Chess.com FAQs and Discussion on Cheating

                                  GM in Deep Cover

 How can you be so damn sure? Have you analyzed the games? Have you checked the engine correlation? Because if you haven't, you'd have to be God Almighty himself to tell!! That's one thing that hasn't been mentioned here is that while it is easy to see that someone did not use a computer - because of things like tactical oversight, inaccuracies and bad calculation - it is impossible to be sure someone cheated simply because there were no mistakes, inaccuracies, etc in one game only!! Humans do occassionally play near perfect games, plus, the game might have seemed near perfect to you but, once subjected to computer analysis, many inaccuracies, mistakes and missed opportunities become patent. I must have played over 1000 games online (not on this site alone) and I have not met a single opponent who I could say I am sure was cheating. Not one. Not only that. I sometimes analyze the games after and I find very many inaccuracies on the part of my opponent even when I was beaten in what  I thought was a perfect game.

I think that this paranoia about the chess internet being infested with cheaters is more harmful to online chess than cheating itself. It is harmful because it tears away at the fabric that human communities are made of - TRUST. You can see easily where this paranoia is leading when you consider that the only way you can confirm that your opponent is cheating during the game is by using a search engine yourself!!!! Otherwise you have no way to tell for sure.

Yes, there are cheaters and maybe (though I doubt it) there are more than 0.1 percent BUT the vast majority are not cheaters and that is good reason enough to trust.

@Karl_ I had a poor choice of words, I'm not sure if I meant blunder or mistake. Or inaccuracy. I guess it depends on how one defines them. Perhaps "tactical error" or "tactical oversight" or something else.

Chess.com should seriously consider offering an Advanced Chess category of play.  I've been looking at the games of top players here and it's pretty obvious that in many cases, they play Advanced Chess against other highly rated players and play normal chess against lower rated or titled players.  The notion that they're all mental defects who get off on crushing human players is b.s. pure and simple.  Advanced Chess is a skill and anyone who denies that doesn't know what they're talking about.  That's why the best Advanced Chess players crush engine-only players. 

Offer the ability to play Advanced Chess without the stigma of cheating.  Let people who enjoy playing it play without fear of being branded a cheater.  The incessant sarcastic and caustic posts from the likes of Costelus and Oprah are doing FAR more to damage to this site than any cheater has ever done or will ever do.   Offering the ability to play Advanced Chess while stepping up the efforts ban people who cross the line in "standard chess" is the only way that I can see to stop the spread of this cancer. 

I had high hopes that posting the names of cheaters would solve the problem but unfortunately that's not the case.

How, exactly, is that obvious?  

Good god! Just took a look at the list of cheaters and it is staggering, even for such a big site! Cheaters should be ashamed of themselves and for making the game's name dirty...

I am of the opinion that catching advanced players in turn-based is not as important as improving live chess. I'm sure that chess.com would readily admit that they could improve their cheat detection - but for now (as you can see from the caught list) it's doing the job they expect it to do.

I'm also sure that they would love to get to the point that improving their cheat detection is their #1 priority.

Wow.  Tremendous performance.  Congratulations.

Is that you, Ulf? 

1. There is a player here who played 6 games againt Julio Becerra. He won once and he drew the remaining 5 games. It is quite clear to me that this person is an amateur better than a GM. I really have no idea about cheating and I don't want to imply anything. It is just a fact.

2. Titled player on ICC means IM or GM. NM does not mean titled on ICC.

3. Cuendilar, I am completely baffled: your OTB rating is 1850, but your playing strength is well above a super-GM. After all, none of the top players in the world is able to defeat today's chess engines. 

Costelus, point 1, that individual says in his profile he has never lost OTB to a GM and beat Bent Larsen when Larsen was in his prime, although he has lost to IM's and NM's.  I agree he is playing at very strong GM level, but he mentions that he has an OTB rating at one point (it is slightly lower then on here he says, I'm sure he wrote that before he hot over 2800 here) but doesn't actually give it. 

