Chess.Com getting sued.

[NickFaulks]

They are probably aware that if they ban a non-US player, any legal action would be in a US court, which would be very expensive, and the court would probably favour a US person.

I don't see why CAS wouldn't take the case, perhaps not Justin's but the Sunway Sitges example, where cash prizes are involved. Chess.com might then be in difficulties.

[Roger de Coverly]

2) At the very highest level of the professional game where very advanced security level could probably be implemented, paid for out of the enormous savings available from avoiding expense on venues and associated costs in particular.

Outside of the current crisis, a high budget event may well be able to afford the services of a local arbiter as physical witness to no consultation.

[Roger Lancaster]

It’s not really a direct answer but fundamentally any attempt to detect cheating using circumstantial evidence alone, based on single games is flawed. It can only seriously be attempted based on a pattern evidenced over (at a minimum) several games.

Even relatively weak players will sometimes play nearly faultless games. Even the strongest players will not do so all the time.

I agree but two points come to mind. First, because of username anonymity, Chess.com presumably can't be sure whether a user is a grandmaster or a total novice. As a mere human, I'd find that knowledge useful in deciding whether a "nearly flawless game" was likely to be bona fide or not.

Secondly, what evidence is there that Chess.com actually look at "a pattern evidenced over (at a minimum) several games"? And, if they do, which games? For example, if I've played 50 games recently with four showing a high correlation with computer assistance but the other 46 much lower, do Chess.com base their assessment on the four games or [and this would likely arrive at a different conclusion] the complete range?

[Roger Lancaster]

They are probably aware that if they ban a non-US player, any legal action would be in a US court, which would be very expensive, and the court would probably favour a US person.

I don't see why CAS wouldn't take the case, perhaps not Justin's but the Sunway Sitges example, where cash prizes are involved. Chess.com might then be in difficulties.

FIDE article 11.1 states that "The players shall take no action that will bring the game of chess into disrepute"

Now, if I publicly asserted that someone was a cheat, I would expect to be challenged as to evidence. If I adamantly refused to disclose any evidence, and particularly if I persisted in my assertion, I would expect to be liable to sanction as violating article 11.1. Of course, FIDE might very reasonably regard me as insufficiently important to worry about - but it would be another matter if a top GM publicly called out another top GM.

Article 11.1 relates only to the conduct of players but presumably has an implied extension to others. For example, if an arbiter made a similar unsubstantiated assertion, one wouldn't expect him or her to escape sanction on the grounds of not being a "player" - and presumably, in principle, that doesn't apply uniquely to arbiters.

I realise that FIDE has limited, if any, power [beyond a pronouncement] over the activity of an online chess platform. However, if a player publicly accusing another player of cheating would offend against its statutes, can FIDE pretend not to notice when an online chess platform does just this?

[Joseph Conlon]

I don't know how the engine detection codes at lichess or chess.com work, but examples where I would regard engine correlation, by itself, as highly suspicious would be

(a) Someone who has the choice of an easy exchange off into a trivially won endgame (say K R + pawns vs K + pawns), but instead plays out a tactically complicated mate in 8 - I think the latter choice is essentially inhuman.

(b) Perfect engine matching in an endgame like K+R vs K or K+ B+B vs K, where everyone can give checkmate but probably via very different move orders.

(c) (at blitz) Perfect play of sharp tactical lines across multiple openings. Everyone has pet lines they have prepared, know well and can play accurately, (and I agree with previous posters that engine matching for e.g. correspondence chess means little to nothing) but to do this in game after game across the whole range of openings requires either strong GM strength or an engine.

There are other (in my opinion) clear tells of engine use - e.g. playing excellently through sharp middle games at 5 seconds / move and then continuing with the same 5 seconds / move once a queen up with obvious routes towards checkmate. Therefore if I suspect engine use I always play until checkmate rather than resigning, as I think this makes naive engine use really obvious (who needs 5 seconds/move to play out K+Q vs K?)

