Author: Robert Hyatt
Date: 19:25:54 06/16/99
Go up one level in this thread
On June 16, 1999 at 01:04:40, KarinsDad wrote: >On June 16, 1999 at 00:08:46, eric guttenberg wrote: > >> >>In a 7-round tournament luck is clearly a major factor, especially >>where the competitors are machines running programs within 20-30 >>elo points of each other. It may not be the biggest factor but >>anyone who has seen a lot of computer v computer results knows that in >>a tournament where the top contenders only play 3 or 4 games against >>other top contenders the result might be very different than if the >>top programs played 40 games against its top rivals. >> >>eric > >I disagree. > >Luck is not a major factor at all. > >Luck is one of those terms that is bandied about when some improbable event >occurs. However, you have 30 chess programs here. So, on average, they each have >about a 3.3% chance of winning. But just because the average chance to win is >low and hence the chance to win of the winning program is low (i.e. improbable), >does not mean that whichever program wins will be lucky. It means that one of >many improbable events will be the final result. The winning program will be >skillful. Chess is a game of total skill, not luck. Backgammon is a game of >skill and luck. > >If there were 30 humans playing in the tournament and you were one of them and >you won the tournament, then you would not say that you won it because you were >lucky. You would say that you won it because you had a higher winning percentage >than any other competitor (or maybe some more colorful euphemism). Four to six >round chess tournaments occur every week in the world and a lot of these are >split up in to 200 point differential sections. Nobody (generally) goes around >saying that the winner of these tournaments is lucky, so why would they say that >for a seven round tournament? > >When Anand played Karpov for the World Championship, the previous elimination >rounds consisted of the winner of two games (and quick chess tie breaks) per >round. Only 2 games! But, nobody said (at least to my knowledge) that Anand was >lucky to beat Shirov, Gelfand, and Adams after winning his first 3 rounds. >Everyone just took it in stride that Anand would be playing for the Championship >because he was victorious. End of story. > >The program that wins the WCCC will be one that plays extremely well. > >KarinsDad :) We'll just have to agree to disagree on this one. I have personally played in over 20 computer chess tournaments. I watched Belle and chess 4.6 play a tactical battle in 1978 where both were failing high, failing low, in a wild tactical game... and luck definitely played a role... because when you have two moves that look equal, and one wins and one loses, and you play one of those moves without _knowing_ the outcome, luck plays a role. Have you never, as a human, played a move, and then _after_ you played it said, "wow, I didn't realize that instead of controlling d5, this actually wins a pawn" or "wow, I didn't realize that I controlled d5, but I lose to a move I overlooked." Yes there is skill in those decisions, but there can be a bit of luck. Not that the outcome of a game is a coin-toss away... but I'd bet that some small percentage of outcomes is a 'coin-toss' decision... maybe 5% or so...
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.