Chess Player Strip Searched

[Roger de Coverly]

I repeat that it would be helpful to chess if you and others here made your comments directly to the ACC, which does contain people who would listen.

The ACC have no public forum and have made no statements inviting submissions, In any case why would they take any notice of random submissions by chess players? Hypothetically you could write to them and they could reply. If that isn't in public, they could say the opposite to the next correspondent. FIDE itself has yet to publish the draft paper. If it does so, will it accompany it with caveats that it's a draft for discussion and unlikely to be adopted as written?

[NickFaulks]

Nick, forgive me for saying so, but insulting Roger and me for not going directly to FIDE when neither of us have a mandate to do isn't really very helpful. Of course with my 4NCL hat on I'm raising my concerns with my national federation - that was the purpose of my discussions with Dave Welch as the ECF's chief arbiter.

I think it is Shaun and I who have been insulted. We have told you that any reasoned comments on the ACC's proposals would be greatly welcomed, and the reply seems to be that we are viewed as a couple of unprincipled liars who would use any such communications for some perverse purpose.

Of course you both have mandates - you each have a FIDE ID. In your case, your known 4NCL position would add weight, even if you felt the need to say that the views expressed were solely your own ( which I'm sure they're not ).
.

[Mike Truran]

the reply seems to be that we are viewed as a couple of unprincipled liars who would use any such communications for some perverse purpose.

And which part of any of my posts has in any way suggested that you and Shaun (whoever he is) are "unprincipled liars"?

[NickFaulks]

the reply seems to be that we are viewed as a couple of unprincipled liars who would use any such communications for some perverse purpose.

And which part of any of my posts has in any way suggested that you and Shaun (whoever he is) are "unprincipled liars"?

None of yours. I didn't actually say that you had, but I realise that it looks like that and apologise unreservedly.

[John McKenna]

Who is Shaun?

I think this may be he -

In some cases (including my own), members can wear all three hats.

Who are you, and what hats do you wear?

Apologies for barging in without a proper introduction. Apart from the kind description from Stewart, I organise a weekly rapid event in Canberra (where I live), plus a weekly chess club. In all these events, people can draw when they wish, phones going off do not lose the game (unless they then ring a second time), and players normally have 30 minutes to show up for their games. And when I make proposals to FIDE concerning rule changes, they normally start with "At the local events I run ...."

The above is from the thread linked to below -

http://www.ecforum.org.uk/viewtopic.php ... 76#p100976

[Roger de Coverly]

Who is Shaun?

Shaun Press is an Australian player and organiser, based, I believe, in Canberra. The reason he is a person of influence is that he also has eligibility for Papua New Guinea, which gives him rather more clout over FIDE than the average Australian chess player.

He's not in this year's PNG team, being replaced by a Yorkshire player.
http://chessexpress.blogspot.co.uk/2014 ... to-go.html

[shaunpress]

Mike Truran has sent an email to members of the FIDE/ACP Committee* listing concerns with the proposed regulations as they apply to the 4NCL. This has kicked off a discussion about these points, and hopefully this will result in improved solutions. Many thanks Mike.

There are a couple of points related to the overall discussion concerning the regulations that are also worth considering. Firstly, the problem that the FIDE/ACP ACC is tasked to solve is the use of computer assistance during games. Obviously the ACP think this is a serious problem, as the livelihood of their membership depend on a fair playing field. But as not everyone is a chess professional, should such regulations apply to all events? To my mind the compromise is to let the organisers decide whether they wish to implement such measures or not. However such an approach does create its own issues. Should organisers be required to inform all entrants that "FIDE Anti-cheating regulations are not being used in this event" so that players can make an informed decision about whether they will play or not? Depending on the level of an event there is of course an expectation that organiser adhere to certain standards (eg you don't have to bring your own equipment to a 4NCL event), so are Anti-cheating measures part of those standards?
The second point is about the deliberations of the committee. In fact there are two bodies that exist for the moment, although the distinction is fairly minor. There is a FIDE/ACP Anti-Cheating Committee, which is comprised of both FIDE and ACP people. There is also the FIDE Anti-Cheating Commission (ACC), which consists of FIDE appointees. There is a degree of overlap between the two, but the FIDE ACC will be the ongoing body from August 2014. For now any comments (including my own) are part of the discussion process, and are in no way official pronouncements. We all speak as individuals, just as no one on this forum would presume to speak on behalf of all English players.

Shaun Press
Secretary, FIDE ACC

[Ian Thompson]

There are a couple of points related to the overall discussion concerning the regulations that are also worth considering.

I think the biggest issue in these regulations is the use of statistical evidence to convict someone. Three points:

1. Will the game analysis procedure detect the 'occasional cheater', particularly if they are a good player to start with? For example, a player whose true strength is 2500 who uses computer assistance only on critical moves during a game, so maybe 5% of his moves are computer assisted?

