Author: Dann Corbit
Date: 10:54:09 01/19/01
Go up one level in this thread
On January 19, 2001 at 06:50:30, Walter Koroljow wrote: >On January 18, 2001 at 17:44:24, Dann Corbit wrote: > >>The problem is that ELO calculations simply do not work that way. You cannot >>combine different opponents into a single foe and produce a composite ranking. >> >>I think also that the assumption of sporadic single measurements of different >>events will be the same as repeated measurements of the same experiment is >>seriously flawed. I believe that time will prove I am correct in this >>assessment. >> >>I might be wrong. But as I see it, your experiment has no controls whatsoever. >>How can you possibly reach any kind of conclusion? > >Well, we will just have to disagree. Perhaps some acknowledged expert will >help? Anyone out there? I do want to clarify one point. I am not disagreeing with your math, which I think is correct. I am disagreeing with your model.
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.