Author: Dann Corbit
Date: 16:50:47 03/06/04
Go up one level in this thread
On March 06, 2004 at 15:21:39, Geert van der Wulp wrote: >On March 06, 2004 at 09:42:26, Albert Silver wrote: > >>On March 05, 2004 at 21:54:11, GuyHaworth wrote: >> >>> >>>A common confusion is to try to compare FIDE ELO ratings (about a community of >>>human players) with SSDF ELO ratings (about a community of computers). >>> >>>This is understandable, as the metric used is referred to as 'ELO' in both >>>cases. However, 'ELO' is the measurement system rather than the measurement, >>>the 'scale' rather than the 'weight'. >> >>(snip) >> >>>I feel that less confusion would certainly be the result. >>> >>>g >> >>The following was posted here about 3+ years ago: >> >>http://chessprogramming.org/cccsearch/ccc.php?art_id=82028 >> >>Subject : SSDF ratings are 100% accurate >> >>Posted on December 12, 1999 at 08:49:08 >> >>Hi all, >> >>As the issue of SSDF ratings, and their comparative value with USCF or FIDE >>ratings, has been a recurring theme and a number of threads have sprouted >>recently, I thought I'd share my opinion (self-plagiarized) as I think it is >>relevant and might shed some light on the matter. >> >>SSDF ratings: inflated or not? >> >>Here's what I think: the ratings are not inflated in the least bit. Sounds crazy >>doesn't it? But it's not. People get too caught up trying to make these futile >>comparisons between SSDF ratings and human ratings whether USCF, FIDE, or >>whatever. The point is, and it has been repeated very often, there simply is no >>comparison. The only comparison possible is that both are generated using Elo's >>rating system, but that's where it ends. Elo's system is supposed to calculate, >>according to a point system, the probability of success between opponents rated >>in that system. The SSDF rating list does that to perfection, but it is based on >>the members of the SSDF only. If you put Fritz 5.32 on fast hardware up against >>the Tasc R30 or whatnot, it will pulverize the machine. The difference in SSDF >>ratings accurately depicts that. It has NOTHING to do with FIDE or USCF ratings. >>The rating of Fritz, Hiarcs, or others on the SSDF rating list depicts their >>probability of success against other programs on the SSDF list, and that's it. >>It doesn't represent their probability of success against humans because humans >>simply aren't a part of the testing. If you want to find out how a program will >>do against humans then test it against humans, and then you will find it's >>rating against them. The SSDF rating has nothing whatsoever to do with that. As >>was pointed out, I believe the SSDF ratings pool is a pool that is COMPLETELY >>isolated from all others and as such cannot possibly be compared with them. >> >>Albert Silver That's on the one hand. On the other hand, there is quite likely to be *some* correlation between SSDF strength and FIDE strength. In other words, Kasparov will probably pulverize a 15 year old chess machine with a 6502 processor and Shredder on a 1.5 GHz box will give a 2000 Elo player fits (most likely) at 40/2 or slower. However, we don't know what that correlation is. The only way to find out would be to measure, and in that case we could form a single coherent list. >You are completely wrong. A rating can only be 100% accurate if the participants >have played an infinite number of games. The SSDF would make itself rediculous >if they claimed to be 100% accurate. They give statistical confidence intervals >with all the ratings that they publish. I think we can say that they are 99% accurate, give or take the error bars. ;-)
This page took 0.01 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.