* Game moderators - the cheat police

Originally posted by no1marauder
I made a number of proposals on page 17 of this thread regarding how I see the system functioning (which Nemesio stated he endorsed "100%"😉. I'll repost them here:

1. Investigations of cheating would be undertaken only where there was: A) A very sharp rating rise of an established player;
B) A new player who comes in and wins a certain num ...[text shortened]... engine use as much as possible as other sites do and has 90% of the subscribers here want to do.

I think that #1 makes many solid suggestions here.

Particularly that there has to be a route back into the community for offenders. I'd propose allowing a new clean account with fees to be repaid, but there need to be some approved route.

One correction to #1's post: only 82.5% of votes have been in favour of the games police (9% against), after 155 vote cast.

Originally posted by The Swine Down Hope
I think that #1 makes many solid suggestions here.

Particularly that there has to be a route back into the community for offenders. I'd propose allowing a new clean account with fees to be repaid, but there need to be some a ...[text shortened]... n in favour of the games police (9% against), after 155 vote cast.

I consider a "no opinion" vote as a null; you might has well count everybody who hasn't voted at all. 128 for; 14 against is 90% in favor in most elections I've ever heard of.

There's one part of my proposal I might change. I said this:

I would support an amnesty only if the player who had cheated confessed to the site admins and was placed on some sort of probation to be monitored by the cheat police, although his name need not be made public.


I would still be in favor of an amnesty program under most of the conditions mentioned, but in reconsidering the matter I would support public disclosure. I personally am trying to judge my abilities against other humans; I'd like to know if I the player who beat me in the past was an engine. And I also think that other players should have the choice to not play a past admitted engine user if they so choose. So far, Russ seems to be supporting a "death penalty" of account termination however, so this point might be moot.

Originally posted by no1marauder
I consider a "no opinion" vote as a null; you might has well count everybody who hasn't voted at all. 128 for; 14 against is 90% in favor in most elections I've ever heard of.

It's almost 91% if you count it as 'people who are not * not against the CP'.

P-

Originally posted by no1marauder
I consider a "no opinion" vote as a null; you might has well count everybody who hasn't voted at all. 128 for; 14 against is 90% in favor in most elections I've ever heard of.

There's one part of my proposal I might change. I said this:

I would support an amnesty only if the player who had cheated confessed to the site ...[text shortened]... e supporting a "death penalty" of account termination however, so this point might be moot.

Well, in either case the feeling is pretty much with the introduction.

But no I don't think you can ignore the 'no opinion'. Particularly as players, who see pros and cons of the system, would cast their vote the using the no opinion option as a proxy for unsure. If the voting was 9 for, 1 against and 70,000 no opinion/ unsure, would you still quote that as 90% of votes have been in flavour?

BTW I'm strongly in flavour of the introduction, but feel its important to follow the kind of process you have outlined at the top

Originally posted by derek9037
Your statement is unacceptable.

If just ONE innocent player is branded a cheater, then the system will have failed.

FWIW I cannot wait for something to be done, but I also wish to see only genuine cheaters kicked.

Then you must be opposed to every legal system ever
invented because, indubitably, innocent people are, in
fact, punished by the law for crimes they didn't commit.

It is a sacrifice that a small part of the community
(honest, high-level players) may have to endure for
the good of the entire site.

You live a life of unacceptable consequences all the time.
The people in power do the best they can (we hope) to
minimize the error.

I see no reason why RHP should be any different, especially
since the cheating factor seems to be rather substantial given
what little we've seen already.

Nemesio

Originally posted by derek9037
It's just I cannot go along with that statement (to accept that there will be innocent casualties). I hope it doesn't happen and I accept it looks as if all steps will be taken to see it doesn't.

Well, while I hope it doesn't happen, I expect that it will
very rarely happen. No system of judging that involves
humans will be purely objective and, as such, innocents will
be lost.

You've got to do a cost-benefit analysis. If 1999 true cheaters
are thrown out along with 1 honest guy, is it worth it? I say
absolutely because the harm that 1999 cheaters do is greater
than the good that one honest guy does.

Where does the equation equal out? I don't know, but it is
my expectation that the cheat police will, under Russ' guidance,
have a high enough 'batting average' such that any accidental
innocents being lost will be kept to an utter minimum.

Nemesio

While we have to accept the risk, I think we should be able to avoid all false positives. Particularly as - if it's that important to the player in question - they should be allowed to play OTB with a trusted member (to prove their strength).

Though its very unlike to go this far in reality. As I said previously:

'Most of the benefit from any policing activity is its deterrent effect: after catching blatant cheaters the most we can do is make is so difficult, to cheat, that it becomes easier for them to play it straight.'

I offer my collaboration to the police cheater.
In the 1994 I wrote 3 chess books !!
I can test the cheater with 3 minutes games in yahoo.com chess site
It's free !! 16 GAMES AT LEAST
I defeat a lot of master and some International Master in quick games
5 minutes or 15 games so I can understand if a chess player is strong or no
I won chess tournament with international rated player so I suggest:
3 MINUTES AND NO MORE because it's impossible play with the computer .
In playchess.com my best rating (3 minutes games is 2397 ELO) but my level is 2150/2250 so I'm no Kasparov but if a chess player win against me is a good honest player.
I hope that other good player follow my idea, it may be no the best but I believe it's interesting.
( Same nick name in Playchess.com best rating 2233 this year )
Bye El Bruto

Originally posted by no1marauder
I made a number of proposals on page 17 of this thread regarding how I see the system functioning (which Nemesio stated he endorsed "100%"😉. I'll repost them here:

1. Investigations of cheating would be undertaken only wher ...[text shortened]... as other sites do and has 90% of the subscribers here want to do.

