Author: blass uri
Date: 00:40:46 01/07/00
Go up one level in this thread
On January 07, 2000 at 03:11:31, O. Veli wrote: >On January 05, 2000 at 06:10:07, blass uri wrote: > >>3)The rating is also not perfect in predicting the playing strength of humans. >>Suppose an 1600 player based on previous topurnaments train some years against >>computers without playing in tournaments and after many years of training go >>back to tournaments and gets performance of 2600 in 9 games(including draws and >>wins against GM's). >> >>It is logical to assume that he deserves more than 2200 but >>the rating system is not going to give this player even 2000. > > The rating system rates the last 80-100 games that you play. If 1600 was based >on 9 games, then the above player would get 2100. If 1600 was based on 80 games, >it would be much lower than 2100, and that is logical. S/he has to continue that >performance for some time to show the real strength. If after the training, the >player really becomes a 2600 player, then it would take a couple of tournaments >(8-9) to show the real rating. On the other hand this performance could be a >fluke, and the rating system considers this. I considered the fact that the performance can be misleading so I said that the guy deserves at least 2200. I did not say that he deserves 2600 but I think that it is illogical to assume that the performance in 9 games can be wrong in more than 400 elo. a good rating system should give the best guess of the real strength of the player based on the history. In this case I think that the best guess should be more than 2200(maybe 2300). My example is an extreme example and I know that it does not happen usually but I wanted to prove that the rating system can be improved. Uri
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.