< Earlier Kibitzing · PAGE 19 OF 19 ·
|Sep-25-14|| ||Everett: I praise Gelfand for the same reason; a professional who time and again shows how good he is when it counts. Not always of course, but more often then would be expected. He understands that all tournaments and matches are not created equal, and prepares accordingly.|
|Sep-25-14|| ||Absentee: <Everett: Truth is, Petrosian won the most important event during that time, namely, the 1966 WC, despite ELO nonsense and predictions.>|
I bet. Elo ratings weren't introduced until 1971.
Now maybe you can take some time off and explain me why it counted the most (no, higher stakes aren't a good argument).
|Sep-25-14|| ||Everett: ELO/chessmetrics, pedant. You bet.
Of course higher stakes.
I suggest you never become a coach of any team which season ends with a qualifying tournament.
|Sep-25-14|| ||Absentee: If I'd really wanted to be pedant, I'd have pointed out that Elo is a name and not an acronym. :-)|
Would you also care to explain why raising the stakes makes an event more meaningful? And why hitting a peak during the WC qualifiers is more commendable than hitting a peak during another run of the mill top 10 tournament?
|Sep-25-14|| ||alexmagnus: <despite ELO nonsense and predictions>|
Nonsense? First learn what Elo tells and what it doesn't. Elo predictions are <probabilities>.
The difference is significant. Not only do they use different formula, chessmetrics was also computed "after the fact" (and the formula was "adjusted" to get the "correct" predictions).
|Sep-25-14|| ||alexmagnus: And history of WC matches in Elo time speaks for Elo, by the way. Nobody ever won as the lower rated player when the difference was 80+ points. Twice a match was won with 60+ points differece, on both occasions by Kramnik - but Kramnik's rating against Topalov was still "damaged" by Kramnik's illness 2004-2005. So there is only <one> counterexample. Kramnik-Kasparov, where Elo gave Kramnik a 4% chance. Yes, once in a very big while a player with 4% chance wins. Big deal?|
|Sep-25-14|| ||alexmagnus: Here all the calssical WC matches in Elo time:
Fischer-Spassky - Fischer won as <125 point favourite>
Karpov-Korchnoi I - Karpov won as <60 point favourite>
Karpov-Korchnoi II - Karpov won as <5 point favourite>
Karpov-Kasparov I: no declared winner. Rating favourite: Kasparov by <10 points>
Karpov-Kasparov II: Kasparov won as <20 point outsider>
Karpov-Kasparov III: Kasparov won as <35 point favourite>
Karpov-Kasparov IV: draw (Kasparov retains the title as <40 point favourite>
Karpov-Kasparov V: Kasparov won as <70 point favourite>
Kasparov-Short: Kasparov won as <150 point favourite>
Kasparov-Anand: Kasparov won as <70 point favourite>
Kramnik-Kasparov: Kramnik won as <77 point outsider>
Kramnik-Leko: Kramnik won as <29 point favourite>
Kramnik-Topalov: Kramnik won as <70 point outsider>
Anand in Mexico was rating favourite.
Anand-Kramnik: Anand won as <9 point favourite>
Anand-Topalov: Anand won as <18 point outsider>
Anand-Gelfand: Anand won as <64 point favourite>
Anand-Carlsen I: Carlsen won as <95 point favourite>.
So, of 16 <matches> with a <declared result>, only four were won by rating outsiders. In two of them the rating difference was insignificant (18 and 20 points respectively), and Kramnik's rating against Topalov was "damaged"...
|Sep-25-14|| ||alexmagnus: <calssical> classical*|
|Sep-25-14|| ||diceman: <EdZelli:
Do you know the meaning of lackluster?>
Petrosian in the 1971 candidates cycle,
he only had 1 win in each match.
|Sep-25-14|| ||Everett: <alexmagnus> I'm not saying ratings don't have their place. Apologies if my above posts do not make that clear.|
And I only mention chessmetrics because there is no Elo rating before 1971. I know there is a difference.
