Only Chess
06 Mar 08
12 Mar 08
Originally posted by Richardt HansenBFD on the leaked pm, you can do better than that.
@ Pawn Qween, eldragonfly, no1
Since you all agree that my posting are childish, logic is my enemy and{add what ever you like} lets have a look at the facts then. (Correct me if I'm wrong):
- It has not been proven that Cludi is an engine user (even though someone claims it)
- The game mods were dismissed before a conclusion regarding the Cl ...[text shortened]... s that Cludi is an engine user - nevertheless you are convinced that he is !!! Why is that ?
And it has not been proven that cludi is NOT an engine user. Know it. Understand it. Worship it.
12 Mar 08
2 edits
Originally posted by SydrianCorrect. Game selection is critical if you are using statistical inference from a batch of games. Any selection criteria imposed have to be unbiased with regard to the probability of matching an engine. You can't start with a large group, weed out all those games with low match-ups, and apply a statistical test to the balance. That would totally invalidate the results.
Lets assume for instance the accused played 50 games that were against opponents of sufficient quality. For the statistics of the analyzed games to have any REAL meaning, all 50 games should be looked at. If for instance only 20 of these games were hand picked for being suspicious, statistically speaking, that stacks the deck against the accused. From a poin erry pick "suspicious" games, if in fact that is what is done when a player is accused here.
12 Mar 08
2 edits
Originally posted by Gatecrasher:That alone makes me feel 110% better about the system.
Correct. Game selection is critical if you are using statistical inference from a batch of games. Any selection criteria imposed have to be unbiased with regard to the probability of matching an engine. You can't start with a large group, weed out all those games with a low match-ups, and apply a statistical test to the balance. That would totally invalidate the results.
12 Mar 08
Originally posted by Sydrianthere's a huge fundamental difference between selecting games with statistically less 'noise', compared to cherry picking 'high matchup' games. the latter introduces great bias, the former improves the quality of the sample. the important thing is that the sample the results are measured against, eg. the huge amount of master games, is selected similarly. and in fact, selecting games against far inferior players would introduce more bias because the master games are almost exclusively against roughly equal opponents.
There is one major thing that's driving me nuts with this whole situation. We were told time, and time again that knowing how cheaters were discerned could help potential cheaters avoid detection. For this to be true, only two possibilities are left open to me. Either knowledge of the system was used to get painfully close to the "threshold" without crossing ...[text shortened]... s is preformed, so this post purely speculative, and highly opinionated. : DISCLAMER!!!
12 Mar 08
1 edit
Originally posted by eldragonflyIf you read Sydrian's post and my reply to him, you get the idea of what can happen with data selection if there is a lack of objectivity.
Can you expand on this? including the objective part, that is.
You can make statistics prove anything if you are able to manipulate the data selection.
12 Mar 08
2 edits
Originally posted by GatecrasherOh so true.
If you read Sydrian's post and my reply to him, you get the idea of what can happen with data selection if there is a lack of objectivity.
You can make statistics prove anything if you are able manipulate the data selection.
edit: i read Sydrians post, he is concerned about how sample selection is achieved and the sample size, not about specific data selection/match-up criteria, but i will echo another point here, it is possible that { the former game mod that shall not be mentioned } used a chess engine only in that one game, therefore would then the idea of sampling games become basically a worthless exercise? Or is this where the idea of highly rated gm moves are in a narrow band comes into view.
12 Mar 08
Originally posted by GatecrasherNow that sounds kind of scarey...Guess you wouldn't want any of the Game Mods pissed at you if this is the case :'(
If you read Sydrian's post and my reply to him, you get the idea of what can happen with data selection if there is a lack of objectivity.
You can make statistics prove anything if you are able manipulate the data selection.
12 Mar 08
1 edit
One thing driving me nuts here. If cludi was cheating and that certainly has not been proven then I see a major flaw in the whole system.
As a game mod cludi would know how games were chosen, he would also know the criteria that needed to be looked at to determine if a player was cheating (i.e. he would know what constitutes "engine moves" ). Knowing this he would be able to avoid the situation.
Consequently he should not have been suspected let alone "caught". So either cludi is a fool or he is probably innocent. I don't believe him to be a fool which makes him innocent.
If he is still perceived to be guilty I think in this instance the proof should be published and if it is not then I think a public apology should be forthcoming and this witch hunt should stop. I think this is all giving an impression that this site is totally infested with engines and that nothing is done about it whereas I feel the truth is the opposite - this site has less cheaters than many others and far more is done than most sites,
It is not unusual for false accusations to be made if someone loses even if they are made in good faith. Mistakes are made and until proven otherwise we must assume a mistake has been made here and hope cludi will return as he had so much to offer.
12 Mar 08
Originally posted by Dragon FireI had been wondering that Exact thing myself!!! You put it into words much better than I could have, well said !!!
One thing driving me nuts here. If cludi was cheating and that certainly has not been proven then I see a major flaw in the whole system.
As a game mod cludi would know how games were chosen, he would also know the criteria that needed to be looked at to determine if a player was cheating (i.e. he would know what constitutes "engine moves" ). Knowing ...[text shortened]... assume a mistake has been made here and hope cludi will return as he had so much to offer.
12 Mar 08
Originally posted by GatecrasherI know this only too well, I used to work for my local Borough Council 😛
If you read Sydrian's post and my reply to him, you get the idea of what can happen with data selection if there is a lack of objectivity.
You can make statistics prove anything if you are able to manipulate the data selection.
