Cheating in chess

[NickFaulks]

To my mind that reasoning is flawed particularly when it claims performance in a single game.

Leaving aside the fact that the entire z-score approach is flawed, relying on an assumption of normal distribution which is completely unjustified.

That is of course also true of the rating system, but when you are talking about data within 2 sigma of the mean it doesn't matter much. When you are invoking variations of 4+ sigma, it really does!

Five years ago, when I discussed this point with an earlier head of Fair Play, her response was that they had no choice but to persist with their z-score, because without it they had nothing.

[Alex Holowczak]

You do wonder what's going on here.

This:

Related reading: https://wearetwofamilies.org/our-journey

In a nutshell, the Fair Play Commission, along with their supervisor on the Management Board, believed that the way to solve Fair Play in chess was to take the responsibility for its execution at FIDE events off arbiters. Their reasoning for this is that arbiters both don't care and don't know what they're doing. They've now understood that arbiters do care, but even now they persist with the second reason.

Things had been rumbling on for a long time, with Fair Play reporting arbiters to the Ethics Commission, attempting to appeal their decisions to the tournament Appeals Committee, and so on. I first became aware of the mess at the Olympiad last year, where there were a number of forfeits imposed on players for silly reasons.

In any event, most of the world's top arbiters believe that fair play is, and always was, the arbiter's job and not a separate job - and this was a gross misstep from FIDE as a policy decision. If the training needs improvement, then no problem - FPL should work with ARB so that they can deliver better training. In the same way that TEC should be working with ARB to improve training in tie-breaks, or QC should be working with ARB to improve training on rating and titlenorms.

You complain about z-scores, but you should understand that policy included them banning sim cards, vapes, sunglasses, pens, and all kinds of other trivial items. These were far more destructive policies, which to the best of my knowledge, persist at FIDE events to this day to some degree. It isn't clear to me the extent to which a Chief Arbiter can simply ignore all this and do something more sensible, but the sort of Chief Arbiter that would hasn't been appointed to a FIDE event in recent times, so this remains an unanswered question in my mind.

Leaving aside the fact that the entire z-score approach is flawed, relying on an assumption of normal distribution which is completely unjustified.

It depends how you use it. If you are using it in an OTB tournament to advise you on who to look out for, it's fine.

[NickFaulks]

It depends how you use it. If you are using it in an OTB tournament to advise you on who to look out for, it's fine.

I understand that, but they insist on a veneer of accuracy which does not exist.

It does not inspire confidence that FPL regulations contain gibberish like

"For online chess, a threshold of 4.25. This represents a natural
frequency of 0.000011, one in almost 100,000. Considering the
observed rate of cheating in online chess, this yields appraised
odds about 1,000-to-1 against the null hypothesis of fair play,
toward the high end of the 99% to 99.9% confidence range
regarded as meeting the CAS criterion of “comfortable
satisfaction” in the context of chess.
b) For over-the-board chess, a threshold of 5.00. This represents
a natural frequency of one in almost 3.5 million, and is similarly
placed with regard to an in-person observed cheating rate."

[Alex Holowczak]

It depends how you use it. If you are using it in an OTB tournament to advise you on who to look out for, it's fine.

I understand that, but they insist on a veneer of accuracy which does not exist.

It does not inspire confidence that FPL regulations contain gibberish like

"For online chess, a threshold of 4.25. This represents a natural
frequency of 0.000011, one in almost 100,000. Considering the
observed rate of cheating in online chess, this yields appraised
odds about 1,000-to-1 against the null hypothesis of fair play,
toward the high end of the 99% to 99.9% confidence range
regarded as meeting the CAS criterion of “comfortable
satisfaction” in the context of chess.
b) For over-the-board chess, a threshold of 5.00. This represents
a natural frequency of one in almost 3.5 million, and is similarly
placed with regard to an in-person observed cheating rate."

