4 ms·

> Chess, in theory, with absolutely perfect play, is a draw. True, but that doesn't mean that you can't treat chess like other sports and try to incentivize wi

> Chess, in theory, with absolutely perfect play, is a draw. True, but that doesn't mean that you can't treat chess like other sports and try to incentivize wins - for example giving win/loss/draw a 3:1:0 point ratio. The world championship is not a good format to decide "who's the best at chess", and anyway we already know who that is right now. Might as well treat it as a spectator sport and add some drama in my opinion.

dwighttk 4y ago
How would point ratio help if every game is a draw? Maybe if it was a round robin tournament that would help top two players competing for wins against less highly rated people and then drawing against each other. Like Liverpool and Man City
- arbitrage 4y ago
  My feel is that from a game theoretical approach, a 3:1:0 scoring ratio would encourage enough players to at least try for a win sometimes that it would restore some of the interest and competitiveness to the sport.
  - franknstein 4y ago
    Would you play tic-tac-toe for a win with that scoring ratio?
    - somenameforme 4y ago
      This is an inappropriate analogy. Chess at the highest levels is not drawish because the game is inherently drawish but because the meta of it is. The easiest way to illustrate this is with openings, though it applies throughout the game in different ways. Against e4 the Najdorf defense was once the opening of champions, being a major part of the repertoire of players like Kasparov and Fischer. In modern times it's an increasingly rare guest at the highest level. It's not because it's considered unsound or even slightly dubious - it's a rock solid opening that gives black real winning chances. But the problem is it also gives him real losing chances. It's complex, difficult to play, and if you get outprepared by your opponent you may lose without him even having to make a single move himself, which is really one of the worst feelings in the world. So instead the meta has largely shifted to openings that are more about minimizing risk where black, more or less, aims for a draw - and usually gets it. Changing the risk:reward ratios in a sufficiently extreme way is most certainly capable of changing the meta.
      - Scea91 4y ago
        This reminds me of football analogy where goals scored at away matches count for two in competitions like Champions League. I am not good at chess but would it help if win as black was rewarded more than win as white?
        somenameforme 4y ago
        The current trend has to been to increase the value of any win, with things like 3 points for a win and only 1 point for a draw. In my opinion increasing the points based on color is an interesting idea, but it's unlikely to be seriously considered because of the fact that it would disrupt the "symmetry" in chess both the sort of abstract fashion, and also in the specific - such as requiring tournament pairing systems based on score to be completely reworked. One tie break system, armageddon, does break the symmetry by giving black draw odds (he wins if the game is drawn) but less time in a blitz game. But it's very poorly regarded and generally only used as a this-game-MUST-have-a-decisive-result last resort.
  - sgjohnson 4y ago
    This implies that top-level players aren't trying for a win.
somenameforme 4y ago
An alternative is to have tie-breaks for every game, playing increasingly rapid time controls until the result is not a draw. 1. Classical game, win = 5 points. 2. Rapid game, win = 2 points 3. Blitz game, win = 1 point. 4. Armageddon game, win = 0.5 points. This point layout might backfire and make the classical games even less interesting because the huge edge for winning is probably going to motivate extreme do-not-lose style play, which trends towards draws. But, nonetheless, wins in classical should be weighted well above the tie-breaks.

dwighttk 4y ago

How would point ratio help if every game is a draw? Maybe if it was a round robin tournament that would help top two players competing for wins against less highly rated people and then drawing against each other. Like Liverpool and Man City

arbitrage 4y ago
My feel is that from a game theoretical approach, a 3:1:0 scoring ratio would encourage enough players to at least try for a win sometimes that it would restore some of the interest and competitiveness to the sport.
- franknstein 4y ago
  Would you play tic-tac-toe for a win with that scoring ratio?
  - somenameforme 4y ago
    This is an inappropriate analogy. Chess at the highest levels is not drawish because the game is inherently drawish but because the meta of it is. The easiest way to illustrate this is with openings, though it applies throughout the game in different ways. Against e4 the Najdorf defense was once the opening of champions, being a major part of the repertoire of players like Kasparov and Fischer. In modern times it's an increasingly rare guest at the highest level. It's not because it's considered unsound or even slightly dubious - it's a rock solid opening that gives black real winning chances. But the problem is it also gives him real losing chances. It's complex, difficult to play, and if you get outprepared by your opponent you may lose without him even having to make a single move himself, which is really one of the worst feelings in the world. So instead the meta has largely shifted to openings that are more about minimizing risk where black, more or less, aims for a draw - and usually gets it. Changing the risk:reward ratios in a sufficiently extreme way is most certainly capable of changing the meta.
    - Scea91 4y ago
      This reminds me of football analogy where goals scored at away matches count for two in competitions like Champions League. I am not good at chess but would it help if win as black was rewarded more than win as white?
      - somenameforme 4y ago
        The current trend has to been to increase the value of any win, with things like 3 points for a win and only 1 point for a draw. In my opinion increasing the points based on color is an interesting idea, but it's unlikely to be seriously considered because of the fact that it would disrupt the "symmetry" in chess both the sort of abstract fashion, and also in the specific - such as requiring tournament pairing systems based on score to be completely reworked. One tie break system, armageddon, does break the symmetry by giving black draw odds (he wins if the game is drawn) but less time in a blitz game. But it's very poorly regarded and generally only used as a this-game-MUST-have-a-decisive-result last resort.
- sgjohnson 4y ago
  This implies that top-level players aren't trying for a win.

somenameforme 4y ago

An alternative is to have tie-breaks for every game, playing increasingly rapid time controls until the result is not a draw. 1. Classical game, win = 5 points. 2. Rapid game, win = 2 points 3. Blitz game, win = 1 point. 4. Armageddon game, win = 0.5 points. This point layout might backfire and make the classical games even less interesting because the huge edge for winning is probably going to motivate extreme do-not-lose style play, which trends towards draws. But, nonetheless, wins in classical should be weighted well above the tie-breaks.