Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: next deep blue

Author: Robert Hyatt

Date: 06:18:39 01/24/00

Go up one level in this thread


On January 24, 2000 at 02:25:19, Howard Exner wrote:

>On January 23, 2000 at 22:41:21, Robert Hyatt wrote:
>
>>Check out the players.  USCF (at the top) is typically no more than 50 points
>>above FIDE.  Ken Sloan has a detailed list on his web page in the department
>>here...  and he did a lot of analysis to correlate FIDE to USCF. the +100 or
>>more is generally wrong.
>
>Grab a Chess Life issue that has the USCF ratings and slap the results of those
>players 2500+ USCF into a database
>along with their Fide ratings. There should not be an overwhelming number of
>such players. It would be interesting to compare the difference between the
>2500+ group with the 2300 - 2500 group. That though would be a little mork work.


This is what Ken Sloan did.  He took the FIDE database, the USCF database, and
found the common players.  He then "fit" a conversion formula to this data, and
found that +100 was way too big.  He even made it more fine-grained by doing
this "class by class".  (IE on the USCF 200-point breakpoints).

I never remembered the exact data because I didn't consider it interesting.
I do recall his asking me in the hall about what I thought the correction factor
should be, and when I responded with the usual +100, he said _wrong_.  And
showed me his results...

Bob



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.