Author: Bella Freud
Date: 11:03:37 11/17/99
Go up one level in this thread
On November 17, 1999 at 05:47:57, Shep wrote: >On November 17, 1999 at 01:18:49, Jouni Uski wrote: > >>Yes, if Chess Tiger gets 2726 rating from SSDF games, it goes to >>place 4. in PCA world list (after Kramnik and BEFORE Anand). Ridiculous or ?? > >Apart from the problems involved with comparing disjoint rating lists >(SSDF/FIDE), it is also problematic to compare "performance ratings" with >"ratings". >Remember both Kasparov and Karpov had tournaments (Linares) with performance >ratings beyond 2900, yet Karpov has never been 2770+ on the FIDE list IIRC. >Fritz 6 supposedly had a 2800+ performance rating at the Frankfurt Masters this >year. See my point? > >--- >Shep Easy task for a programmer. C, Basic, anything. Imagine a program that is objectively a 40% program against one opponent, or even a 40% program against a pool of opponents. Eg it wins 40 per 100 of its games over a massive run of games. Or substitute any numbdre you like 30%, 60%. Write a program to generate a bit stream 0010101100111 etc, 1000 bits long. Where each bit has 40% chance to be 1, and 60% chance to be zero. Tell us the win percentage out of 1000. The number of 1 bits divided by 10. Chances are it won't even be 40, probably close though. Now tell your program to take random streams of say 10, 15 or 20 bits within the sequence. Tell it to print out the percentage win rate of the first 10 streams. If you do it, you'll get numbers like 12,45,18,63 ....... Try it. And think about the implications for short run grades, or TPF ratings. Bella
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.