Author: Uri Blass
Date: 03:03:22 06/24/05
Go up one level in this thread
On June 24, 2005 at 04:42:48, Graham Banks wrote: >Amir, > >I've stayed out of this spat till now. Terry did not treat you very well in >attacking you from the start, but he has made an attempt to bury the hatchet and >you've basically thrown it back in his face which is very disappointing >behaviour on your part. It won't endear you to some of those members who >supported you. > >Graham. I will not say that Terry is stupid but he posts nonsense in this discussion. The difference between fide elo and uscf elo is not relevant when the fide elo is used to calculate fide performance. If his claim is that there is a significant difference between fide formula and uscf formula to calculate performance then one of the formula is illogical and should not be used. The fide calculator certainly has big bugs because even with 6 out of 6 against average X you get only rating of X+75 that is absurd. I find nothing illogical in the results of the uscf in the case that all the opponents are equal. I can use the fide calculator also after knowing that the fide calculator give logical results when the result is not more than 50% I start from the assumption that adams play 6 games against 3000 player and ask what result he need to achieve to get his rating. 1 out of 6 give adams performance of 2727 1.1 out of 6 give adams performance of 2738 It seems that there is no difference betwen 1 and 1.04 or 1.05 and 1.1 so I guess that the fide rating calculator simply round the number but it is clear that adams need something between 1 and 1.1 out of 6 against 3000 player to get fide performance that is equal to his rating and it means that 3000 player needs something between 4.9 and 5 to get fide performance of 3000. 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.