Author: greg moller
Date: 11:41:26 05/04/99
Go up one level in this thread
On May 03, 1999 at 17:56:21, Will Singleton wrote: >On May 03, 1999 at 14:43:12, Dann Corbit wrote: > >>Everyone knows that computers are the best in the world at blitz already... > >Well, maybe not. :) > >Standard Blitz Bullet >2720 Ferret(C) 3137 Dlugy(GM) 2937 Data(C) >2720 OzsO(C) 3132 Ferret(C) 2930 WchessX(C) >2682 matemastr(C) 3101 Mofongo(C) 2914 Mofongo(C) >2677 RAINHA(C) 3076 FlyingPike(GM) 2912 crafty(C) >2676 Lynnett(C) 3058 crafty(C) 2855 Box(C) <snip> I wouldn't trust the blitz ratings on ICC. This is a place where most players duck one another all the time, handpick their opponents and only play each other when the rating differential seems favorable. All this has a negative impact on the rating system's accuracy. ICC should implement a system whereby it _forces_ players to regularly play others in their rating bracket, possibly by generating random server pairings, whenever a player's open for matches, in order to keep their ratings active AND accurate at the same time. At the moment ratings seem dictated by players' whims rather than their strength. Over-inflation at the top is clearly one of the farcical by-products, although there may be other factors contributing to that. Also, the list shown above is subjective (as the header suggests). Ratings can change very fast(another fallacy) on ICC, and picking a random top-ranking list could just as likely show a Computer on top. Incidentally, to find out the 'best' blitz players list, don't use the _best_ command, which only gives human players (and then not all of those, either), but use the _rank_ <playername> command, just as Will (correctly) did. Regards, gm
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.