@ Fat Lady
Big mistake by me for mixing up English Tal and David Tebb - obviously I mixed up the two names ... my mistake. Glad the post was removed - it made no sence what so ever then !!! Thanks for the correction !!! (and for the removal of the post)
@ no1marauder
Do you think that 86,9% is a high number ? And if you do at what basis do you draw that conclusion ? Do you have any idea of the program behind the calculations of theese numbers ?
If I was to decide whether or not this number was high I would need alot of additional information - for instance:
- What are the numbers for other +2200 players
- How are the intervals estimates calculated
- How are the p-values achieved
My thoughs are that the program developed by Gatecrasher (unless he is a hardcore probability theorist) only can be considered as a good indication on whether or not a player is using an engine and the 0,01 procentile is just a number being agreed on by the mods.
I don't know why the site admins decided to "scrap the entire Game Mod team" and neither do you !!! They never got a chance to finish their job - so actually we don't know if Cludi would have been found guilty or not - but you seem to have drawn the conclusion !!!
Also remember that Cludi is a 2100+ OTB player - not that big of a difference to a 2200+ player .....
Let me finish by saying that I myself are of course biased. I have talked alot with Cludi during the many hours I have spent on this site and have no reason what so ever to belive that he is a cheater.
Also if we try to look at this a bit more objectively there is really no evidence what so ever that Cludi has cheated - only speculations.
Cludi has decided to finish his clan games and clan league games before leaving - with all this attention he will probaly not use engine support (not saying he has ever done it) - so let's see how well he does in his remaining games - and let's see if there will be a huge difference compared to his games i 2006-2007.
Originally posted by Richardt HansenHave you seen the evidence NOT generated by Cludi?
@ Fat Lady
Big mistake by me for mixing up English Tal and David Tebb - obviously I mixed up the two names ... my mistake. Glad the post was removed - it made no sence what so ever then !!! Thanks for the correction !!! (and for the removal of the post)
@ no1marauder
Do you think that 86,9% is a high number ? And if you do at what basis do ...[text shortened]... ames - and let's see if there will be a huge difference compared to his games i 2006-2007.
P-
Originally posted by Phlabibit
Have you seen the evidence NOT generated by Cludi?
P-
No - and if someone else gave me another number - say 91,7% I would'nt know what to do with that either. Like I wrote - we need additional information if we wan't to make conclusions.
Originally posted by Richardt HansenEven Gatecrasher has said there is not enough evidence to prove cheating yet, the game mods didn't get far enough into the review to make a final decision before Russ blew it up.
Originally posted by Phlabibit
[b]Have you seen the evidence NOT generated by Cludi?
P-
No - and if someone else gave me another number - say 91,7% I would'nt know what to do with that either. Like I wrote - we need additional information if we wan't to make conclusions.[/b]
I also do not have the evidence anymore, having been locked out of the game mod system. It's just important to realize (in my mind anyway) that Cludi offered up a lower number that was still borderline running his own review and the review offered by Tebb and work done by Gate came up higher in the range of engine use.
P-
Originally posted by tomtom232Please explain, otherwise i'm sure you'll agree that you have somehow wandered toward an irrelevant and possibly biased conclusion. Sure the consequences are not the same, i'll give you that, but that does not somehow magically remove or void the strange and "play innocent" behavior of the player in question. Ya' see it's a question of honor, not of strange and unreliable statistics generated via chess engines. And also the question of motivation should be examined here.
No, not valid. There are consequences to leaving a trial while there are none for leaving a chess site.
Originally posted by HindsteinFalse dichotomy; bandwagon fallacy.
Silence by the admin team in the past only showed that there is not enough evidence to ban and so therefore anyone that people suspect should be considered innocent... until proven otherwise. This should have been the case here too. No ban - not yet guilty.
Originally posted by Richardt HansenI've analyzed all my games here, OTB and a few other sites using Fritz + at least 100 other games. That's about a 1000 games using all kinds of players at all kinds of ratings.
