Quote:
Originally Posted by BobJoeJim
What exactly are the tiebreakers for this event? It's looking like there's a decent chance it will end up mattering, I can't find them anywhere on the official website, and you mentioned in your latest video that you helped devise them so I was hoping you could fill us in
I'll mention them in detail in my next video hopefully, if I remember.
Basically it's none of this endless series of blitz games crap that you usually see, however it'll still be controversial, although will at least be a "real" chess game that the fans always claim to want to see.
Basically assume its a 2 way tie (there are other systems for bigger ties but I'm going to stick with this one for now since it's simplest). The two players play one game of 45+5, so it's a pretty slow time control. The catch is that black gets draw odds.
Now you've probably seen this scenario a million times in which a bunch of bumbling organizers decide that they know what the fairest time odds are and so they just randomly decide that some time is fair (usually this number changes from year to year), randomly assign one color to one of the players, and then invariably someone bitches about how it's unfair to be white or unfair to be black or whatever etc etc.
So the format here is as follows: Both players make one sealed bid in which they bid on how much time they are willing to give away in order to choose their color.
So if it's Kamsky vs Nakamura for example in the tiebreak, Let's say Kamsky is willing to start with 35 minutes to choose his color, while Nakamura is willing to start with 30. This would mean that Nakamura bid the lowest time, so he would get 30 minutes, Kamsky would get 45 minutes, and Nakamura could either choose to have white or take black and draw odds.
The upsides:
1. The game is a relatively real chess game with no ridiculous time scrambles happening on move 15.
2. For media it's great because it's known that by the end of the game, the champion will be determined. There won't be 10 more blitz games going back and forth, and there won't be the final armageddon blitz game which is really super random looking.
3. It's completely fair to all parties. No one has an inherent advantage. No one can complain that they were cheated by the rules being shaded towards one color.
4. There will be a financial bonus if you win, so that even though black gets the title on a draw, there will be a few thousand dollars riding on them winning the game, so if they are totally winning they aren't always going to just accept a draw immediately, and so the actual real chess part of the game will be more legitimate than if black has absolutely no reason to win.
5. Please note that the entire tiebreak has to last a maximum of 2 hours, due to timing involving closing ceremony and etc. This format allows for this, even with a 3-4 player tie.
The downsides:
1. Of course I hate someone winning something on a draw, however I also hate all other tiebreak systems I've seen. The nonstop blitz games are annoying, and while G 45 isn't 40/2 SD 1, the chance of a well played game, worthy of the US Championship, goes way up.
2. It could be considered confusing, however the players have known about it for months and have had ample time to consider the ramifications of what their bids should be.
3. The one thing that really bugs me is there is no provision for 5+ players if there is a tie. Right now they just eliminate everyone under 4th place on tiebreaks. This is one aspect of this system that I disagree with and I think it'd be pretty easy to allow for more players, but the organizers didn't want to deal with. Fortunately it's unlikely that more than 4 players will tie for first, but I've certainly seen crazier things.
Anyway I like the system personally, and it was shown to some top players/ tournament participants on the committee and they passed it. However I'm almost 100% certain there will be huge criticism if this tiebreak occurs
Ideally I'd like to use this system but make the time control much slower and make sure the incentive for black to win is still very significant (I'm not sure if it's quite as significant as I'd have liked), but due to time constraints it's not possible.
Last edited by curtains; 05-14-2009 at 05:15 AM.