- Blind mode tutorial
lichess.org
Donate

Suspicious Chess App Idea?

@HollowLeaf

Thanks for the blog. Enjoyable to read with the metrics well explained.

That work led me to an idea I want to explore more openly: building a suspicious game detector. Not as an anti-cheat system, and not as a tool to accuse or label players — but as a way to provide context after a difficult or confusing game.

A 'suspicious game' detector will simply lead to accusing and labeling. One game cannot be a base. Of course it's 'just' a signal, but it doesn't change what someone will think.

Entering a position, waiting for analysis, and then playing the move introduces a rhythm that differs from natural human thinking. Sudden delays before critical moments, followed by instant replies in quiet positions, can create patterns that feel unnatural when viewed across an entire game.

Is it not the exact opposite? Why is delaying during a critical moment sus? Obviously people will think longer in critical positions, actually its been proven through a study which finds that people think longer when the benefit of thinking is higher: https://lichess.org/@/RuyLopez1000/blog/position-vs-time-the-trade-off/4o148A7t

@HollowLeaf Thanks for the blog. Enjoyable to read with the metrics well explained. >That work led me to an idea I want to explore more openly: building a suspicious game detector. Not as an anti-cheat system, and not as a tool to accuse or label players — but as a way to provide context after a difficult or confusing game. A 'suspicious game' detector will simply lead to accusing and labeling. One game cannot be a base. Of course it's 'just' a signal, but it doesn't change what someone will think. >Entering a position, waiting for analysis, and then playing the move introduces a rhythm that differs from natural human thinking. Sudden delays before critical moments, followed by instant replies in quiet positions, can create patterns that feel unnatural when viewed across an entire game. Is it not the exact opposite? Why is delaying during a critical moment sus? Obviously people will think longer in critical positions, actually its been proven through a study which finds that people think longer when the benefit of thinking is higher: https://lichess.org/@/RuyLopez1000/blog/position-vs-time-the-trade-off/4o148A7t

@g6firste6second said in #3:

someone made something similar already based on the humaness of a move at any rating, but feel free to code anyway -

https://yoshachess.com/article/oracle-best-human-like-chess-engine/#evaluations-in-chess

Thanks for sharing, I have been working on chess stuff for a while, and I have not heard of this project, but it does sound unique.

Combine GPT with stockfish for predicting human moves, that is definitely a unique and interesting take.

@g6firste6second said in #3: > someone made something similar already based on the humaness of a move at any rating, but feel free to code anyway - > > https://yoshachess.com/article/oracle-best-human-like-chess-engine/#evaluations-in-chess Thanks for sharing, I have been working on chess stuff for a while, and I have not heard of this project, but it does sound unique. Combine GPT with stockfish for predicting human moves, that is definitely a unique and interesting take.

@RuyLopez1000 said in #2:

A 'suspicious game' detector will simply lead to accusing and labeling. One game cannot be a base. Of course it's 'just' a signal, but it doesn't change what someone will think.

Indeed, hence why bringing in the whole profile, opening performances, where and how they spend time to see if there is anything out of the norm. And to be combined with the rest.

@RuyLopez1000 said in #2: > A 'suspicious game' detector will simply lead to accusing and labeling. One game cannot be a base. Of course it's 'just' a signal, but it doesn't change what someone will think. Indeed, hence why bringing in the whole profile, opening performances, where and how they spend time to see if there is anything out of the norm. And to be combined with the rest.

@RuyLopez1000 said in #2:

A 'suspicious game' detector will simply lead to accusing and labeling. One game cannot be a base. Of course it's 'just' a signal, but it doesn't change what someone will think.

I could be naive, but I actually believe that cheating is far lower than people suspect, so I think that most games will have a lower probability, and people can get on to the next game as opposed to stewing about it.

@RuyLopez1000 said in #2: > A 'suspicious game' detector will simply lead to accusing and labeling. One game cannot be a base. Of course it's 'just' a signal, but it doesn't change what someone will think. I could be naive, but I actually believe that cheating is far lower than people suspect, so I think that most games will have a lower probability, and people can get on to the next game as opposed to stewing about it.

I would certainly not install a suspicious chess app

I would certainly not install a suspicious chess app

@TotalNoob69 said in #7:

I would certainly not install a suspicious chess app

Thankfully it will be a web app :)

@TotalNoob69 said in #7: > I would certainly not install a suspicious chess app Thankfully it will be a web app :)

I think this would be a terrible idea. Just the existence of such an app will reinforce beliefs that cheating is a widespread problem. People will complain about false positives and false negatives. Public Shaming will skyrocket. The current approach by lichess works: You can anonymously report people for cheating suspicions, they take their time to review allegations and then silently take action (or not). Public shaming is forbidden precisely because it creates an atmosphere of distrust and suspicion everywhere.

You code a lot of good stuff, but this one I would prefer to not exist.

I think this would be a terrible idea. Just the existence of such an app will reinforce beliefs that cheating is a widespread problem. People will complain about false positives and false negatives. Public Shaming will skyrocket. The current approach by lichess works: You can anonymously report people for cheating suspicions, they take their time to review allegations and then silently take action (or not). Public shaming is forbidden precisely because it creates an atmosphere of distrust and suspicion everywhere. You code a lot of good stuff, but this one I would prefer to not exist.

@CommonElk said in #9:

I think this would be a terrible idea. Just the existence of such an app will reinforce beliefs that cheating is a widespread problem. People will complain about false positives and false negatives. Public Shaming will skyrocket. The current approach by lichess works: You can anonymously report people for cheating suspicions, they take their time to review allegations and then silently take action (or not). Public shaming is forbidden precisely because it creates an atmosphere of distrust and suspicion everywhere.

You code a lot of good stuff, but this one I would prefer to not exist.

I will code this for myself, and then test to see what it looks like. My personal hope is that the likelihood would be extremely low (I could be very naive), and I hope that this will reinforce that most people are not cheating.

The issue with with reporting is that the feedback cycle is too long. And the app will not be a simple score, but lots of analysis around the game.

@CommonElk said in #9: > I think this would be a terrible idea. Just the existence of such an app will reinforce beliefs that cheating is a widespread problem. People will complain about false positives and false negatives. Public Shaming will skyrocket. The current approach by lichess works: You can anonymously report people for cheating suspicions, they take their time to review allegations and then silently take action (or not). Public shaming is forbidden precisely because it creates an atmosphere of distrust and suspicion everywhere. > > You code a lot of good stuff, but this one I would prefer to not exist. I will code this for myself, and then test to see what it looks like. My personal hope is that the likelihood would be extremely low (I could be very naive), and I hope that this will reinforce that most people are not cheating. The issue with with reporting is that the feedback cycle is too long. And the app will not be a simple score, but lots of analysis around the game.