@ Fat Lady
Big mistake by me for mixing up English Tal and David Tebb - obviously I mixed up the two names ... my mistake. Glad the post was removed - it made no sence what so ever then !!! Thanks for the correction !!! (and for the removal of the post)
@ no1marauder
Do you think that 86,9% is a high number ? And if you do at what basis do ames - and let's see if there will be a huge difference compared to his games i 2006-2007.
There's no doubt that 86.9% of your moves being one of Fritz's first three choices IS a high number. I'd have to see the actual data to determine whether it was "overwhelming" proof in my eyes. To me, saying a move matches the first three choices as evidence of engine use can be both underinclusive and overinclusive. "Under" because they may be no real difference between the 4th or 5th choice and the 1st choice (perhaps a couple of a hundreths pawn), so that any variance would be stylistic only. "Over" because there are many positions where the first choice is so obviously superior as to make the use of the 2nd or 3rd choice a mistake or even a blunder.
So yeah I have an idea behind the calculations of these numbers.
It would be a bit hypocritical for Cludi to now question the validity of a program that he has used as a tool to decide whether others should be banned. Sauce for the goose ................
The conclusions I have drawn are: 1) That Dave Tebb was justified in making an issue of this after the Site Admins scrapped the Game Mods prior to a decision being made in Cludi's case; 2) That Dave's motivation to do so had nothing whatsoever to do with the 2007 Championship; and 3) That if Cludi really believes in his own innocence, he should await the decision of the Game Mods when they are reconstituted. I have not, repeat not, drawn the conclusion that Cludi is beyond a reasonable doubt an engine cheat.
Originally posted by Richardt HansenWrong, there is evidence. His own secret "statistical" analysis showed that the probability that was cheating was extremely high, his rather inappropriate comparison to Tals games notwithstanding.
There are no evidence what so ever that Cludi has cheated - if there was he would have been banned.
Originally posted by eldragonflyPersonally I think a comparison to a contemporary Correspondence GM would be of interest.
Wrong, there is evidence. His own secret "statistical" analysis showed that the probability that was cheating was extremely high, his rather inappropriate comparison to Tals games notwithstanding.
Originally posted by Fat LadyThe analogy of the child molester is a good one as it does provoke thought. When the safety of children is in jeopardy, swift actions have to be taken to prevent the unthinkable from happening. However, for the sake of argument, consider the repercussions of a false positive situation....
OK, rather than a murderer analogy, how about a child molester instead? Murderers, for the most part, only kill once, whereas kiddy fiddlers will keep going until they're stopped.
What if you had evidence that a teacher or someone else in authority was molesting the children in their charge. Perhaps you were only 90% certain of your facts (I'm not sure h ...[text shortened]... let that pass!). Would you allow them to carry on with their job until you were 99% certain?
There have been many cases of false accusations against teachers. The scenario goes like this: Child falsely accuses teacher, teacher gets immediately suspended, 12 months later the trial begins, 1 month later teacher is found to be not guilty and child admits to making the whole story up, child goes back to school and continues life as normal, teachers life is ruined and never works in education again.
It is a similar story to a false positive here, but perhaps the main difference is that in a child molestation case these swift actions are taken as a direct result of the severity of the consequences of what would happen if the accused did indeed turn out to be guilty - the safely of children is paramount. I see no such danger here, it is only a chess website. Comparisons to child molesting and murder are perhaps a little over the top.
One thing is certain, in the same way as when Alistair Stewart, presenter of Police Camera Action police show on ITV was found guilty of drink/driving, he was dismissed from his job, any game moderator under investigation should not be allowed to continue the job of judging others.
However, I stand by my original beliefs, that any player should be considered innocent until proven guilty. If 0.1% is the threshold and it is not proven to be within these levels, then so be it - verdict should only be guilty when it is proven. Any other situation, whether there is some evidence or not (and not to others that evidence of computer-like moves is not the same as proof of cheating) should always be treated as innocent.