Cheating in chess

Discuss anything you like about chess related matters in this forum.
Roger de Coverly
Posts: 19343
Joined: Tue Apr 15, 2008 2:51 pm

Re: Cheating in chess

Post by Roger de Coverly » Sun Oct 25, 2020 11:10 am

Matthew Turner wrote:
Sun Oct 25, 2020 10:42 am
Isn't this relatively simple? If this has an observable impact on a player's play then it is cheating, if it doesn't, it isn't.
What about the mathematics of coincidence? A player applies skill and judgement to select moves. An engine applies the rules and methods that have been programmed or self learnt. The two may match in their move choice. That may be a coincidence unless it can be shown to be "too often". Equally it may not be a coincidence, but not cheating either, given the legality of consulting with engines and books written with engine assistance before a game starts.

Empirically lichess appears to have a short fuse in terms of implementing bans.

A bottom line is that whilst over the board chess can be regarded as an "engine and consultation secure" environment for the most part, online chess cannot without major surveillance measures.

Wadih Khoury
Posts: 278
Joined: Sun Jul 12, 2020 8:14 pm

Re: Lichess criteria?

Post by Wadih Khoury » Sun Oct 25, 2020 11:24 am

Joseph Conlon wrote:
Sun Oct 25, 2020 10:57 am
Wadih Khoury wrote:
Sun Oct 25, 2020 10:06 am
Roger de Coverly wrote:
Sun Oct 25, 2020 9:53 am


Is the adult able to watch the game in real time with no server delay? Does the spectator just see the game or assessments and analysis as well?
You can try it yourself and spectate any game on lichess. There will be no analysis nor position assessment.

However, should you have the bad idea of launching an analysis of this position in lichess itself, and happen to share the same IP with the player, the player will be very likely flagged.

Most people focus a lot on computer assisted cheating, as it is the most obvious, but I am now of the belief that the most insidious and hard to catch cheating is when another player is helping.
I would imagine that it is essentially impossible to detect any case where one player 20 or so grading points higher is assisting another; the play would be better, but within the reasonable realm.
Even a helper 50 ECF grade higher would be difficult to prove.
Especially when dealing with juniors that can increase at 30 points / 6 months.

User avatar
JustinHorton
Posts: 8393
Joined: Mon Aug 04, 2008 10:06 am
Location: Somewhere you're not

Re: Cheating in chess

Post by JustinHorton » Sun Oct 25, 2020 11:27 am

Matthew Turner wrote:
Sun Oct 25, 2020 9:50 am
Most of us will know someone who has been banned by a platform. The bottom line is that in the vast majority of these cases it is because they have been cheating. It is difficult to view things objectively unless you get over this.
It is really unclear to me what is meant by "the bottom line" here.
Matthew Turner wrote:
Sun Oct 25, 2020 9:50 am
4....... If they are just starting out, then this is completely normal and nothing is going to happen.
I see no convincing reason to be confident of this.
Matthew Turner wrote:
Sun Oct 25, 2020 10:42 am
Isn't this relatively simple? If this has an observable impact on a player's play then it is cheating, if it doesn't, it isn't.
What constitutes "an observable impact" does not always appear to be "simple".
"Do you play chess?"
"Yes, but I prefer a game with a better chance of cheating."

lostontime.blogspot.com

Neville Twitchell
Posts: 72
Joined: Mon Feb 22, 2010 10:22 am
Location: Harlow, Essex
Contact:

Re: Cheating in chess

Post by Neville Twitchell » Sun Oct 25, 2020 11:40 am

Regarding the Lichess website, though it is not strictly relevant to the question of cheating, there are some slightly odd features.
One is its extreme interpretation of what does or does not constitute a drawn position, so that one can lose on time even where the opponent does not have mating material simply because one has, say, a lone pawn. Thus one can lose on time with King and pawn versus King and knight on the grounds that it is theoretically possible to set up a position where one is mated with, say, King on a8 and pawn on a7 and opposing King on c8 and knight on c7 (White's last move presumably being a7?? rather than Ka7).
Another feature that often irritates is the tendency to be told that "letting time run out rather than resigning" can lead to a temporary suspension. I have been given this reprimand several times, none of them occasions where I had deliberately let time run out, but simply had very little time left in a difficult position. I have never actually been suspended, so I don't' know how often one has to trangress before a suspension is actually imposed (if it ever is) and whether anyone has been suspended on that basis, but I would be interested to know the algorithm that determines whether this is imposed.