I agree online chess is not suitable for serious long-play games with proper money at stake, but in certain ways it is much superior to OTB chess - always available and a means for those like myself who no longer / can't play OTB chess due to family commitments to still play. There's also the chance to play in the same events as some of the strongest players in the world and even occasionally play them - eg via lichess last year I got to play Carlsen at blitz, which is the sort of experience I would never have via any conventional OTB tournament.

[Jonathan Bryant]

30 players got kicked out of the Sitges online tournament apparently

[Adam Raoof]

30 players got kicked out of the Sitges online tournament apparently

Do you know how they determined who to kick out?

[John Hodgson]

Possibly this has been raised elsewhere, but I wondered about this.

"Internet chess provider anti-cheating software will be in place, with provider sanctions and rights of appeal (if any) as set out in the provider’s terms and conditions; in the event that a player is banned by the provider any appeal must be made to the provider (if permitted by the provider’s terms and conditions). The 4NCL has no involvement in this process." (4NCL website)

Remembering the banned player on this thread: I do not know whether Lichess has the same issues as chess.com, but if so this raises the possibility that players may be banned who have not actually cheated. We are relying on an online provider without appeal to the 4NCL or ECF, particularly as we do not seem to know how the anti-cheating software works (does the provider give precise reasons for the ban)?

Quite a few refugees from the OTB world started out at 1500 and have played each other (I have sunk to below 1400). The amount of theory in these games might be more than expected of a 1500, even if there is no other evidence of cheating. I do not know whether this would be sufficient on its own to result in a player being banned.

Before covid-19 most players could play both OTB or online chess if they wished, and perhaps an online ban was not particularly onerous. But under current conditions banning players in effect means that they are unable to play either form of competitive chess. I would not wish to exaggerate this or think I am referring to myself, but this might upset someone's mental health, particularly as it would be considered unjust. The 4NCL and ECF should consider their policy in this regard.

[David Sedgwick]

Possibly this has been raised elsewhere, but I wondered about this.

"Internet chess provider anti-cheating software will be in place, with provider sanctions and rights of appeal (if any) as set out in the provider’s terms and conditions; in the event that a player is banned by the provider any appeal must be made to the provider (if permitted by the provider’s terms and conditions). The 4NCL has no involvement in this process." (4NCL website)

Remembering the banned player on this thread: I do not know whether Lichess has the same issues as chess.com, but if so this raises the possibility that players may be banned who have not actually cheated. We are relying on an online provider without appeal to the 4NCL or ECF, particularly as we do not seem to know how the anti-cheating software works (does the provider give precise reasons for the ban)?

Quite a few refugees from the OTB world started out at 1500 and have played each other (I have sunk to below 1400). The amount of theory in these games might be more than expected of a 1500, even if there is no other evidence of cheating. I do not know whether this would be sufficient on its own to result in a player being banned.

Before covid-19 most players could play both OTB or online chess if they wished, and perhaps an online ban was not particularly onerous. But under current conditions banning players in effect means that they are unable to play either form of competitive chess. I would not wish to exaggerate this or think I am referring to myself, but this might upset someone's mental health, particularly as it would be considered unjust. The 4NCL and ECF should consider their policy in this regard.

Very well put, John.

I and others made similar points a few days ago:

I may be wrong, but I have the impression that a significant proportion of those who have been banned in OTB chess are players of master strength and above. Honest players of that strength - the vast majority - are very keen to see the cheats exposed.

Nevertheless, the FIDE Fair Play Regulations are designed to protect the innocent as well as to identify and punish the guilty. Chess.com don't appear to care about the former at all and it doesn't seem to be of much importance to Lichess.

For most people, being banned from Chess.com would not affect their employment or other earnings. In my case, as a Category A International Arbiter, it could result in my never working again.

I am intending to play in the forthcoming European Online Chess Championship organised by the European Chess Union, but only because in that event the decision on whether to expel a player rests not with Chess.com but with the Chief Arbiter of the tournament.

