3 ms·
chess.com should release the cheating probabilities of every titled player who has ever registered and verified on chess.com . That would be the only fair thing
by otherfiftycents 3y ago
chess.com should release the cheating probabilities of every titled player who has ever registered and verified on chess.com . That would be the only fair thing to do. It should be in fact updated dynamically and visible on a player's profile at all times.
- LeifCarrotson 3y agoI agree, but the standard rebuttal is that this would allow cheaters to quickly learn the algorithm, which relies heavily on secrecy to disguise the tools they use to detect cheating. What say you?
- BurningFrog 3y agoYeah, if you make your security features public, it is so much easier to circumvent them.
- otherfiftycents 3y agoWhatsoever algorithm they use is some variation of taking the diff of engine moves vs. historically played moves vs. moves played at a certain rating. This is not complicated, contentious nor does it sound like it is worth of a trade secret. Even if they do wish to use a private or proprietary standard to ban or suspend a user officially, they can still use a public standard (can even publish the code on Github for example) and show a standing statistic on a player's profile page using that public algorithm.
- cmcaleer 3y agoThat would be an extremely uncreative implementation. Positional complexity, time taken, mouse position/movement/clicks/arrows drawn, piece 'touch' instances, and historical performance in similar positions are all things that could conceivably be easily tracked and potentially serve as legitimacy signals.
- bnralt 3y agoParticularly since the report says that dozens of other Grandmasters cheated as well[1], and we’re not even being told who they are. [1] https://www.wsj.com/articles/chess-cheating-hans-niemann-report-magnus-carlsen-11664911524 https://www.wsj.com/articles/chess-cheating-hans-niemann-rep...