Roger de Coverly
Posts: 19343
Joined: Tue Apr 15, 2008 2:51 pm

Re: Cheating in chess

Post by Roger de Coverly » Sun Oct 25, 2020 11:48 am

Neville Twitchell wrote:
Sun Oct 25, 2020 11:40 am
Thus one can lose on time with King and pawn versus King and knight on the grounds that it is theoretically possible to set up a position where one is mated with, say, King on a8 and pawn on a7 and opposing King on c8 and knight on c7 (White's last move presumably being a7?? rather than Ka7).
It's implied by the FIDE Laws of Chess that you can lose on time with King and Knight against King and Knight. All that has to be possible is the construction of a help mate from the material on the board. Not all servers may support this interpretation, chess.com might be one as there were some special rules when it hosted the Online Olympiad recently.

The King and Knight position has come up within recent memory in an important OTB women's event during an Armageddon game.
Last edited by Roger de Coverly on Sun Oct 25, 2020 12:00 pm, edited 1 time in total.

User avatar
JustinHorton
Posts: 8393
Joined: Mon Aug 04, 2008 10:06 am
Location: Somewhere you're not

Re: Cheating in chess

Post by JustinHorton » Sun Oct 25, 2020 11:52 am

Neville Twitchell wrote:
Sun Oct 25, 2020 11:40 am
Another feature that often irritates is the tendency to be told that "letting time run out rather than resigning" can lead to a temporary suspension. I have been given this reprimand several times, none of them occasions where I had deliberately let time run out, but simply had very little time left in a difficult position. I have never actually been suspended, so I don't' know how often one has to trangress before a suspension is actually imposed (if it ever is) and whether anyone has been suspended on that basis, but I would be interested to know the algorithm that determines whether this is imposed.
Dunno, but every time I've won a game like that it's been a question of several minutes. Presumably sometimes it's quite accidental (the phone rang, somebody came to the door etc etc) but it's an annoying problem.

I have no idea how they calculate their "games completed" figure, while we're on the subject.

I've served a number of short suspensions for refusing opponents who have wanted to join games I've set up. Well I say short, the last couple have been of nearly half an hour's duration so approaching the point where it could be problematic.

Getting information out of lichess does seem to be prodigiously difficult.
Last edited by JustinHorton on Sun Oct 25, 2020 11:56 am, edited 2 times in total.
"Do you play chess?"
"Yes, but I prefer a game with a better chance of cheating."

lostontime.blogspot.com

John Hodgson
Posts: 62
Joined: Thu Jul 29, 2010 10:13 am

Re: Cheating in chess

Post by John Hodgson » Sun Oct 25, 2020 11:54 am

Matthew Turner wrote:
Sun Oct 25, 2020 9:50 am

4. Why have they got a low rating? If they have been deliberately throwing games or allowing someone else to use their account then this is against the rules - so they should be banned
I'm sure this topic has been covered before, but the problem with ratings on Lichess for classical chess is that most games are played in teams ordered by ECF OTB grading and all players started at 1500. Many of my games on Lichess have been against players averaging ECF 190-200 but with Lichess ratings of 1400-1900 and I assume many players are in the same situation. In those circumstances it is impossible to gain a lichess rating with a similar value to their ECF or FIDE rating.

I suspect Blitz ratings is much less of a problem because games are not played in such a closed environment.

How would your rating advance playing in a closed environment against fellow GMS all starting with a 1500 rating rather than an average 2500 rating?

John Upham
Posts: 5149
Joined: Wed Apr 04, 2007 10:29 am
Location: Cove, Hampshire, England.
Contact:

Re: Cheating in chess

Post by John Upham » Sun Oct 25, 2020 11:59 am

Roger de Coverly wrote:
Sun Oct 25, 2020 11:48 am

The King and Knight position has come up within recent memory in an important OTB women's event during an Agamemnon game.
Roger,

What is the difference between an Agamemnon game and an Armageddon Game?

I was not aware that the King of Mycenae got involved in such matters.
British Chess News : britishchessnews.com
Twitter: @BritishChess
Facebook: facebook.com/groups/britishchess :D

John Upham
Posts: 5149
Joined: Wed Apr 04, 2007 10:29 am
Location: Cove, Hampshire, England.
Contact:

Re: Cheating in chess

Post by John Upham » Sun Oct 25, 2020 12:04 pm

John Hodgson wrote:
Sun Oct 25, 2020 11:54 am

How would your rating advance playing in a closed environment against fellow GMS all starting with a 1500 rating rather than an average 2500 rating?

