cheating on chess.com

Cheating sucks.  Cheating is bad, obviously.  This 'argument' is like two ships passing in the night.  Look, since there is obviously a difference in definition here I think that the best solution is to make it patently obvious for anyone who plays CChess which definition their opponent is using.  Let those who consult programs get a big (C) next their name (the C is for computer consultation not cheater!) and then that way everyone can go home happy.  

As to the inevitable (and obvious) argument: yes, I am aware that program users are breaking the rules of this site, and I do not condone that, but an internet community should be allowed to define itself.  To that end, we should accommodate those who wish to use any and all aids, and those who believe that computers spoil the game.  Engine aid is only illegal, in my opinion, when the users don't disclose it before the game (like plagiarism, don't pass it off as completely your own analysis).  I don't think that any competent chess player has grounds for an objection if they make the first move with full knowledge that they are playing against a 'robocop.'  

I hought I had commented already, but as I cannot find my comment I had better have another go.

If people cheat at chess they only cheat themselves. What enjoyment or satisfaction can there be in making a move that  a programme tells you to make?

The real joy in chess is in the intellectual challenge in figuring things out for oneself, exercising the imagination, clinging on when things go wrong, extricating oneself from almost impossible catastrophes,  following a successful plan serenely,  and  being diabolically and stubbornly inventive when necessary,  as well as recognising and acknowledging someone else's brilliance, in which one had a share,  gracefully. Everyone likes to win. Hopefully everyone is passionately competitive. But  no cheat ever really wins.It as a programme or an engine or a Grandmaster at one's elbow. How sad is that? The 'winner' does not love the game or experience the 'rush' of a real win.

 Exactly my sentiments Brianpeter, if your going to log onto Chess.com and use an engine WHY BOTHER!  It's a pointless act to even pay for a membership.

 I caught my partner cheating one time and let me tell you, I was not pleased one bit!!  Talk about throwing a fit!  I never thought I was the type to break things.  I don't condone violence, but screaming and binge eating wasn't making me feel any better!

Hey... what about removing the ability to download game PGNs until the game is concluded. This might interfere with engine users since if they really want to cheat they must create the board in the engine from scratch. Might sound easy enough, but imagine doing this for every one of your games. I'm sure this would weed out several engine users.

It`s actually pretty easy to check out an opponent for cheating. Of course, you need a strong engine to check out the moves. The best choise will probably be Rybka 3, rated 3150, but other engines will do also. If your opponent has scored 85 % or more, compared to the engine, he might be cheating. But when you do this, you have to be thorough, considering a lot of different things. First of all, you start your check out after the theoretical opening phase. That will be book moves, not to be taken into consideration. So you have to figure where to start! Then you have to give your engine enough time to investigate every single move properly. Some moves are clearer than others, because there will only be 1 or 2 choices worth playing. When there`s several good choices, you must let the engine run for longer time. But in cases like this, more than 1 move can be a strong choice. Therefore you must rate it with a high percentage. That will be an approximately measurement. So the overall measurement will not be perfectly precise. A good idea will be to run over more than 1 game. That might give you a clearer picture. And then you will have to remember, that this is only a tool, not a perfect verification. There are actually strong and honest players out there! If any questions come up, I`ll be happy to try to answer them.

In other words, it is pretty hard to tell for sure whether someone is cheating in "correspondence" chess based on just one game.

Note that when Hans Berliner won the World Correspondence Chess Championship, he scored about 88% against top engines. But we know he wasn't cheating because back then, any master could beat the best engines (which all ran on mainframe computers). Even though he had an OTB rating of only 2350, this shows how well a diligent player can do, without cheating.

That`s right! DavidForthoffer, you can`t tell for absolute sure, when an opponent is cheating. Not the best example with Hans Berliner. He won the championship going on from 65-69. There wasn`t any strong engines back then. So it doesn`t tell how far an honest player can go against cheaters. And especially not today. Even Fischer and Kasparov will get beaten, I believe. Compared to Rybka 3, Kasparov will score approx. 91%. And he sure didn`t learn the game over night!

Is there an option on these packages to give the engine a 'character' - does it make a difference to the percentage if the computer is set to play aggressively, defensively or whatever?

jonnyjupiter: when you have to defend, you put it on defense and so on, according to the character of the position.

Parallel conversation problem alert.

There are two different ideas here. First, there is the score against a computer. Second, there is the correlation compared with computer moves.

