Borislav Ivanov is BACK!
Sort:
[COMMENT DELETED]
Oct 17, 2013
0
#502
adamstask wrote:
well argued Professor. Not because of titles, certainly. And yes, professor, I heard of Lilov because of the thread on Ivanov on chess.com, and I heard of Ivanov just from surfing forums to read on chess.com.
And the analogy to the murderer of the killer of Jesse James, just brilliant. I read up on the O'Kelley man on the English wikipedia; there's a lot more on him in the English than in the Dutch, and its worth the read.
Thanks, I added the English version to my post. It was much more interesting indeed. :-)
@Loek: I just changed it from 'Professor' to 'Loek'. Sorry about that :-) and thanks again for that link to O'Kelley. What a life!
Indyfilmguy wrote:
What if the guidelines call for more than "there's an interest in them"? Also, how would you define "interest"? That two people care enough to write about something and nobody else cares enough to propose it for deletion? Please explain.
675 posts on this thread, that's interest.
This paper posted earlier is defintely worth a read.
http://www.cse.buffalo.edu/~regan/chess/fidelity/ACPcover-and-report.pdf
Regan is a real scientist and a statistician but his analysis is also pretty empty (but much better than the Bayes calculation earlier). Pretty much "statistical move matching" is based on the idea that Ivanov's moves ought to match Rybka (odd) at something like the same frequency as other players of his rating or even of higher ratings. They don't. We knew that before any statistical test. So we can throw all the statistical tests in the world at it and come up with pseudo-science that says "He cheated" but it is just based on transforming that initial observation.
The fact is that Regan is doing a statistical test based on a sample size of 1 (Ivanov), not the moves as he sort-of claims. Statistical tests based on sample sizes of 1 are usually not very good.
This stuff is obviously not published in any journal and I would pan it to death on any journal I referee for (which is down to all of 1 right now, but has surely been higher).
Pretty much any statistical argument about Ivanov contains some assumption like "2300 players don't have results like 3000 players" or "the moves of 2300 players cannot reliably be like 3000 players" or "Nobody has a performance rating that varies 1200 pts from weekend to weekend", assumptions that are probably true but believing them means there is no need for statistics.
Incidentally, a more important area than chess cheating is test cheating and I've done some expert witnessing on answer sheet correspondence between test takers. Many test providers will void the results (and possibly give other penalties) if one person's answers closely match the answers of someone else in the room. I think this is crappy proof of cheating because the correspondence could simply be because they have the same source of mistakes, like the studied together or used the same bogus study materials. I've had less success than failure with these arguments, but I think it's mostly because I am arguing for people who actually did cheat.
[COMMENT DELETED]
Indyfilmguy wrote:
adamstask wrote:
Indyfilmguy wrote:
What if the guidelines call for more than "there's an interest in them"? Also, how would you define "interest"? That two people care enough to write about something and nobody else cares enough to propose it for deletion? Please explain.
675 posts on this thread, that's interest.
It's local interest on Chess.com, do you realize there is a broader world out there?
...You people seem to sorely lack any sense of proportion in life, common sense...
Out where? chess.com is part of 'the wider world'. You had scoffed at the notion that 'interest' was that "two people ...care...to write about something...". I pointed to 675 posts.
Some of you people sorely lack the ability to deal with reason, evidence, or even common decency.
Indyfilmguy wrote:
Yeah, so are you saying that 675 posts on a chess discussion forum should be enough to justify an article on a general interest encyclopedia like Wikipedia?
No that's not what I'm saying. You have a penchant for misconstruing. However, like I wrote earlier, you should open a thread on wikipedia, and/or you should go and argue your case on the Afd pages on wikipedia.
Oct 17, 2013
0
#509
Mr. O'Kelly shot the man who shot Jesse James. Don't know what the man who shot the man who shot Abraham Lincoln has to do with it. I did not introduce that man. And no, I don't think one of those two men have more or less importance. Abraham Lincoln had and Jesse James was a well known/notoriously known person, but the murderers of their murderer weren't.
There have been articles in several chess magazines about this issue in which Lilov was mentioned too and whatever one thinks of it, his video messages has brought the allegations of cheating done by Borislav Ivanov more into the footlight.
When you go to youtube, then can you see that his videos have been viewed several thousands of times. KingCrusher mentions the video of Lilov too. The Bulgarian Chess Federation reacted on it. A lot of Bulgarian GMs reacted on this using the videos of Lilov as an argument. In articles about Ivanov on chessbase.com is Lilov mentioned. If Valeri Lilov deserves a page on wikipedia then is that due to his videos about the cheating of Borislav Ivanov.