[Richard Bates]

I don't know how the engine detection codes at lichess or chess.com work, but examples where I would regard engine correlation, by itself, as highly suspicious would be

(a) Someone who has the choice of an easy exchange off into a trivially won endgame (say K R + pawns vs K + pawns), but instead plays out a tactically complicated mate in 8 - I think the latter choice is essentially inhuman.

Not sure about this one at all - think you would have to qualify this substantially by the nature of the game and the motivations of the individual. Chess is not simply about competition and winning, it is also for many people about creativity and art - the latter often delivering more enjoyment than the result. And even more so once the importance of the result is taken out of the equation - which frankly is the situation for the vast majority of online chess. And what’s more, sometimes playing brilliant games and/or sequences of moves which are subsequently shown to be justified by computer analysis does not mean that they were played in the sure knowledge that they were sound. Human instinct can sometimes align with the calculating machines and result in the same outcomes.

[Jonathan Bryant]

30 players got kicked out of the Sitges online tournament apparently

Do you know how they determined who to kick out?

No idea. There was a reference made to the number of complaints of cheating made each round. I don’t know whether they only followed up those allegations or whether they were also proactive.

I saw a tweet from a chess news organisation of which I hadn’t previously heard. There was a series of tweets, one of which stated that the people were banned "because the probability of them using an engine was higher than 99,5 percent."

Which is obviously a completely meaningless statement.

Then then quote the Sitges organiser so I imagine their information came from him. And I wouldn’t be at all surprised if the Sitges organiser doesn’t have the faintest clue how people came to be banned and just accepts at face value what chess.com have told him.


I understand why organisers and chess.com (and similar websites) want to present this idea that detecting cheating is going swimmingly but it would be much better in the long run if they came clean.


My only real interest in the tweet was the number they say were banned. Of course whether 30 is "a lot" or "not many" depends on how many people were in the tournament. I’ve no idea about that but I’d want it to be about 3000 before I’d conclude that the number banned was negligible. And I assume it wasn’t.

[Jonathan Bryant]

The 4NCL and ECF should consider their policy in this regard.

Yup.

Again, it’s entirely understandable why they (and others) would adopt the "Nothing to do with us" shoulder shrug response and just contract out the cheating detection to the host sites. I suspect, however, that this attitude will sooner or later be tested and found wanting.

After all, if I was to hold a chess tournament in a venue and to the surpise of me and everybody else the roof should fall in during that chess tournament ... well that’s one thing.
If I were to hold a chess tournament in a venue at which i’d previously been told by many people that the roof was in danger of falling in ... well that’s something else.

Like David, if I was publicly banned from a chess website for cheating I’d very possibly lose my job. In those circumstances it might very well be worth taking legal action against those responsible (given that I wouldn’t actually have been cheating). If this was an ECF event or a 4NCL or whoever, then, yes I’d very possible be looking to include them as well as chess.com or whoever.

In a less extreme way, the reputational damage of being associated with processes that are patently problematic is also likely to increase motivations to a make a change.

[Roger de Coverly]

My only real interest in the tweet was the number they say were banned. Of course whether 30 is "a lot" or "not many" depends on how many people were in the tournament. I’ve no idea about that but I’d want it to be about 3000 before I’d conclude that the number banned was negligible. And I assume it wasn’t.

Details of the tournament are on chess-results.

http://chess-results.com/tnr526437.aspx ... ES&flag=30

A number of people quit before the tournament finished. Look at the number "not paired" in round 9. Reports on earlier rounds suggested that the organisation was somewhat chaotic.

[NickFaulks]

I understand why organisers and chess.com (and similar websites) want to present this idea that detecting cheating is going swimmingly but it would be much better in the long run if they came clean.

Not the nature of the beast.

[Jonathan Bryant]

Not the nature of the beast.

True. And I suppose it’s not realistic to expect the ECF to be on top of online cheating issues when they haven’t made much of a fist of dealing with them in the face to face world.