Larsen in his prime? It might be an important distinction here, before or after the match with Fischer. Larsen did play differently after he lost.

As for the person MM78 is talking about, I'm sure he had or has an OTB rating of about 2700-2800. Otherwise, I can't explain how he defeated some engines.

You forgot FM, which is titled on ICC.

But you're right, NM is not titled on ICC or FICS. :-(

This is why I am aiming for my FM title - to get the shiny badge on these two sites! :-)

cuendillar, achmatova has a FIDE rating of 2256 :-).  I think a lot depends on the engine used and how long the user lets it run.  I got some wins and draws and banned players and also got a draw in an unreated game against a guy who was using an engine by agreement.  I was surprised as he had Fritz and Rybbka, but he explained he was only letting it run for a minute or so. 

I seriously doubt any human player's ability to beat any of the top engines, even with correspondence time controls. My assumption is the players either didn't let it run long enough against good players and so lost or else had a poor engine that beats average p[layers 100% buy struggles against good players.

I do think thought having watched a couple of your games from start to finish (against a strong friend of mine) you are playing consistently at IM level at least. 

I don't understand how many people can be rated 2050 and achieve 2600+, while others, with the same OTB rating, have a chess.com rating of about 1900.

About Arno Nickel: please, you and those who believe that a human can defeat an engine, please do your homeworks! The information is just everywhere, here is the top hit of google. Read the first sentence:

http://en.wikipedia.org/wiki/Arno_Nickel

I would rather believe in Santa Claus rather that in what you said. Arno Nickel defeated Hydra while being assisted by a weaker program (strong enough however to check for his tactical mistakes). That's what I've always said: a strong human player + an engine is much stronger than a lone computer. No human in the recent era has been able to defeat a computer. The last achievements of Rybka was playing with an exchange down against a 2700+ GM and drawing 3 times. Read here the whole story:

http://www.chessbase.com/newsdetail.asp?newsid=4919

Conclusion: An average engine on a personal computer is much stronger than any of the human players in the world. Surprisingly, it is defeated by the human players at chess.com. 

Erik should be really happy: chess.com is definitely the place were the future world champions started to play.

It may simply come down to the way they manage their time.  I notice that cuendillar has only five concurrent games at the moment and has a time per move up over ten hours.

This likely affords him ample time to "eliminate cheap mistakes" to use his words.  A similarly OTB rated player with 50 concurrent games and an average time per move under, say two hours, simply won't have the same level of diligence for each move and can be expected to be lower rated as a result.

OTB play is OTB play -- one game at a time and with largely consistent time controls, and so the comparison between OTB ratings is fair.  Here, it should not come as a surprise that each users usage pattern can skew their rating up or down from their OTB rating resulting the disparity you've cited.

If I only actually _knew_ someone at this high level, then I could vouch for them. But unfortunately I don't.

I disagree with the notion that you can "eliminate cheap mistakes" just by thinking longer - I thought that this was disproved awhile back. The idea was that if a player doesn't see a tactical element in the first bit of time then they are very likely to miss it altogether. Maybe sometimes you can. In fact, definitely you can sometimes, since I have. You know - a day into the 3 day thinking time, and you're off doing something else and then poof! WAIT A MINUTE, I can just play Qg5+ if he plays Ng6! But for sure - you cannot actually ELIMINATE cheap mistakes. I am sure of that. Eliminate is a very strong word and means reduce to zero. I do not believe that it is possible.

It has never been the case for the past correspondence players. Nobody until now, not even the correspondence world champions (from before the computer era) have been able to eliminate these cheap mistakes. As in the case of OTB, most correspondence games were decided due to tactical errors. Sooner or later in the game a tactical mistake appeared.

Again, this can be checked easily. Instead of coming here and saying (repeating the same things over and over), why don't you do your homeworks? Take a correspondence game at the highest level from the past and analyze it!

Because I'm not the one making the accusations.  Over and over, I might add.

The Grobe: so, you come here and defend/express your doubts on my accusations, but you prefer not to have an informed opinion? You prefer to argue with me and others only for the purpose of arguing.