Not for his own accomplishments as a chess player.
Oct 17, 2013
0
#510
Porn Stars have Wikipedia pages. Sports athletes have Wikipedia pages. I'm a fan of Wikipedia and use it often but an argument over who should and should not have a page seems silly. Wikipedia isn't the Encyclopaedia Britannic. They aren't all that cautious over who has a page up. That's half the fun.... If Porn Stars and tennis players can have a page why not a chess player or two?
Oct 18, 2013
0
#511
In this article is prof. K. Regan (member of the FIDE anti-cheating commitee) mentioned too:
Oct 18, 2013
0
#512
Valeri Lilov, if he wants to put up a web page on wikipedia is fine, there is no harm. Other Fide-masters have web pages. And so do other kind of 'performers'.
What I think Wikipedia takes issue with, and have taken offence with, is that Lilov has been editing his own page, and using it as a kind of 'MEDIA OUTLET". Updating, changing, putting his opinion, etc...
Wiki pages are supposed to be fact based, impartial, and most of the time written by volunteers, whose information gets fact-checked by other volunteers, and then it is LEFT ALONE.
Unless someone suggest a GOOD REASON for changing it first, there is a kind of informal permission you have to take, not just whenever someone FEELS like chainging it, no matter how good the reason. That is what Facebook is for.
Because Lilov may have done that a few times, messing with the page willy-nilly, Wikipedia volunteers have taken issue with that, and like here on the forums, decided to be just done with the whole page, to put a permanent end to it, since his page is not that CRUCIAL.
Oct 18, 2013
0
#513
If anyone is looking for articles which make them feel comfortable, rather than facts which can be verified, then Google Conservapedia. Wikipedias creepy uncle.
Or is it still "And the analogy to the murderer of the killer of Jesse James, just brilliant."
and the analogy to the murderer of the killer of Jesse James, just brilliant. Thank you to Loek for that!
Yekatrinas wrote:
silly this chiaroscuro. Now he is even a referee to some Math statistics journal. Earlier he claimed to have a degree in a science other than Math.
Anyone who believes him is doing it just because he is good in showing off. No scientific clarity in any of his remarks.
I got $10,000 says that your mother will say that my academiic credentials make your look meaningless....
Yekatrinas wrote:
Hehe, no one on this website bets your credentials exist.
The other academics do. Sorry you are a waanabe and a not very bright one at that. But make the bet. You get your Mom to judge and she will say something like "I'm sorry honey, we are proud of you but you are nothing compared to this guy"....
chiaroscuro62 a écrit :
Yekatrinas wrote:
Hehe, no one on this website bets your credentials exist.
The other academics do.
They have been incredibly quiet so far... Where are the academics when you need them ?
here's what chess.com people are saying to IndyFilmguy on another thread where he's on his 'critical thinking' wikipedia attention-seeking "antagonistic 12 year old "bandwagon.
http://www.chess.com/forum/view/general/congrats-to-valerie-lilov-on-his-promotion-to-im-wikipedia
chiaroscuro62 wrote:
This paper posted earlier is defintely worth a read.
http://www.cse.buffalo.edu/~regan/chess/fidelity/ACPcover-and-report.pdf
Regan is a real scientist and a statistician but his analysis is also pretty empty (but much better than the Bayes calculation earlier). Pretty much "statistical move matching" is based on the idea that Ivanov's moves ought to match Rybka (odd) at something like the same frequency as other players of his rating or even of higher ratings. They don't. We knew that before any statistical test. So we can throw all the statistical tests in the world at it and come up with pseudo-science that says "He cheated" but it is just based on transforming that initial observation.
The fact is that Regan is doing a statistical test based on a sample size of 1 (Ivanov), not the moves as he sort-of claims. Statistical tests based on sample sizes of 1 are usually not very good.
Lol. You really believe that the "sample size" here should be the number of players ?
The Houdini matchup percentage of any player who played for some while is around constant ; yes, people can get better (by practice) or worse (medical condition/lack of practice/etc.) but they do not exhibit rating jump. This has been verified on almost all players (Carlsen, Alekhine, Mr. Patzer), including pre-Zadar Ivanov. Yes, it is an assumption, but a well-warranted one.
It is possible to deduce stuff from a sudden change in that percentage. All the arguments that you presented against that fact amount to "no, that percentage changes".