Author: Dann Corbit
Date: 13:39:20 06/15/04
Go up one level in this thread
On June 15, 2004 at 16:20:45, Peter Fendrich wrote: >On June 15, 2004 at 15:33:36, Rolf Tueschen wrote: > >>On June 15, 2004 at 15:14:45, Dann Corbit wrote: >> >>>On June 15, 2004 at 14:30:48, Rolf Tueschen wrote: >>> >>>>On June 15, 2004 at 13:38:56, Dann Corbit wrote: >>>> >>>>>On June 12, 2004 at 16:57:06, Peter Fendrich wrote: >>>> >>>>>>Well after 10 games you can't even rely on the accuracy of error bars and >>>>>>shouldn't use them (based on the bell curve) but the rating is well defined as >>>>>>one value. "x's rating after 10 games is 2739" is a correct statement. >>>>> >>>>>That is misleading and very bad science. >>>>>Why not say that the rating is >>>>>2739.8356245494183672715153891736273563 >>>>>? >>>>>Even though you are not even sure about the leading 2. >>>>> >>>>>>/Peter >>>> >>>> >>>>Is Peter one of the statistical scientists behind the creation of SSDF? I can't >>>>believe it. >>> >>>I think he is just following current practice (for instance a FIDE, USCF, BCF >>>etc. rating does not descibe its own accuracy). >>> >>>When a player is said to have a rating of 2345 what does it mean? We have no >>>idea, although for the very rough ratings they are called 'provisional'. >>> >>>It is much better to describe the ratings like an entry in (for instance) the >>>CRC handbook of chemistry and physics, with a tolerance figure to show how well >>>known and accurately described the rating is. >>> >>>But since nobody else does it, they do not follow the practice either. >>> >>>On the other hand, they provide the necessary data to produce the same >>>information, with their error bars. >> >>Dann - of course!! But isn't this here about what should be the best to do >>(scientifically)? I mean, could we play stupid only because testing has no real >>importance in a sport where winning is the ultimate ratio? I thought a minority >>here should guarantee a certain state of the art... > >We could in fact invent a much better rating system for chess engines. The ELO >system is designed for humans with a sparse number of games and not for hundreds >and thousands of games in long matches. But it works. >IMHO it's however not very practical with another rating system when the ELO >system is the chess rating standard. There used to be a nice web site by Royal C. Jones on alternative Elo calculation methods. I am no longer able to find it. Here is a C++ program that performs his alternate calculations in a simulation: ftp://cap.connx.com/pub/tournament_software/prog10.cpp Here is the letter where I asked his permission to use the code: ftp://cap.connx.com/pub/tournament_software/Re%20Your%20chess%20rating%20systems.txt
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.