What I find interesting is that since 2000, the rating seems to matter less, 4-3 in favor of the favorite.
Another question to ask is whether the top player not reigning as WC makes it to the final. Short and Gelfand certainly not. Leko? Topalov in 2010?
In any case, points remain: ratings have their place, circumstances and stakes matter, the WC cycle is not one event. Thanks everyone.
|Sep-25-14|| ||plang: <What I find interesting is that since 2000, the rating seems to matter less, 4-3 in favor of the favorite.>|
One explanation for this is a trend towards shorter matches where there is more of a chance for an upset
|Sep-25-14|| ||mkrk17: Not sure why so much research here on ratings. Noone is denying the importance of ratings. Everyone agrees that Fab's ratings is absolutely fabulous. Only thing we are saying is that it doesn't automatically qualify him for a shot at the WC. If he is as good as his ratings, then let him succeed in the candidates. Its obvious that you won't get a candidates tournament the moment your ratings shoot up. Let him wait, play, win. Aronian has been ratings no.2 for long time. He didn't make it thru any candidates so far. Am not saying that Fab will not make it. I too want to see Fab taking a shot at the WC. Am just saying that he needs to prove his worth in the candidates.|
|Sep-25-14|| ||Everett: <mkrk17> Well said!|
<plang> good point about the shorter matches.
|Sep-25-14|| ||ljfyffe: You get your fingers burnt.|
|Sep-25-14|| ||nok: That every "classical" game, including 3-move draws, is weighed 1 while every rapid game is weighed 0 in standard Elo is of course a somewhat arbitrary simplifying assumption. Not that it'll stop our little accountants here.|
|Sep-25-14|| ||Everett: <5-14 nok: That every "classical" game, including 3-move draws, is weighed 1 while every rapid game is weighed 0 in standard Elo is of course a somewhat arbitrary simplifying assumption. Not that it'll stop our little accountants here.>|
Yes, the accountants. Tabulating nervous tension and brow sweat into neat little boxes filled with numbers.
|Sep-26-14|| ||Absentee: <nok: That every "classical" game, including 3-move draws, is weighed 1 while every rapid game is weighed 0 in standard Elo is of course a somewhat arbitrary simplifying assumption.>|
It's not an assumption, it's a fact. Games are categorized according to playing conditions, not how they're played. That would be somewhat problematic: how short is a short draw? 4 moves? 9? As long as it's still theory? What if a player spends 10 minutes and the other one an hour? Etc.
|Sep-27-14|| ||Arihant: After Nov 2013 WCC, what was the underlyig philosophy that FIDE had to organize it again in Nov 2014....and so we expect another WCC finals in nNov 2015?|
|Sep-27-14|| ||schweigzwang: FIDE? Philosophy?|
|Sep-27-14|| ||Absentee: <Arihant: After Nov 2013 WCC, what was the underlyig philosophy that FIDE had to organize it again in Nov 2014....and so we expect another WCC finals in nNov 2015?>|
It's supposed to be every two years, with the next match in 2016, unless FIDE changes the rules again.
|Sep-27-14|| ||mkrk17: <Arihant: After Nov 2013 WCC, what was the underlyig philosophy that FIDE had to organize it again in Nov 2014.> The goof up was between 2012 and 2013. Poor Anand had to play 2 WC's in 2 consecutive years. Also, in 6 years from 2007 - 2013, Anand had to play 5 WC's. Isn't that ridiculously unfair ?|
|Sep-27-14|| ||Rolfo: When comparing, maybe the matchcount ( number of defences) could be one factor in the equation. FIDE's regulations is getting more predictable now hopefully|
|Sep-27-14|| ||Everett: After this I think it should settle down to every two years.|
|Oct-04-14|| ||ljfyffe: Quick draw? ---- "l'll do the thin'in around here, and don't you forget it!"|
|Oct-14-14|| ||Arihant: I wonder when shall page on WCC 2014 shall open for discussion at CG!|
< Earlier Kibitzing · PAGE 19 OF 19 ·