Cheating in chess

[Adam Raoof]

Where is the empirical evidence for this?

Yesterday's UK Chess Challenge Gigafinal (the first of two) run on Lichess has already labelled one highly-ranked competitor in each of the Under 12 Boys, Under 12 Girls and Under 14s as "This account violated the Lichess terms of Service". None are so labelled as yet in the Under 18s. These are players who qualified without apparent problems from Megafinals, but then the hurdle was far less (3 or 3.5/6, rather than 5.5 or 6/7).

See:

https://lichess.org/swiss/Dhw2gWEH
https://lichess.org/swiss/NvQLKDMg
https://lichess.org/swiss/BcY6WUbX
https://lichess.org/swiss/QdwFYdGq

This is not evidence that lichess does not wait for a decent sample before banning players based on assistance. These are based on 7 games, which is very decent.

[Alex McFarlane]

These are based on 7 games, which is very decent.

I would think that it is actually more than 7 games as the qualifying rounds were also available for analysis.
The possibility that at least some of these players were under suspicion from the qualifying games should not be dismissed. However, it was the recent results that may have tipped them over the edge.

[Roger de Coverly]

These are based on 7 games, which is very decent.

Not really, given that consecutive chess moves aren't independent of one another. The sample size is a key metric in determining that matchs to engine suggestions aren't just coincidence. For decent players, it's only the critical moves that suggest external advice.

I seem to recall that the Northumberland team withdrew from the 4NCL after only a handful of rounds because of an accusation made against one of their players.

[John Swain]

These are based on 7 games, which is very decent.

I would think that it is actually more than 7 games as the qualifying rounds were also available for analysis.
The possibility that at least some of these players were under suspicion from the qualifying games should not be dismissed. However, it was the recent results that may have tipped them over the edge.

It was mentioned in the Twitch commentary for the Megafinals that players, requiring 3 or 3.5/6, had been caught when they reached this target by cheating occasionally (there are two cases scoring 1 0 1 0 0.5 1 and 0 1 1 0 1 1).

[Matt Bridgeman]

One of the leaders after 3 rounds of the 4NCL Online Congress did not make it to round 4.

[Adam Raoof]

The bottom line is - it is easy to catch people who use assistance, especially online. What we need to work out is how to dissuade players, usually young players, from using assistance. This should be a priority for the ECF and for all of us, not least because chess may remain online for a long while.

It does not help them one bit - they learn nothing from it, and I worry that the attitude that "it doesn't matter, it's only an online game" will creep into other areas of their lives. I have seen juniors doing this for money (which is a criminal act), and for fun, and because it was a laugh (it's not fun for the losers) and all their mates were doing it. I have sympathy for young people who get suckered into using assistance, and I believe in forgiveness. But really, I am shocked at the attitude of some people on this Forum. These children are being groomed, by websites and individuals (often peers, coaches and even parents) that don't care what consequences they suffer as a result, as long as they go along, or qualify, or win money, or get a title they can put on their application for the next school. What kind of example does it set them? These are good kids, being made bad in real time right in front of us.

Adults I have less sympathy for. Most of the time I watch people posting puzzles on Facebook, and from the responses I can often tell who uses an engine to find the solution. It won't do them any good in the short or long run. But they are adults, and it's their call.

[MartinCarpenter]

This shouldn't be such a problem though? Its another reasons why you wait for a half decent sample of games before doing anything.

Empirical evidence suggests that lichess and chess.com don't wait.

Where is the empirical evidence for this?

The successful appeal above presumably demonstrates this to at least some extent? If it was based on a larger sample then I really doubt if he hit preparation in all of the games.

Of course the fact that there was an appeal is promising/comforting in some sense, but you really shouldn't have the anti cheat screening set up to have its conclusions vulnerable to that sort of appeal.

It'll make it incredibly hard to actually nail anyone non down, even if/when they are cheating.

[Alex McFarlane]

What we need to work out is how to dissuade players, usually young players, from using assistance.

