Kramnik was never as strong as Anand but....................

[Uri Blass]

<snipped>

So what happened to Kramnik rating it went down from 2802 in 2001 to 2766 in 2007. Was it that Kramik rating was higher in 2001 since he was avoiding to play stronger players and now the competition is much harder

The difference is completely insignificant, and does not allow us to make any conclusions whatsoever about the relative strengths of Kramnik in 2001 and in 2007. I can understand that average chess players can make the mistake of attaching such significance to ratings, but I find it astonishing that computer chess enthusiasts draw conclusions based on a rating difference of less than 40 Elo points. We all know that in order to reliably measure the strength of a chess program, we need hundreds (preferably thousands) of games. For humans, we would probably need even more games to measure the strength to the same degree of accuracy, because human players play much less consistently. Tord

You could be right in case that games are from random positions but if we talk about games with opening preperations then
I am not sure if humans play less consistently.

Humans are more consistent than computers in the choice of the opening.

Uri