracy of these assumptions can be very high.
You can't "detect" in the sense that people sometimes think.. You can only assume. Some assumption have very very high accuracy. If you called it evidentiary detection as lets say similar to how crimes are "Solved", then you would have to pull more than one "Evidence". Again.. It's considered skilled assumptions. And the accuracy of these assumptions can be very high.
The best way to detect such a subtle cheaters is to check it by a human.
There are some players here I think they kind of specialize in a such recongnizing.
If there is a lot of suspicions they should probably determine.
Especially if you play against such one I've seen it on someones stream.
Playing so badly mixed with so greatly - just so obvious cheating.
It could have been felt in the air and took only couple more games to make sure and ban the guy.
There are some players here I think they kind of specialize in a such recongnizing.
If there is a lot of suspicions they should probably determine.
Especially if you play against such one I've seen it on someones stream.
Playing so badly mixed with so greatly - just so obvious cheating.
It could have been felt in the air and took only couple more games to make sure and ban the guy.
Only problem with that is lichess can't really ban unless there are I don't know how much but probably over 99 percent sure it is a cheat. And subtle cheaters my 'specialize' in it so might be difficult really to prevent sometimes.
This topic has been archived and can no longer be replied to.