< Earlier Kibitzing · PAGE 222 OF 222 ·
|Dec-18-14|| ||alexmagnus: By the way schism: note how Kramnik competed in both cycles before he eventually beat Kasparov.|
|Dec-19-14|| ||Petrosianic: Smart move. With a reunification match perpetually on the horizon, winning the FIDE title could have been equal to winning the PCA Candidates.|
|Dec-19-14|| ||perfidious: <visayan.....No tiebreaking system can take the pace of an outright match between tied players....>|
|Dec-19-14|| ||Conrad93: <By all means you are permitted your fantasies, but Anand was not at all likely to defeat Carlsen, even playing better than last year as he did. It was only with a supreme effort that he kept this match close in the face of a determined titleholder.>|
Another narcissistic and clueless comment from you.
Anand was not at all playing at the same level as he was at the Canditates. Anyone with a brain and some common sense would see and know this.
He had excellent positions against Carlsen and completely ruined them.
If he had the same precision as in the Canditates, he would have never lost.
|Dec-19-14|| ||perfidious: <donkrad: Another narcissistic and clueless comment from you.>|
Sez the projectionist as he looks in the mirror......
|Jan-03-15|| ||GeneM: Bidding.
In most cases where Armageddon Chess is specified in the rules as an if-needed tie-breaker, the rules specify the exact amount of time that both colors receive: but this is so unfair.
How can some guy who devises the rules possibly know the exact amount of times and relative times that exactly compensate for Black's advantage of draw odds?? He cannot know, he is just guessing, and nobody needs his guess.
Instead, the rules should specify the amount of time that White receives, and must have the two players BID for how little time they would be willing to receive as Black (lowest bid earns black pieces), in exchange for the draw odds that Black enjoys.
Bidding is essential for fairness in Armageddon; right?
GeneM , 2015-Jan-03
|Jan-03-15|| ||GeneM: Thought Experiment, about Armageddon:
Suppose White was given 60 minutes, and the players had to bid for Black's lesser amount of time. What would be the sensible amount of time to bid for Black?
Would master-level players still be willing to bid under 9 minutes to get Black and draw odds!?
|Jan-04-15|| ||kellmano: <GeneM> That happened in some tournament a couple of years ago. Sadly I can't remember more than that the chess community reacted with its typical conservatism - This is more like poker than chess. Also, I seem to recall that the amounts bid were surprisingly low.|
|Jan-04-15|| ||Shams: The armageddon game with bidding has been the final tiebreak for several years now in the US Championships. Kamsky has himself participated in two of them.|
|Jan-04-15|| ||kellmano: <shams> cool. That is what I was thinking of. Can you recall what times were successfully bid?|
|Jan-04-15|| ||Shams: <kellmano> If I remember correctly Black has typically ended up with a bit less than half of White's time (in a G/60 or G/45 maybe) but don't quote me on that.|
I dislike the "armageddon" game but if such a game has to decide matters I definitely support the players bidding for Black.
|Jan-04-15|| ||kellmano: <shams> What's the alternative when a tie has to be broken? Bullet seems less fair to me.|
|Jan-04-15|| ||Shams: <kellmano> I don't think classical tournaments and matches should be decided by non-classical games, unless there is no other option. |
I don't see what's wrong with having co-champions. Why shouldn't matches be drawn just like games are?
|Jan-04-15|| ||Kinghunt: <I don't see what's wrong with having co-champions. Why shouldn't matches be drawn just like games are?>|
The point of a match is to produce a winner. If a world championship match is drawn, do we have two world champions? Who plays in the following title match? If that match is also a draw, do we have three world champions?
|Jan-04-15|| ||Shams: <Kinghunt> <The point of a match is to produce a winner.>|
Is it? That's what many non-chessplayers say about individual games of chess. Fifty percent draw rate, are you kidding?
|Jan-04-15|| ||Petrosianic: The point of an individual game is also to produce a winner, but people sometimes fail to accomplish this. It's not possible to make every game produce a winner, but it is possible to create match rules that will guarantee this.|
|Jan-04-15|| ||visayanbraindoctor: <Shams: <kellmano> I don't think classical tournaments and matches should be decided by non-classical games, unless there is no other option.>|
I agree. And there are options that use only classical games in the tiebreaker. One I mentioned above, a two game classical mini-match in which <the Challenger gets two whites. This is his advantage. The sitting champion retains the Title if he ties. This is the champion's traditional advantage that he retains, but for which he must pay for with two blacks. Fair trade-off IMO.>
Variations of this theme are conceivable. For instance a four game classical match, and if this is tied then follows a two game mini-match with rules as above. The purpose is to address two recurrent issues:
1. Coming up with a clear winner using only classical games, or if it's a tie, then a reasonable tie breaking rule that avoids any potential scenario of a chess world schism.
