Author: Dann Corbit
Date: 11:03:39 08/08/02
Go up one level in this thread
On August 08, 2002 at 13:45:22, Russell Reagan wrote: >On August 08, 2002 at 06:07:57, Roger Brown wrote: > >> Russell, let us use a chess argument here. Are you saying that if a player >>rated 2000 went on a rampage in an Open, shredded the field of players all rated >>higher than him/her over the course of several days and several games and met a >>GM (say 2600) in the finals, winning by a half point, he/she wouldn't be the >>best player? > >What if the GM had a momentary slip and hung his queen and gave the 2000 player >an easy win? The GM would beat the 2000 player the vast majority of the time. >Clearly the GM is the better of the two players, but the 2000 player was better >for one game, and you can't conclude anything based on one game. Sometimes you >can't conclude anything over a long match. I think that is the only point he was >trying to make, that no matter what the outcome of both of the tournaments in >October, it won't "prove" anything. Indeed. On the other hand, we should also recognize [and not forget] that the contests do mean something (more than nothing). In a one game match, a 1-0 result would have us tend to believe that the winner is stronger. Of course the error bar is unmentionably large. A 7-0 result would be pretty strong evidence that the winner was stronger. Still not a certainty, but more likely. The more data we gather, the more certain we become of the conclusion.
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.