I agree with the sentiment of the above sentence. However, I'm not sure of the approach.
In virtually every other online game using 'cheats' is regarded as acceptable. In fact, in some games it is virtually impossible to win without using some form of external assistance.

I think it would be too great a task to simply say that cheating is wrong and shouldn't be done. Instead, I think we have to educate youngsters that chess is not the same as many of their usual online games. The etiquette is different.

So, I'm suggesting that we don't try to convince them that getting assistance in general is wrong. Instead we educate them that chess is not like their usual games but requires a higher level of self achievement. A victory is only a victory if it has been achieved by their own skill.

[David Sedgwick]

In virtually every other online game using 'cheats' is regarded as acceptable. In fact, in some games it is virtually impossible to win without using some form of external assistance.

That is a very interesting point. It perhaps helps to explain the phenomenon, regarding which there seems to be considerable evidence, that players cheat in online chess who would not cheat in OTB chess.

I continue to hear officers of various chess organisations saying "We are not worried about cheating if we set up online competitions, because none of our players should ever cheat." I predict that during the months ahead a number of them will be sadly disillusioned, owing to what I have said in the previous paragraph.

[Roger de Coverly]

One of the leaders after 3 rounds of the 4NCL Online Congress did not make it to round 4.

There are some giveaways allowing the player to be identified, although not whether it was a lichess decision or a 4NCL one.

[John Swain]

In virtually every other online game using 'cheats' is regarded as acceptable. In fact, in some games it is virtually impossible to win without using some form of external assistance.

That is a very interesting point. It perhaps helps to explain the phenomenon, regarding which there seems to be considerable evidence, that players cheat in online chess who would not cheat in OTB chess.

I continue to hear officers of various chess organisations saying "We are not worried about cheating if we set up online competitions, because none of our players should ever cheat." I predict that during the months ahead a number of them will be sadly disillusioned, owing to what I have said in the previous paragraph.

I must plead ignorance about other online games and the acceptability of using 'cheats', but I agree that Alex's point is an intriguing one.

However, to continue with the example of the UKCC, I would make three points:

(1) The warning not to cheat was given in the Twitch preamble very clearly before each separate event. It was often pointed out how easy it is to identify cheats. The warning about cheating is also prominent in the regulations on the UKCC website, with an accompanying video.
(2) Children of varying ages can generally accept and abide by right/wrong concepts and that these vary across different activities.
(3) The under 8s and under 10s in the Megafinals were not found to be "violating the Lichess terms of service"; the problems arose with the older groups. Of course, this might be because the younger children were usually supervised by a parent, whereas the older ones were perhaps left to their own devices (literally!)

[JustinHorton]

The bottom line is - it is easy to catch people who use assistance

I am really not sure why you say so, unless we're neglecting the question of false accusations. Catching the guilty without fingering the innocent strikes me as really quite hard.

I have seen juniors doing this for money (which is a criminal act)

We've gone over this a few times on here, including on this thread, and while this may be true, it's not so easy to demonstrate so.

[JustinHorton]

In virtually every other online game using 'cheats' is regarded as acceptable. In fact, in some games it is virtually impossible to win without using some form of external assistance.

That is a very interesting point. It perhaps helps to explain the phenomenon, regarding which there seems to be considerable evidence, that players cheat in online chess who would not cheat in OTB chess.

Isn't it just much much easier, or are do you mean in the sense that people who think it ethically unacceptable OTB will do it online? (Even in that instance the issues aren't wholly separate, as one means of convincing people that it's acceptable online may be the belief that everybody else is doing it.)

[Roger de Coverly]

In virtually every other online game using 'cheats' is regarded as acceptable. In fact, in some games it is virtually impossible to win without using some form of external assistance.

It's how authors and publishers of games have managed to acquire an income stream and build a (big) business. Paying extra gives in game benefits not available to those not paying. The chess equivalent might be that for a fee, you could access an opening tree or endgame tablebase whilst in play.

[David Sedgwick]

Isn't it just much much easier, or do you mean in the sense that people who think it ethically unacceptable OTB will do it online?

I consider that both points apply, but I was focussing on the latter.