4NCL Online

[Neil Graham]

Notification has just been received of the withdrawal of another squad just two days before the final rounds of the preliminaries. This will actually affect play-offs and relegation issues as three teams are involved - this withdrawal means that the teams concerned will finish on zero points.

Does anyone know whether there is an official statement about this?

Alex H has sent out an e-mail stating that the squad concerned has withdrawn from the League "with immediate effect" and detailing the consequences of this that all the teams concerned "will be deemed to have finished in 8th place of any group they happen to be in". Obviously the tables concerned will have to be amended as soon as possible so everyone will be aware of what has happened. I have no idea whether an "official statement" will be made other than the clear advice sent out in the e-mail or whether we will receive another e-mail (cf the Northumbria Vikings one) from the party concerned.

* My post has crossed with Roger's. I purposely didn't name the squad involved.

[Roger Lancaster]

* My post has crossed with Roger's. I purposely didn't name the squad involved.

I had no hesitation in doing so as Alex's email identified them and so it was already public knowledge.

[John Swain]

Surbiton Chess Club has issued a statement:

http://www.surbitonchessclub.co.uk/news ... cement.htm

[Roger Lancaster]

My post timed at 9.07 included a reason [allegation of computer assistance] for Surbiton's withdrawal which was omitted from Alex Holowczak's email. Before anyone draws the wrong conclusion, I should make it clear that my source for this extra piece of information was not Alex.

[Roger de Coverly]

Surbiton have pulled out in protest at "computer assistance" accusation which they strongly deny

There may be a simple solution to all this. Is it the case that lichess don't monitor games flagged as "casual"? If that's correct, then lichess are removed from the decision process and it's down to the Regan black box which is at least under the control of 4NCL arbiters.

[Rhys Cumming]

“ Having investigated the matter, the Surbiton Chess Club committee has accepted the player’s account, together with the evidence he has provided, that he was not using external assistance and believe that a fair appeal process would exonerate him.”

Sounds like they already have enough evidence to be convinced of the player’s innocence so I’m not sure why any more hours of analytic work are required. I would certainly be interested to know what about Regan’s algorithm it is that they disagree with other than not liking the result that is output.

There seems to be a (somewhat understandable) inclination by team captains to believe their players’ innocence against the evidence of the games, lichess’s algorithm and Regan’s. The player may well be innocent (I indeed hope he is) but ‘they denied it and we believe them’ is clearly not sufficient to be so sure of a players innocent to withdraw all teams in protest. The number of parents who think their kids misbehave in school and the number that actually do vary wildly: our inclination to want to believe someone who we know well’s innocence often clouds our judgement.

[Roger de Coverly]

There seems to be a (somewhat understandable) inclination by team captains to believe their players’ innocence against the evidence of the games, lichess’s algorithm and Regan’s.

The main problem is that no evidence of perceived guilt is presented.

If the allegation was that someone on the same IP address viewed the game and looked at the assessments, that's a concrete charge. It might not have been the player, rather a parent or sibling. It's also partly the fault of the server in supplying real time analysis of games in progress.

[Matthew Turner]

Roger,
You only get a ban if you play unnaturally well, better than anyone played in the whole of the OTB 4NCL. So the idea that a player could be banned because a parent accessed some analysis from the same IP address is total nonsense.

[JustinHorton]

You only get a ban if you play unnaturally well, better than anyone played in the whole of the OTB 4NCL.

What's your reason for saying so?

There seems to be a (somewhat understandable) inclination by team captains to believe their players’ innocence against the evidence of the games, lichess’s algorithm and Regan’s

What's "the evidence of the games" in this instance?

(Surbiton statement)The onus is on the individual to provide evidence that they have not used external assistance

How would you practically do this?

[Paul Cooksey]

I assume this is the same issue Keith Arkell has been talking about on facebook, with the possibility of his team withdrawing. They lost to a match in which a player subsequently banned by lichess played.

The situation we have is that players cheat, players get very upset when they think someone has cheated against them, and we cannot be 100% sure if anyone has cheated.

I'm a bit doubtful cameras are the answer. Some things just don't work, and I think serious chess online is one of them.

[Richard Bates]

FWIW I get the impression that it’s not that difficult for a decent player to have high correlation of moves with the lichess algorithm on occasion, particularly in certain types of positions which are positional and relatively simple in nature. I am not surprised that a couple of the very recently banned players in Div1 have been playing chess of that nature. It’s not even as if it requires you to play the computers moves, just ones that aren’t worse than the computers best moves.

I would also question whether the lichess retrospective analysis engine is actually that strong/accurate, particularly when analysing games at slower time limits.

But then we still (I think, happy to be corrected) have no idea whether the lichess method will ban people on isolated good games, or takes performances over a range of games into account. Or what in fact it does do. Basing it on a small sample is obviously flawed - then again a large sample could conclude that the variance between the good and the bad is too great. For example, am I at risk because I lost several games in a row a couple of days ago (including some ridiculous losses/blunders) and followed it up a few hours later by winning 11 in a row at a much higher standard? Probably a difference of several hundred points in observed strength across the two sessions!

I had plenty of games in the OTB world where my opponent’s play has been near flawless - I don’t get phased by that so not sure why it should be that surprising when the same happens online.

[Richard Bates]

Roger,
You only get a ban if you play unnaturally well, better than anyone played in the whole of the OTB 4NCL. So the idea that a player could be banned because a parent accessed some analysis from the same IP address is total nonsense.

I think that’s just demonstrably false.

[Matthew Turner]

You only get a ban if you play unnaturally well, better than anyone played in the whole of the OTB 4NCL.

What's your reason for saying so?

This is what the 4NCL captains were told at their meeting with Ken Regan and seems to be borne out with the evidence that I have seen.

[Matthew Turner]

Roger,
You only get a ban if you play unnaturally well, better than anyone played in the whole of the OTB 4NCL. So the idea that a player could be banned because a parent accessed some analysis from the same IP address is total nonsense.

I think that’s just demonstrably false.

This is what the captain were told at their meeting with a Ken a Regan and does indeed seem to be borne out by the evidence I have seen.

[Matthew Turner]

FWIW I get the impression that it’s not that difficult for a decent player to have high correlation of moves with the lichess algorithm on occasion, particularly in certain types of positions which are positional and relatively simple in nature. I am not surprised that a couple of the very recently banned players in Div1 have been playing chess of that nature. It’s not even as if it requires you to play the computers moves, just ones that aren’t worse than the computers best moves.

Exactly the same players have been unable to do this in the 4NCL, so why is there a difference in the online version?
(The superstars in the Isle of Man or Candidates were unable to do it either)