2. The regulations state that the game analysis tool has a false positive error rate of 1 in 1,000,000 and then goes on to say that this means a false positive will occur once every 20 years, based on the number of games currently recorded. That is a clearly flawed calculation:
a) FIDE may currently only get 50,000 games per year, but that number would increase under these regulations.
b) Under these regulations FIDE would received a biased set of games (i.e. a higher than average number of games where unusual results occurred), so the error rate on the games received would be higher than on an unbiased set of games.
c) In view of b), I think the false positive rate would be more accurately calculated based on the total number of games played/rated. That would result in a much more frequent false positive rate than once every 20 years.

3. If a convicted person took their ban to a court (e.g. the CAS) would they uphold a conviction based primarily on statistical evidence?

If the answer to either point 1, or point 3. is "No", then the regulations are going to be ineffective.

[MartinCarpenter]

Well you can't even begin to work out the false positive rate until you know the actual rate of cheating The ratio of valid to false accusations rather matters of course - even rather small false positive rates can make for huge problems for screening procedures etc.

I presume that it'll be used to inform/target further investigations which makes it all much less crucial. Occasional cheaters can't really be caught except red handed somehow At least it'll deal with the idea of people just playing computer moves a lot.

[NickFaulks]

I think the biggest issue in these regulations is the use of statistical evidence to convict someone. Three points:

1. Will the game analysis procedure detect the 'occasional cheater', particularly if they are a good player to start with? For example, a player whose true strength is 2500 who uses computer assistance only on critical moves during a game, so maybe 5% of his moves are computer assisted?

I may be alone on the committee in finding the statistical approach a bit of a sideshow. They are very excited because it is new and scientific, and looks like something you can actually do. In my opinion it addresses ( and, to be fair, probably solves ) last year's problem. Yes, the small community that uses a computer every move to replicate its play will be found out and, once they know that they're going to be found out, deterred. That's a good thing so far as it goes, but point 1 above is correct. The sort of cheating about which we should really be worried will not be spotted, certainly not to the point where is it overwhelmingly significant.

[Paolo Casaschi]

They are very excited because it is new and scientific, and looks like something you can actually do.

In order to qualify a testing procedure as "scientific" there's a fundamental requirement of transparency. This means the details of the testing procedure should be publicly available up to allow anyone to replicate those tests and their results.
The proposal paper has some vague reference to peer review of the methodology but restricts the access to the ACC testing server to selected official and arbiters. In similar contexts it is often argued that secrecy of the testing procedure is mandatory to avoid cheaters working around the tests. Unfortunately, a secret testing procedure, defined and operated by the ACC prosecutors and impossible to replicate by anyone else would completely undermine the legitimacy of those procedures. Especially since those test results are suggested to be used as sole evidence for a cheating conviction. For comparison, it would be as if the athletics anti-doping tests were run exclusively by a single laboratory; if that laboratory checks the athletes blood for undisclosed substances, using undisclosed procedures and ultimately decides if the athletes has used performance enhancers; on top of that, that single laboratory is owned and operated by the prosecutors in cheating accusations.

[Paolo Casaschi]

I presume that it'll be used to inform/target further investigations which makes it all much less crucial.

Check your assumptions, the paper recommends the statistical analysis (with the given risk of false positive) to be used as sole evidence for a conviction. In other words, in Little Britain style, you are banned for 5 years because "computer says so"

[Roger de Coverly]

The ratio of valid to false accusations rather matters of course - even rather small false positive rates can make for huge problems for screening procedures etc.

Let's not also lose sight of the problem that a game can legitimately be the product of computer analysis. At least three possible reasons, (a) you recall and are following your home analysis, (b) you are following a line recommended by an author or (c) you are following a previous Correspondence game.

[Matt Fletcher]

The ratio of valid to false accusations rather matters of course - even rather small false positive rates can make for huge problems for screening procedures etc.

Let's not also lose sight of the problem that a game can legitimately be the product of computer analysis. At least three possible reasons, (a) you recall and are following your home analysis, (b) you are following a line recommended by an author or (c) you are following a previous Correspondence game.

But surely all of these can be discounted as false positives, hopefully before but if necessary after any allegation (player can point to the analysis / book / previous game). And what are the chances that someone would manage to follow analysis for a whole game in this way for a significant proportion of a tournament?

[NickFaulks]

I presume that it'll be used to inform/target further investigations which makes it all much less crucial.

Check your assumptions, the paper recommends the statistical analysis (with the given risk of false positive) to be used as sole evidence for a conviction. In other words, in Little Britain style, you are banned for 5 years because "computer says so"

It does say that, but I don't believe that is really Ken Regan's intention, and it's his baby. The wording needs tightening up to confirm that the test will only be used to convict when there is independent supporting evidence, but in reality I don't think this is a frightening part of the paper.