GM tournament draws

i know i've seen this topic on here before, but the corus chess tournament that is currently running made me annoyed again.

i really think that they should try to alter the scoring system in order to make games more excited. i'm tired of watching two GMs make 10 moves and call it a draw.

it has been suggested (and used in the past) that a scoring system where the 3 points is given for a win, 1 point for a tie, and 0 points for a loss may be a possible solution. i haven't really spent any time thinking about it, but i want to try and come up with a better scoring system that makes the goal of winning more lucrative for the players. any thoughts?

that's a good point. a minimum number of moves is required for a draw to be offered (obviously this would not include draws by three-fold repetition and draws of that nature).

although, i could still see players wanting a draw and after 10 moves and then both players simply making pointless waiting moves just to hit the magic number where a draw can be made.

obviously draws can't be erradicated, but i believe there could be a way to emphasize going for a win a little more.

There is an unwritten rule at Corus, 30 moves or 3 hours. It is not strictly enforced.

Invitations, or should I say no invitations, could be part of the solution. (You can only go so far though before nobody is invited.)

The prize for best win in each round, currently 500 euro (around $700 US) could be increased.

increasing prize money is a great idea that i didn't even consider.

also, maybe every time somebody wins a game there could be an addition to the overall prize money as well. more incentive for everyone to try and go for a win.

At the London Chess Classic they had a winners pool and 3 points for a win 1 for draw.

And yet it didn´t change the amount of wins and draws.

And how has this been a problem during Corus? Most draws has been actual games.

I think the problem with changing the scoring system can be highlighted by the use of game theory: (Prisioner's dilema, in particular)

If two GMs play two games then the maximal outcome for both players (in terms of scoring points) in a system where wins count for 3 and draws 1 is simply for both players to win one and then lose one. (two draws= 2points, win and loss=3pts)

Therefore, what we might see instead of 10 move draws is 10-move losses! (you deliberatly lose the game with the expectation/hope they will do the same...therefore you both end up with 3 points instead of 2. 

It is in the mutral interest of both players to co-operate rather than defect. Since GMs often play each other at numerous tournaments a 'Tit for Tat' strategdy would have room to florish.

{{ http://en.wikipedia.org/wiki/Prisoner%27s_dilemma }}

{{ http://en.wikipedia.org/wiki/Tit_for_tat }}

Secoundly, I would like to point out that draws are part of the game- the very nature of chess- why should not be trying to get rid of them in the first place. With that said, since no spectator really wants to see uncontested "10 move draws" we probably should add rules to make this more tricky...but even so, there is very little that can be done that can stop GMs agreeing to a result before the first move has been played. 

yes i agree with you're game theory example. i also don't think that 3 points for a win and 1 for a draw is ideal either. however, your example still is true if 2 points for a win and 1 for a draw. simply agreeing to draw with each other still "saves" each player as you stated. and i strongly feel that this happens currently (no matter how the games are scored) and that this is not "good" for chess in general imho.

and i also want to state that i am not trying to argue that we should erradicate draws (as i stated in post 3). just that we should try to increase the desire to win over drawing (which is honestly debatable right now in chess i think). i really believe that many GMs would be perfectly fine to agree to draws before even making the first move and that is not good for chess.

I got an Idea, do it like they did it in the 1800s, replay draws

that's actually not too bad an idea. although who knows how long it could go on for.

and yet that is exactly my point. the way the scoring system is set up it is beneficial for players to play for a draw (which on ocassion doesn't even need to be played for because both players will just agree to a draw after 10 moves - or possibly even before the match even begins if that was actually allowed).

and therein lives my issue. a change which either makes it worth more for a win or less for a draw. so far, replaying draws i think may be the best suggestion, but i may try to create some sort of scoring system if i have the time.

Even with 10000 points for a win and 1 for a draw, it's still beneficial to play for a draw in un-winnable positions. And those are rather common when you're playing as black in Corus A.

I like the Sofia rule -- simply outlaw draw offers. And otherwise I don't think there's a huge problem. Chess played at 2700+ level is a pretty drawish game, so it goes.

yes i agree that playing to draw in a lost position is definitely necessary and should not be changed or looked down upon. my issue is with games where there is so much play left and there is no clear advantage yet.

and yes, any simple scoring system like that would not make a difference. obviously the scoring system would have to be more advanced than simple points awarding.

outlawing draw offers does have potential. simply have the players play out to the draw (or three-fold repetition or stalemate or whatever). that could lead to some boring games though, but not every chess game can be exciting i guess.

Your issue is with games where there is a lot of play left, and that are still drawn extremely quickly. And you got annoyed by the current Corus.

How many such games have there been so far in the main group? Two or so out of 50+ games?

On game theory:

"however, your example still is true if 2 points for a win and 1 for a draw."

You are wrong in thinking this: mathmatically, you are of course correct, however, what you failed to consider was risk.

[A] Suppose I resign a game 3 moves in on the expectation the my oppeant will do the same. If he does indeed 'co-operate' then there is no problem... but what if he choses to defect (i.e play his best next game)? 

[B] Now suppose we agree a 3 move draw,  perhaps next game he choses to co-operate and offer a draw in which case we both net two points. But what his he choses to defect? that means I have to play my socks off next game!

Sceneraio [B] is allways more preferable, since there is less risk of being screwed over by defection:

After the completion of the first game the possible outcomes are

[A] 1-1 {Best case}, 0.5-1.5, 0-2 {Worst case}

[B] 1.5-0.5 {best case}, 1-1, 0.5-1.5 {Worst case}...

Looking at the best and worse cases we can see that stragedy [B] (offer the draw in game 1) offers better results if our oppenant choses to defect. (if he co-operates, then both stragedies are of equal value...but since there is always a chance (i.e risk) he might defect, [B] is clearly better because their is less risk: we mitigate the risk of defection  )

just the one day where there were two disappointing draws (yesterday i think...?)

yes increasing the prizes/prize money would definitely help i agree.

ah, okay i get your point. but that is exactly why i think a possible change in the system is required in order to place less value on draws.

I agree trigs

Yesterday (round 7) had Van Wely-Karjakin, a draw in 16 moves. That was one of the games I was thinking of. I don't blame them, Van Wely had five losses in a row (because he tries to win too hard?), and Karjakin was ill. But that is indeed an example of an early draw where there was play left.

If you mean the round before that, that had the quick draw in Anand-Carlsen, but that was a game where there was just no play left, and the same was true for Ivanchuk-Nakamura. In both cases it was a matter of repeat moves or have a worse position.

Chess is a draw, you know. As Black at top level you must assume that a draw is a good result (since there are many opening lines in which White can force a draw and Black is worse if he avoids it), and it's unavoidable that sometimes White plays into these lines, because you can't expect them to know absolutely everything there is to now.