All right,my thoughts on that:
1.A) ok,good point
B)I agree with that,but it might mean a big extra workload
C)if n can be 1,then yes,but how can you judge if the complaint is justified,without examining the game(s)?Hence,any complaint must be examined,at least briefly
2.Inform of being investigated,and give opportunity to admit or deny,yes,very good.
Allowed to study the evidence used?No.You'll be teaching them how you catch 'em,and they'll learn,improve their ways,come back and cheat better.
3.Ah!The most important point.'Clear and convincing evidence'.Very good,except....what is that??Who here has the time and the knowledge,to investigate all the games?And I think it will be many,many games.
4.Sounds good

What to do with a caught individual is up to russ,it's his site.

I mostly agree with you,but you don't adress the most important problem.It lies with the evidence.It's easy when someone's games all match 100% with an engine.The problem is,what if it's only 90%,is that enough too?And surely matching moves is not all that must be looked at.I don't think many volunteers know what they're getting into,nor do they have the time or the knowledge for proper investigation.
But that's just what I think,to each his opinion.

Originally posted by El Bruto
I offer my collaboration to the police cheater.
In the 1994 I wrote 3 chess books !!
I can test the cheater with 3 minutes games in yahoo.com chess site
It's free !! 16 GAMES AT LEAST
I defeat a lot of master and some International Ma ...[text shortened]... ame in Playchess.com best rating 2233 this year )
Bye El Bruto

Cant cheat in under 3 min games? Please......... 90% of people over 2000 on yahoo cheat. They have plug in programs and let the engines play 1 minute games. You can find them all over the net.

There are people on Uchess using 1 min engine games.

Frankly all the other information is unintresting, if you have an OTB rating feel free to post it.

i definitely agree with grayeyes here .... to my understanding short time limit games is the computer chess engine's strongest game.

the longer the time limit the greater the chance the human has to win.

it has to do with the method of analysis, humans gain more by taking more time and planning, computers just get bogged down in the huge number of possibilities.

i think computers can be really strong at correspondence also - especially with a little human guidance from time to time.

having the suspected cheater play extremely fast could have its use though ... it could force the cheater to use one engine exclusively, and make comparison to an engine's choice of move simpler.

http://www.leblitzer.com/sutra68.html

Yahoo engines. They have their own interface and let the engine play the games, set the engine to a 1 min game and then all they have to do is let it go.

Seems theres programs for Pogo there too.

Although I cant find the programs on the site, I have a copy of CMT, if anyone is intrested in seeing how it works.

DSPC would have had the same type of program here.

Originally posted by flexmore
i definitely agree with grayeyes here .... to my understanding short time limit games is the computer chess engine's strongest game.

the longer the time limit the greater the chance the human has to win.

it has to do with the method of analysis, humans gain more by taking more time and planning, computers just get bogged down in the huge number of po ...[text shortened]... eater to use one engine exclusively, and make comparison to an engine's choice of move simpler.

They can swap engines, the interfaces are built to use the normal standard engines. I just took a look at which engine was in the CMT zip file and it was Crafty, but on the board they talk about Fritz and shredder too.

Originally posted by flexmore
i definitely agree with grayeyes here .... to my understanding short time limit games is the computer chess engine's strongest game.

the longer the time limit the greater the chance the human has to win.

it has to do with the method ...[text shortened]... ively, and make comparison to an engine's choice of move simpler.

All very interesting, but... If a player is accused of cheating, why on earth would he/she "prove" their guilt by using an engine in a trial blitz game? Surely they should just play the blitz game humanly and badly. It would at least prove that they did not use an engine in that instance.

Playing a blitz game tells us nothing about how good or bad someone is at playing correspondence chess, it only tells us how good or bad they are at blitz chess.

I've only played a handful of blitz chess games, and I spent the entire time in a state of white panic. I sucked badly, missing all sorts of threats and opportunites which, given enough time, and a more thorough analysis I would have seen quite easily.

The whole trick of improving my correspondence game has been to stop myself from playing impulsive moves, to consider the positional elements, to formulate a plan, to thoroughly calculate variations. It would be lovely to do all that in a split second, but for us non-savant human mortals, it's just not possible.

A blitz litmus test would tell us nothing, unless the suspect is stupid enough to use an engine to play it.

Perhaps more useful and illuminating would be a "walk through" by an accused person giving the thinking behind each of their moves in a suspected RHP game. Computers and Humans move for very different reasons. Humans are unlikely to choose a move because of some obscure tactical advantage 10 moves deep. A computer will. Computers are not too concerned about symmetry and harmony. Humans generally are. To have a human explain why they made a "computer" move could be quite revealing.

But even then, there is no single magical indicator of cheating. It is a complex issue that cannot be simplified or reduced to a single factor. You have to examine the entire body of evidence, and arrive at a fair conclusion based on the evidence. It's the only way to go.

Originally posted by Gatecrasher
All very interesting, but... If a player is accused of cheating, why on earth would he/she "prove" their guilt by using an engine in a trial blitz game? Surely they should just play the blitz game humanly and badly. It would at least prove that they did not use an engine in that instance.

Playing a blitz game tells us nothing about how good or bad ...[text shortened]... of evidence, and arrive at a fair conclusion based on the evidence. It's the only way to go.

gets my recomendation!

out of interest ...i am different from you .... when i play fast i rely almost completely on positional assets and lose my tactics horribly.
when i slow down the thing i try to bring into my game is tactics.