2. Minimize financial costs.
|Jan-19-15|| ||Everett: Kramnik's fading in this tourney still comes off to me as a surprise. |
If there is a psychological issue with Levon, look no further than Aronian's first game to understand his current problems.
|Jan-19-15|| ||alexmagnus: <visayan> Can you prove, mathematically, that the tradeoff is fair?|
Actually I even doubt such a proof is possible. Too many uncertainties (how bis is White's advantage, how high is the draw rate etc.).
|Jan-19-15|| ||Rolfo: Everett, Aronian may be temporarily burnt out, lost the flame|
|Jan-19-15|| ||Mr 1100: How about “long armageddon chess” (i.e., with near-classical time controls)?|
So, Black has draw odds, but White gets more time. White starts with 105 minutes (1h 45m) and Black starts with 90 minutes (1½h ).
After 40 moves, add 50 minutes for White, 35 minutes for Black, and from move 40 to move 60, both players get 30 second increments, and if the game continues beyond move 60, then White gets 30 second increments, and Black gets 20 second increments.
Of course, you could get some ∼2600-rated players to play maybe a 100 trial games with these time controls, carry out a statistical analysis, and then adjust the time controls accordingly, so neither side has an unfair advantage.
Now you could have a twelve-game match of standard classical games. If that ends in a 6-6 tie, then the players play a further “mini-match” of five “long Armageddon chess” games (as proposed above), which I expect would be guaranteed to produce a decisive result one way or another. That might be considered more acceptable than a single 6-to-5 minute game being used to decide the outcome of a match.
Another option might be, two tie-break days. On day 1, they play a mini-match of 8 short Armageddon games – i.e. 6 minutes for white, 5 minutes for black. However, if this 8-game match ends in a 4-4 tie, then on to tie-break day 2, where they play another mini-match, this time with 7 games, or perhaps 9, which would again be guaranteed to produce a decisive result. Again, the point being, this might be more acceptable than one single short game being used to decide the outcome of a “classical” match.
Or, some of you might think that the idea is just a little frivolous. But I’d like to know what you folks think.
|Jan-19-15|| ||AylerKupp: <<Mr 1100> How about “long armageddon chess” (i.e., with near-classical time controls)?>|
As good an idea as any and probably better than most. My one suggestion, to eliminate wear and tear on those 2600-rated players, is to determine the initial, hopefully somewhat reasonable time controls, by having chess engine vs. chess engine games. Then, depending on the results, adjust the times for both sides until an acceptable balance of wins vs. losses is found. Then the actual time controls can be further adjusted according to the result of human vs. human competition.
A few additional points:
1. Since we are only interested in wins and losses, any draws should probably be ignored. Alternatively we can use the winning percentage for both sides.
2. In order to establish an "equal opportunity" for both sides, the number of wins for White and Black would not necessarily have to be equal. If you look at game databases, the winning percentages for White and Black are in an approximate 55/45 ratio, not 50/50. So an "equal opportunity" set of time controls must take this into account.
3. In order to reach a statistically significant (e.g. 95% confidence level that the result was not due to chance) "equal opportunity" set of near-classical time controls, probably many more than 100 games at each time control would be necessary, particularly if draws are discarded.
But, all in all, a suggestion worth investigating.
|Jan-20-15|| ||Everett: <Jan-19-15 Rolfo: Everett, Aronian may be temporarily burnt out, lost the flame>|
You may be right.
|Feb-20-16|| ||Hawkman: If there were 2 Candidates tourneys where the 2 winners played each other, the 2nd one would actually have greater depth at the bottom with all 8 seeds being higher rated than Svidler. The 8 seeds by the live ratings would be:|
#2 in the world Kramnik
#7 in the world MVL
#9 in the world Liren
#10 in the world So
#11 in the world Eljanov
#13 in the world Chao
#14 in the world Harikrishna
#16 in the world Grischuk
|Mar-26-16|| ||activechess55: Giri is an eternal optimist. That's why his glass is always half full.|
< Earlier Kibitzing · PAGE 222 OF 222 ·