What I don't understand is - what fun or satisfaction can you get , knowing that a computer chose your moves for you !! There's no skill in that - especially when the other guy is at the disadvantage of having to rely on his own feeble brain. These guys  might as well play themselves - if they're not already playing WITH themselves , that is!

A good and sound point, che-moi!

That`s right ozzie!

I'm not sure if this is possible on the site (or if it's already a feature) but I've seen it on other websites...

Computer Accounts.

i.e. accounts that gain a (Computer) tag and a different coloured name so that people know they'll be playing a computer.

Sure it won't remove cheating but it'll atleast make people feel a little more comfortable knowing that if people want to use a computer and they're honest, they'll go with a (computer) tagged account rather than a normal one.

Also in reply to the "scoring" against computers idea... I don't think that's an accurate way of "testing". The positions in amateur chess aren't the same as those that tend to arise in professional chess. Then different computers and different programs suggest different moves as being good.

As a similar comparison, I've been able to "score" 90%+ with GM games (i.e. choosing a player and predicting their next move during world championship matches and other high level tournaments)  does that make me a GM? I don't think so. Though I'd like to dream that it means I have GM potential. Hahaha

all questions answered here:

Cheating at Chess on Chess.com - Questions and Answers

Erik, very well put!

You should put my games though, especially the one on Live Chess recently where I placed my queen en prise for no reason at all. Qd5!! as black, with white's knight on c3. Anyways, he didn't take it. A computer would never engage in such tomfoolary, now would it?

Or would it?

This whole discussion of catching cheaters strikes me as absurd. Any one with a modicum of intelligence should understand that cheating by letting the engine play all your moves is only the (visible) tip of the iceberg. Pure idiot cheaters are no doubt catchable. But there's a whole invisible world of cheat possibility that is undetectable.  You will never, ever, be able to detect that someone has checked an engine first to determine whether they should offer draw or play on. You will never be able to determine that a player tested a speculative sac to see how it played out engine vs. engine before deciding to play it or not... especially if the player decides to play a "bad" sac because he sees that the refutation is very difficult and not likely to be found by his opponent. This kind of sophisticated cheating is not only not detectable, the moves (which don't correspond to the engine's first choice, at all) would seem to prove the player wasn't cheating.

Probably the simplest cheat, so simple and supposedly common that I heard one player admit openly he does it because he thinks all correspondence players do it, is to "blunder check" his moves. He comes up with his own moves, but checks with the engine to make sure each move isn't a "blunder" -- the thing is, to a chess engine, a move that steps into a hard-to-see four move combination evaluates as a blunder... it IS a blunder to master -- and that kind of "blunder checking" will push a mediocre player's rating into the stratosphere (assuming his opponents are not also cheats) It's grotesque cheating but impossible to check in individual games... it would require a lot of games to be checked, and a sophisticated analysis of moves and potential blunders.  

I applaud chess.com's efforts to root out cheaters. But I have no illusions that they can catch them all, or even most... they will only snare the most pathetic of them.

The cleverer idiots will elude detection, and one simply has to make peace with that, or not play correspondence chess, or play with engine use allowed and acknowledged. But I've met some wonderful players here and played the best chess I've ever played here... what a loss to me to leave because, yeah, I've also run acoss the occassional suspicious character. So far, the great guys outnumber the _possible_ jerks (I have no proof of any misconduct) at least 20-1. 

I think with a little common sense you can avoid most problems (players who never, or almost never seem to lose, strike me as good candidates for avoidance... when someone runs up an 80-3-1 record or some such nonsense, I think it's time to tell Mr. Ghost-of-Capablanca to consider joining an actual chess club and getting himself a real rating, since he seems to have "talent.") And if the occassional tournament draw is one of these gifted types... ehhh, deal, worse things have happened.

sorry for running on at the mouth...

How the +&£* do you 'check' to see if someone is using an engine? All you can do is ask them. So far as draws when ahead in material, I often offer draws to weaker players when substantially ahead materially and positionally because I feel that it's either a) boring to play with a queen, three minor pieces and a pawn to the good or b) it isn't fair to beat someone whose rating is so much lower or c) I don't want to discourage someone just starting out who has shown a hint of ability.

Note that many players resign when offered such a draw so it saves me much time. As far as losing rating points, so what?

chawil: By using an engine yourself, to determine how good your opponent`s move was according to lets say Rybka. See post 383.