Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: how not to calculate performance

Author: Sune Fischer

Date: 20:45:49 10/30/04

Go up one level in this thread


On October 30, 2004 at 21:59:36, James T. Walker wrote:
>
>As long as you realize you are making a "best guess" and not giving a real
>rating that's fine.

It was of course a back of the envelope, I have not derived it rigorously.

I think a more accurate guess can be made if you solve for the case where the
binomial distribution should give 50% or more for the X straight wins.

> The problem is that in real life untill you actually score
>some real points you cannot get a score which is anything but a guess.

It will always be a guess as long as all you have is a finite sample.

>The
>formula assumes you are 400 points below the average opponents and that is true
>in both cases cited above.  Nobody says it's perfect.  It's only an attempt to
>give a provisional score untill some concrete data is available.  The 400 >points below your opponents average is as good a guess as any and easy to >calculate.

The provisional score is actually a different story. You might be thinking of
the Glicko system where a provisional score is used to accelerate the process.

The 400 is more like a rule of thumb, it's not accurate and it's certainly not
independent of the number of games.

>Also how did you come up with a win expectancy of 1/4 point?  Simply because
>that's the next possible score closest to zero?

Imagine that you have an infinitely large bag and that this bag contains
infinitely many chips.

You pull out 4 chips, all of them red.
How certain can you be that all of the chips in the bag are red?

-S.
>Jim



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.