Author: Ricardo Gibert
Date: 14:25:23 03/02/05
Go up one level in this thread
On March 02, 2005 at 14:50:53, Drexel,Michael wrote: >On March 02, 2005 at 13:57:31, Robert Hyatt wrote: > >>On March 02, 2005 at 12:19:13, Terry Giles wrote: >> >>> >>>Hi CCC friends, >>> >>>After playing through some of the very short "grandmaster" draws at Linares, I >>>feel that it's about time some of these tournament organisers started to award 3 >>>points for a won game in an effort to get the players to try and win a game >>>instead of fearing a loss. Chess today, well at least at the stratospheric >>>heights of the "super-grandmasters", is far too technical and theory laden for >>>most of the 'general' public to really appreciate and most of it has already >>>been prepared and analysed at home. Something needs to be done to liven up the >>>game, before the machines take over! >>> >>>Terry ;-) >> >> >>Why don't you play some tournaments to see what it will do. Of course, some of >>us would need to modify our programs to take advantage of this. Right now we >>assume loss=0, draw = .5 and win = 1. I could sort of tweak Crafty to >>understand this by twiddling with the draw score, but it would begin to think >>that draw=loss, win=good, which is not exactly right. >> >>But it would be interesting to have some real data to see what this would do to >>the game, when suddenly trying for a win is worth the risk. > >The idea is IMO nonsense. >The illogical 3-points rule does not make any sense at all in chess. > >An idea that makes some sense is from the Persian game Shatranj and was >recommended by Emanuel Lasker. > >He proposed to introduce again a stalemate-win (Pattsieg). > >For example: >4 points for mate or resignation of the opponent >3 points for the player who delivers stalemate >2 points for a draw >1 point for the stalemated player >0 points for a loss > >This certainly wouldn't avoid short draws but it would make the game less >drawish and might reward players who tend to fight in the long run. > >Michael I've always rather liked this idea. I remember once thumbing through an endgame book and realizing that for the most part, the variations would remain the same, while only the evaluations would change. There would be added material e.g. some KBK endings would be interesting, while virtually nothing would be lost. Another idea to make chess less drawish is to make repetitions illegal. The drawback is that some endings would be pretty insane. You would be spending half your time trying to figure out whether a move is legal, because it might repeat the position from 20 moves ago. Ugh! In blitz, you can forget it. Fortunately, there is a simple fix. Just make trivial repetitions illegal i.e repetitions of the position from 4 ply ago. This is very much like the difference between ko and super ko in the game of go. Repetitions with longer cycles would be treated the same way they always have. If there are no other legal moves, then it would be stalemate. This rule would make sacrificial attacks in the middle game a bit more prevalent, since often the defenders only out is to repeat. Instead, he will be compelled to play a move that loses instead. More significantly, a lot of endings that were formerly draws would then be wins e.g. [D] 4k3/8/8/3Kp3/4P3/8/8/8 w - - Normally drawn, with the repetition rule change, white would win with 1.Kxe5 Ke7 2.Kf5 Kf7 3.Ke5 and now 3...Ke7 would repeat, so Black must choose a different move and go on to lose. The only significant drawback is the change to the game would be quite substantial making it unacceptable to too many chess players.
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.