|
< Earlier Kibitzing · PAGE 1181 OF 1181 ·
Later Kibitzing > |
Sep-12-08
 |
| Rolfo: I don't think frogbert or anyone else get upset with this slo arguing because slo is proven wrong for so many times. I just wonder, since slo doesn't want to be named a troll, why can't he make things even by admitting when he is wrong or accepting some hard facts when getting them on a teaspoon? Only explanation, he wants to be a troll |
 |
Sep-12-08
 |
| slomarko: another ad hominem attack on me by <Rolfo>. what do you mean that i was "proven wrong for so many times?". the only troll here is you. |
|
Sep-12-08
|
| slomarko: <zarg: 19...dxc4
Carlsen giving weak pawns odds, for what? > don't you see that white was planning c5? |
|
Sep-12-08
|
| Rolfo: slo, since you never admit anything anytime, I will not argu with you. Troll is a wellknown term used in these forums, I therefore used it. If it makes any difference I may very well use trollish behaviour instead of naming you a troll. I didn't said you allways show this behaviour, but it certainly appears from time to time |
|
Sep-12-08
 |
| frogbert: now we <only> need
* carlsen to beat anand!
* topalov not to beat ivanchuk
* aronian not to beat radjabov
and carlsen will play blitz tie-break for the tournament victory tomorrow. if those 3 things come true (not very likely, i admit), then only two people might share 1st place (and enter the blitz tie-break), and one of them will be carlsen. carlsen to be official number one in the october list? not likely, but it happens if both topalov and carlsen win tomorrow. if topalov wins and carlsen doesn't, then topalov is number one, and if topalov doesn't win (against ivanchuk), then ivanchuk is the official number one on october 1st. now, i wonder - if carlsen would win (something that would be great for him no matter what happens in the other games) - would carlsen prefer a) topalov to win, to become number one in the rating list? or b) ivanchuk to at least draw, in order to get to play for the 150 000 euro? he can't get both... :o) |
|
Sep-12-08
|
| slomarko: <Rolfo: slo, since you never admit anything anytime, I will not argu with you. Troll is a wellknown term used in these forums, I therefore used it. If it makes any difference I may very well use trollish behaviour instead of naming you a troll. I didn't said you allways show this behaviour, but it certainly appears from time to time> <Rolfo> if you have any problem with anything i wrote please report me to the admins but stop with this ad hominem attack because it generates off-topicness. |
|
Sep-12-08
|
| slomarko: <now, i wonder - if carlsen would win (something that would be great for him no matter what happens in the other games) - would carlsen prefer a) topalov to win, to become number one in the rating list? or b) ivanchuk to at least draw, in order to get to play for the 150 000 euro? he can't get both... :o)>
he should prefer the cash of course. |
|
Sep-12-08
 |
| rogge: Yeah, I think he'd prefer to win the tournament. Slim chance :) |
|
Sep-12-08
|
| Rolfo: < now we <only> need * carlsen to beat anand!
* topalov not to beat ivanchuk
* aronian not to beat radjabov >
If the probabilities for each of the above is let say 0.33, 0.66, 0.66 it should be by multiplication 14% chance for Carlsen to win the tourney :) |
|
Sep-12-08
|
| frogbert: rolfo, i think even that is optimistic, from carlsen's point of view. :o) i'd guess there is a 5-10% chance for carlsen winning the event at this point. |
|
Sep-12-08
 |
| alexmagnus: Unfortunately the Elo system may predict only <scoring> probabilities, telling nothing about how much of this scoring is <win>. So it is impossible to calculate... |
|
Sep-12-08
 |
| zoat22: Disappointing that Carlsen even couldn't beat Radjabov... |
|
Sep-12-08
|
| frogbert: so far, it's 2-2 in mini-matches:
the fighting carlsen
* beat aronian 2-0.
* lost 0-2 against topalov
* beat radjabov 1,5-0,5
* lost 0,5-1,5 against ivanchuk
tomorrow the outcome of the mini-match against anand will be known. until now, carlsen has no drawn minimatch, and it'll be interesting to see what the current 0,5-0,5 standing between him and anand turns out as in the end. my gut feeling is that carlsen will play too hard for a win and lose, or simply don't get much from the opening, still play for a win, and it'll have the same effect. i hope my stomach is wrong. ;o) |
|
| Sep-12-08 |
| barbababa: <alexmagnus> <Unfortunately the Elo system may predict only <scoring> probabilities, telling nothing about how much of this scoring is <win>. So it is impossible to calculate...> On the contrary, Elo rating predicts (statistically) very accurately what are the probabilities of player A to win/draw/lose agains player B. This has been even shown empirically to work very well (also reported in some chessbase article, cant remember where, sorry), especially when the rating differences are < 150 (surely the case in Bilbao Grand Slam Chess Final). It is very easy to calculate e.g. the probability of Carlsen winnning some tourament according to Elo ratings (at least with any computer). |
|
Sep-12-08
|
| alexmagnus: How? Normally a difference in rating can be translated into the scoring probability (i.e. probability to win+0.5*probability to draw). But how do extract the "winning part" of the result? |
|
Sep-12-08
|
| alexmagnus: It is impossible. Just because different players have different drawing profiles.... |
|
| Sep-12-08 |
| barbababa: You are right, sorry. I did it earlier for betting, but used also win/draw/lose probablities of different players. I did it by first defining the expected score in 100 games and then took into account the players draw probabilities to get the distribution of win/draw/lose agains each other. |
|
| Sep-12-08 |
| barbababa: <alexmagnus> E. g. for tomorrow I would have used the average of Carlsen's drawing probability with white and Anand's drawing probability with black.
Together with the Elo expected score it fixes the draw/win/lose probabilities. Then you just have to find bets where the odds are badly twisted. |
|
| Sep-12-08 |
| barbababa: And with Elo expected score I meant Carlsen's white Elo rating against Anand's black Elo rating expected score.
That means Carlsen white and Carlsen black are hanlded like two different players with different elo ratings. Betting would probaly be profitable even with these rules but almost nobody makes bets for chess games so there is not much to bet without ruining the odds. |
|
Sep-13-08
|
| slomarko: Anand is playing really badly in Bilbao. this is Carlsen's golden opurtunity to toast him! |
|
| Sep-13-08 |
| chesstoy: Does anyone know what Anand said in the press conference when the moderator asked him for his reaction on him and Carlsen working together? He spoke in Spanish, so could use some help. Thanks! |
|
Sep-13-08
|
| rogge: <Slomarko: Anand is playing really badly in Bilbao. this is Carlsen's golden opurtunity to toast him!> Yes, don't think he will, though...
To me Anand's performance is surprisingly bad, even considering the Kramnik-match coming up. Topalov has done better than I expected. |
|
Sep-13-08
|
| slomarko: i haven't seen Anand playing so badly in a long long time. i actualy think he's sandbagging a little bit. |
|
Sep-13-08
|
| rogge: Maybe? Anyways, looking forward to the wcc match! |
|
Sep-13-08
|
| slomarko: i'm not too interested. who wants to see number 5 in the world playing number 6? instead what should FIDE do is kick Kramnik and Anand out of the wcc match and invite Ivanchuk and Moro to play for the title. they are cleary the best players at the moment and they play highly spectacular chess. |
|
 |
|
< Earlier Kibitzing · PAGE 1181 OF 1181 ·
Later Kibitzing > |