But this doesn't really effect people. (Until it does, I suppose!) On the other hand, the practices imposed at the Olympiad, and the event in Prague earlier in 2025 which resulted in another protest letter signed by many of the teams in the Seniors in one of the playing halls, were a real issue.

[Roger de Coverly]

It depends how you use it. If you are using it in an OTB tournament to advise you on who to look out for, it's fine.

I suppose at OTB tournaments, other activities would have to be a possible cause for concern. These might include being absent from the board for long periods out of sight, unusual clothing or behaviour. It occured to me that if a player's excuse for absence was frequent toilet visits then perhaps they should be offered half point byes or withdrawal from the tournament.

I never did understand the objection to sim cards. If the point was that it could be mated with a phone to provide a banned device, then surely a phone even without a sim card can run chess sottware and it's the phone that's the problem not what's inside it. I'm aware that the Appeals Committee at the Olympiad overruled the Fair Play Commission on this one.

[Alex Holowczak]

It depends how you use it. If you are using it in an OTB tournament to advise you on who to look out for, it's fine.

I suppose at OTB tournaments, other activities would have to be a possible cause for concern. These might include being absent from the board for long periods out of sight, unusual clothing or behaviour. It occured to me that if a player's excuse for absence was frequent toilet visits then perhaps they should be offered half point byes or withdrawal from the tournament.

I never did understand the objection to sim cards. If the point was that it could be mated with a phone to provide a banned device, then surely a phone even without a sim card can run chess sottware and it's the phone that's the problem not what's inside it. I'm aware that the Appeals Committee at the Olympiad overruled the Fair Play Commission on this one.

Well, there we have it. After 15 years on here, we finally agree on something. If that doesn't highlight the nature of the mess FIDE got themselves into with this, nothing can.

[NickFaulks]

Another issue, which has not been mentioned and about which Alex may know more than he can reveal, is financial. There is presumably a finite amount of money to be shared between Fair Play Arbiters and ( as I would call them ) real Arbiters, with both wanting as much of the pool as possible.

[Kevin Thurlow]

"FIDE – the International Chess Federation – remains fully committed to fair, transparent, and unified standards in chess."

Really, when did they start?

It is nice to see a thread when I think the recent posters are talking sense!

[Alex Holowczak]

Another issue, which has not been mentioned and about which Alex may know more than he can reveal, is financial. There is presumably a finite amount of money to be shared between Fair Play Arbiters and ( as I would call them ) real Arbiters, with both wanting as much of the pool as possible.

At an online FPL Commission meeting during a FIDE Congress during covid, it was made very clear to those in attendance that FIDE wasn't paying them for their work on the Commission, and so they needed this to be paid and rewarded in lieu of being paid by FIDE for Commission work. This is now denied, of course, but I commented on it in the Zoom chat at the time and Arkady responded privately. I won't disclose what he said.

There were numerous objections to the cost of the additional staff that organisers didn't want. The organisers who bid for FIDE events also submitted a corresponding petition complaining about that. But it certainly wasn't a finite sum being split up amongst more people - these were additional fees that organisers were expected to bear.

[NickFaulks]

these were additional fees that organisers were expected to bear.

Yes, I do remember that. Even so, it feels a bit like the Chancellor saying that increased taxes on employers and landlords will have no impact on employees and tenants - that isn't how economics works.

[Helen Milligan]

At the recent World Seniors in Italy, where I was playing in the 50+ Open, the daily routine to get into the tournament hall included the Fair Play person squeezing my wallet. She told me this was to check for SIM cards and I told her how I felt about this ridiculous idea. As for lingering in the toilets to cheat - given the incredibly small number of toilets available, I think anyone lingering in one of them would have been ... well, shall we say treated harshly by the other players! (I hope that limiting toilet facilities is not a deliberate Fair Play policy, by the way!)

