Author: James T. Walker
Date: 05:51:47 12/18/99
Go up one level in this thread
On December 18, 1999 at 04:08:52, Dave Gomboc wrote: >On December 17, 1999 at 15:25:24, Dann Corbit wrote: > >>On December 17, 1999 at 02:54:31, Dave Gomboc wrote: >>[snip] >>>>>The normal distribution does not accurately predict the occurance of large >>>>>upsets. Elo himself discussed this in his book, which you can reference for >>>>>further details. >>>>Of course, for a single contest, it does not accurately predict anything. So >>>>what does he say about large upsets? More frequent that predicted? Less >>>>frequent? I don't have the book (does anyone know where to buy it from?) >>> >>>Upsets are more frequent than predicted when using a normal distribution. >>> >>>What I don't understand is why people don't take real tournament results and >>>figure out the correct distribution. Maybe they actually do... <shrug> >> >>That's interesting, because it supports the opposite of the assertion that lower >>rated players will never beat higher rated players if the difference is great. >> >>What is the stated reason for A.E.'s statements? Is it mathematical or >>empirical? > >Empricial. I think that such a claim would have to be made based on >observation, no? > >Dave I suspect the reason empirical data would not agree with predicted mathamatical results is that humans sometimes play tournaments infrequently. Many people improve a lot between tournaments. I once new a player with a 1150 rating that was playing around 1900 chess at the chess club. He simply had not played in a tournament in 2 years. When he showed up at his next tournament there were a few large upsets. Jim Walker
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.