Cheating on chess.com: the final proof

All the online chess sites have a problem with cheaters who use chess software to create their moves, what is strictly forbidden on correspondence chess sites like redhotpawn.com, gameknot.com, chess.com, chesshere.com, chessworld.net etc.

There are chess sites which seriously try to detect and ban cheaters from their site. And there are "cheater-friendly" sites like chess.com. The policy of chess.com: only if we are 100 % sure, we ban cheaters. What means, that only the dumbest cheaters are caught. But acccording to Erik, the owner of chess.com, his site does not face a cheater problem. He states on his site, that below a level of 2200 (= master chess players) the percentage of cheaters on his site is below 1 % and therefore you need not care about that topic.

OK, so far, so good. We did not believe that, that's why we run an interesting field experiment. We wanted to know it exactly. We registered a new account on chess.com, built up a nice profile with a correspondence chess rating of around 1900 and then started to cheat! Of course we were not so stupid to run the chess software on the same PC, many chess sites check for multitask-switching. We created our chess moves using Chessbase Pocket FRITZ 4.0, the worldwide best pocket chess software, run on a Pocket PC IPAQ 114 with over 600 Mhz, quite fast. This programme reached in 2009 an ELO performance of 2900. What means that it plays better then all grandmasters and of course must beat all amateur chess players on chess.com below 2000 level without exception. As we allowed Pocket Fritz just some seconds to think over its moves instead of a tourney chess time control, the performance could have been somehow weaker, lets say only 2700 ELO, still enough to beat anybody on chess.com.

We played 50 games on chess.com. We cheated all those 50 games, but sometimes only used the second best move instead of the best one (Pocket Fritz is the only pocket chess software with a multivariant-modus, it can show the second best move, third best move etc.). Our opponents on chess.com were all rated 1700 to 1900, far away from expert level. According to Erik (you remember: cheating is a very tiny problem on chess.com). we should have won all 50 games, shouldn't we?

You will be amazed to hear the result: a whopping 14 (!!!) % of our games were either drawn or even lost. Against chess.com opponents rated 1700 to 1900 – unbelievable. This is the final prove that chess.com has a serious cheater problem! Not to mention that of couse we have not been caught cheating during those 50 games (what could for sure go on for another hundreds of games without being caught). Maybe you ask now: if those people cheated (and it's a fact, they did), why their ratings are not higher? Well, nearly nobody is so dumb to cheat every game. People cheat just sometimes. Chess.com publishes a list of caught cheaters every months and the majority of caught cheaters on this list are rated below 2000.

Dear Erik: if you read this (hopefully you do): you have created a superbe site, but your site has a serious cheater problem. You should rename your site. maybe cheater.com is the better name.

Considering that chess.com is the most expensive chess site out there (99 $ per year), clients could expect upon my opinion a better cheating-protection.

Rate Complaint

Mark Inappropriate