[Roger de Coverly]

There was a USCF caes where the Regan analysis was a factor in deciding to investigate.

http://ethics.fide.com/wp-content/uploa ... -Yukta.pdf

from which

4.1. During the Universal Open held during June 20-23, 2024, the Chief Arbiter detected RF transmissions in the vicinity of the Respondent. When the Respondent and her opponent were pulled aside in a private area of the playing hall, the Respondent was found with a mobile phone on her person, hidden by her thigh. A subsequent examination of the phone by tournament staff revealed that a chess app had been active. The Respondent was already under observation following remarkable results in a previous tournament. Upon review of her games Professor Regan’s statistics revealed a highly clear positive indication of cheating from the z-score results.

That suggests there exists a gadget that can establish whether there is an active phone or at the very lest something transmitting in the vicinity. It makes sense as phones can talk to other nearby phones and other devices with tranmitters or receivers. It could be difficult to interpret given that the arbiters would likely have an active pairing or display computer and the boards themselves would be transmitting if the wireless type.

[Alex Holowczak]

At the recent World Seniors in Italy, where I was playing in the 50+ Open, the daily routine to get into the tournament hall included the Fair Play person squeezing my wallet. She told me this was to check for SIM cards and I told her how I felt about this ridiculous idea. As for lingering in the toilets to cheat - given the incredibly small number of toilets available, I think anyone lingering in one of them would have been ... well, shall we say treated harshly by the other players! (I hope that limiting toilet facilities is not a deliberate Fair Play policy, by the way!)

I was present at the World Rapid & Blitz Team Championship in June, where wallets were banned. "Not because we think you're cheating, but because it takes us time to scan them."

Unfortunately the messaging didn't quite work, so lots of players came down with wallets, only to be turned away to their annoyance. During round 1, I observed members of tournament staff (i.e. volunteers, organisers) take possession of several wallets belonging to players, and waiting around in the lobby for the players' games to end so they could give them back. Luke McShane was a few minutes late and being barred from entry due to having his wallet, whereupon Malcolm went up to the gate, snatched his wallet off him, and dragged him into the playing area. The Fair Play person standing there could only laugh. To be fair, that was my reaction too.

But this is the point - nothing has really changed. Your experience in Italy is not dissimilar from the experience of many others in FIDE events around the world. It is completely out of proportion, but it is all part of the training administered at FP Seminars. Another reason to abolish them.

[Alex Holowczak]

There was a USCF caes where the Regan analysis was a factor in deciding to investigate.

http://ethics.fide.com/wp-content/uploa ... -Yukta.pdf

from which

4.1. During the Universal Open held during June 20-23, 2024, the Chief Arbiter detected RF transmissions in the vicinity of the Respondent. When the Respondent and her opponent were pulled aside in a private area of the playing hall, the Respondent was found with a mobile phone on her person, hidden by her thigh. A subsequent examination of the phone by tournament staff revealed that a chess app had been active. The Respondent was already under observation following remarkable results in a previous tournament. Upon review of her games Professor Regan’s statistics revealed a highly clear positive indication of cheating from the z-score results.

That suggests there exists a gadget that can establish whether there is an active phone or at the very lest something transmitting in the vicinity. It makes sense as phones can talk to other nearby phones and other devices with tranmitters or receivers. It could be difficult to interpret given that the arbiters would likely have an active pairing or display computer and the boards themselves would be transmitting if the wireless type.

The arbiter found the mobile phone on her person, so she was in possession of a device she should not have been. So we are already at "loss of game". The question then, is "Was this person cheating?" The arbiter found a chess app was active on the phone, which already raises the likelihood of cheating. Then they found a high z-score. In other words, everything was pointing in the same direction; it wasn't a unilateral decision made by the statistics, or even made by the finding of the phone. I don't see that as objectionable.

[NickFaulks]

There was a USCF caes where the Regan analysis was a factor in deciding to investigate.

This was a 1600 player who suddenly scored +10 =1 -0 against 1730 opposition. If you are happy for a player to be put on watch after "a remarkable string of results" ( which I am, although I don't just mean 4/4 ) then you don't need Regan analysis to back it up.