That IS a good idea!
One way to catch a cheater
Sort:
Apr 25, 2008
0
#3
i (At 1500 some) could figure out why a computer did something. but i probably couldn't find the move myself.
But what if a clever cheater has a friend or pays a strong Chess Player at a club cash to explain the moves of the game for him he just cuts and pastes it to the sight and pretends he did it there's no way for anyone to prove it wasn't him.
Apr 25, 2008
0
#5
The point is, despite being imperfect, it WOULD work towards detering cheaters!
The problem is there's no real way of stopping it. I reported a totally blatant case of cheating, 36 straight victories jumping from 13 to 19 hundred, and absolutely nothing came of it. Even if it's obvious there's no proof.
Apr 25, 2008
0
#8
Well actually.. that's not proof... nor is it 'obvious' ...
I personaly jumped from 1600 to almost 2000 in just over a month with only 1 or 2 losses in that time.. my 'excuse' is that I was in a tournament with a bunch of 1600's and had no issue beating them all.. plus I was studying my opponents opening preferences, it doesn't take long and REALLY helped my game (against that individual)..
Point is, there is only 'proof' if you see someone consistently matching up with an engine ... after that, if they can't present a good argument of innocence, I say ban them!
I agree with PerfectGent. I play around 1400. Pretty sad numbers, so you know I don't cheat, Lol! But even I usually understand why a computer or human opponent did something when I look back on it. I can write in chess notation and reviewing a game can see how someone set up a combination. If I wanted to cheat, this rule wouldn't stop me.
ih8sens wrote:
Well actually.. that's not proof... nor is it 'obvious' ...
I personaly jumped from 1600 to almost 2000 in just over a month with only 1 or 2 losses in that time.. my 'excuse' is that I was in a tournament with a bunch of 1600's and had no issue beating them all.. plus I was studying my opponents opening preferences, it doesn't take long and REALLY helped my game (against that individual)..
Point is, there is only 'proof' if you see someone consistently matching up with an engine ... after that, if they can't present a good argument of innocence, I say ban them!
I checked it out in some detail before i reported - the period before the sudden change in ability he was regularly losing to players rated 12-1400, matching his rating which had been steady for a few months. Then after the change he beat all comers all the way up to 1900s. About two days after I accused him his month long 36 game winning streak came to an end, now he's just maintaining his inflated rating. I analysed a half dozen games and though they didn't match my particular engine (chessmaster) it did report 0% error and no disagreement.
As I clearly stated already, it's not proof. But it is obvious.
Apr 25, 2008
0
#11
that could be deemed proof... depends.. it's one of those things that is slightly circumstantial and needs some giveaway somewhere (misc. opening choice, long book lines, greedy play) ...
Well i dont know about obvious, but it was certainly very suspicious when you consider the sort of stuff the person had been posting in the forums. hmm, lets see if i can make up an example:
***
Hello everyone, i was wondering what opening this is?
What about this?
Also, what name is this opening?
****
It's still hard to tell though, because someone might have just started taking advantage of things allowed in cc, like opening books/databases/analysis boards. But in this case, you'd have at least expected some draws due to solid play from opponents.
Any programmers out there? Wonder if a tracking cookie could be made to monitor games being played against games being analyzed? Would stop the single puter user. Then only games being monitored would be rated. Probably would grow into a small program that would need to be downloaded.
A noble idea, but as a frequent chess program player, I can tell you that chessmaster, for example, has a wonderful auto annotation feature and one could just cut and copy the chessmaster's comments. These are not just basic comments, but a very in depth analysis of each move.
hehe, i think when someone annotates their opponent's move with "Yikes!" we could rest our case 
Apr 25, 2008
0
#16
just this week we finished our anti-cheating tools :) we have already banned a few people. over the next few weeks we will be fine-tuning the code to automatically catch cheaters. but trust me - it is only if there is OVERWHELMING evidence. and it IS easy to tell.
erik, do you have some kind of computer program you use to catch cheaters, or is it just based on weather their moves match with a chess engine?
I have been playing online chess for quite a while, and know a little bit about programming--dont let anyone fool you when they say they have anti-cheating tools. The "tools" they refer to are nothing more than this: If enough people complain about a certain person, and if his/her rating is too good to be true, and if the player is copping an attitude, then the "tools" caught the cheater. It is nothing more than speculation and logic. The "tools" are probably right more than they are wrong, but there is no form of software in existence than can detect if someone is using a chess program or not.
And what would be "anti-cheating" tools, Erik?
1. analyze window focus? Although it is lame, playchess "detects" cheaters using this method :))
2. analyze moves and match them against an engine? What if I play a game, I make 5 moves told to me by the computer and then I make a weaker move? But not weak enough to loose. Or I use a weaker engine which plays at let's say 2300. Enough to blow up almost everybody... How would then the magic tool react??
When playing online I was surprised especially by the quality of the opening repertoire. Often my opponents played the best lines in the opening, even when I tried fancy gambits like Smith-Morra or Cochrane. I guess that they were using a database (because after the first 10-15 moves many of them started to make weaker moves). And of course, there are cheaters. On every chess site. I played with some of them...
Ussually cheaters are willing to pay for accounts (because they want to show to the world how smart they are). That's why many chess sites (the big ones mainly) do tolerate cheaters. It's funny, but that's how it is. "OK, you cheated, we deleted your accoun, but you can make another account (if you pay). And if you promise you will not cheat again." Yeah, right :))
On another chess site that I play on the winners of larger tournaments say 100 or more may be asked to annotate the winning game to show understanding of their moves. Sounds like a good idea to me. What do you think?