This thought had also occurred to me : anyone know the answer?
British Chess News : britishchessnews.com
Twitter: @BritishChess
Facebook: facebook.com/groups/britishchess :D

Matthew Turner
Posts: 3326
Joined: Fri May 16, 2008 11:54 am

Re: Cheating in chess

Post by Matthew Turner » Sun Oct 25, 2020 12:08 pm

John Upham wrote:
Sun Oct 25, 2020 12:04 pm
John Hodgson wrote:
Sun Oct 25, 2020 11:54 am

How would your rating advance playing in a closed environment against fellow GMS all starting with a 1500 rating rather than an average 2500 rating?

This thought had also occurred to me : anyone know the answer?
The ratings would stay consistently at 1500 and the objective performance as measured by move matching would stay at 2500. So no observable change, no problem.

Roger de Coverly
Posts: 19343
Joined: Tue Apr 15, 2008 2:51 pm

Re: Cheating in chess

Post by Roger de Coverly » Sun Oct 25, 2020 12:16 pm

Matthew Turner wrote:
Sun Oct 25, 2020 12:08 pm
The ratings would stay consistently at 1500 and the objective performance as measured by move matching would stay at 2500. So no observable change, no problem.
I thought the purpose of Regan and similar tests was to finger those with 1500 ratings playing at a 2500 standard.

But if a genuine 1500 standard player happened to join the closed pool of 1500 GMs and duly got slaughtered, why would the 1500 player not shout about cheating?

NickFaulks
Posts: 6277
Joined: Sat Jan 02, 2010 1:28 pm

Re: Cheating in chess

Post by NickFaulks » Sun Oct 25, 2020 12:19 pm

JustinHorton wrote:
Sun Oct 25, 2020 11:27 am
Matthew Turner wrote:
Sun Oct 25, 2020 9:50 am
The bottom line is that in the vast majority of these cases it is because they have been cheating. It is difficult to view things objectively unless you get over this.
It is really unclear to me what is meant by "the bottom line" here.
It is in reality "the top line" and all else follows from it. That was certainly the point of view fed to the Guardian journalist.
If you want a picture of the future, imagine a boot stamping on a human face — forever.

Matthew Turner
Posts: 3326
Joined: Fri May 16, 2008 11:54 am

Re: Cheating in chess

Post by Matthew Turner » Sun Oct 25, 2020 12:22 pm

Roger de Coverly wrote:
Sun Oct 25, 2020 12:16 pm
Matthew Turner wrote:
Sun Oct 25, 2020 12:08 pm
The ratings would stay consistently at 1500 and the objective performance as measured by move matching would stay at 2500. So no observable change, no problem.
I thought the purpose of Regan and similar tests was to finger those with 1500 ratings playing at a 2500 standard.

But if a genuine 1500 standard player happened to join the closed pool of 1500 GMs and duly got slaughtered, why would the 1500 player not shout about cheating?
Roger,
There is a difference between the Regan tests and those carried out by the websites. Yes the Regan test would flag up a player rated at 1500 and 'objectively' performing at 2500, but when these tests are used by the ECF/4NCL/Chess Scotland/FIDE then they are using OTB ratings.

Roger de Coverly
Posts: 19343
Joined: Tue Apr 15, 2008 2:51 pm

Re: Cheating in chess

Post by Roger de Coverly » Sun Oct 25, 2020 12:52 pm

Matthew Turner wrote:
Sun Oct 25, 2020 12:22 pm
There is a difference between the Regan tests and those carried out by the websites.
Given the secrecy surrounding the methods and tests carried out by the websites, I doubt we know much one way or another as to whether and where their methods differ from the Regan template.

John Upham
Posts: 5149
Joined: Wed Apr 04, 2007 10:29 am
Location: Cove, Hampshire, England.
Contact:

Re: Cheating in chess

Post by John Upham » Sun Oct 25, 2020 2:23 pm

The longer this veil of secrecy hangs over Lichess idiosyncratic banning's the more it will damage them.

It might be that they can pay for alleged quotes from world class players to say that they are privy to the anti-cheating criteria and what a great platform Lichess is.

Various GMs might be able to claim that they have been admitted to the elite inner circle of those who know and they can be smug that the vast bulk of players do not know.

It is fairly obvious that publishing details of the stupid things that might get you banned would help people have confidence in the platform.

They can be no harm in declaring that (for example) :

Having an old analysis window active on another game will get you banned or other examples of accidental behaviour.

I don't expect them to declare their criteria are for concluding player X is not strong enough to play good moves.
British Chess News : britishchessnews.com
Twitter: @BritishChess
Facebook: facebook.com/groups/britishchess :D

Post Reply