Home > Archive > Chess politics > April 2005 > Re: Ratings experts: Please explain





You are viewing an archived Text-only version of the thread. To view this thread in it's original format and/or if you want to reply to this thread please [click here]

Author Re: Ratings experts: Please explain
Mike Nolan

2005-04-11, 5:59 pm

"Angelo De Pa1ma" <adpspammersgotohell@tellurian.net> writes:
quote:

>Why is going 50% against five players rated 2000 a better result than going
>50% against an 1800, 1900, 2000, 2100, and 2200. That's what I demonstrated.
>Why is it?


Do you understand that 'expected performance' is a non-linear function
while 'performance rating' is linear?

Here's a table of the expected performance of an 1800 player against
those other players:

1800 0.5
1900 0.359935
2000 0.240253
2100 0.15098
2200 0.090909

In 5 rounds against a 2000 player, the 1800 player would have a
cumulative expected performance of 1.201265.

Against the field you gave, the expected performance would be 1.342077.

Therefore, the same score would generate a higher post-event rating for
the player who played the 2000 player 5 times than for the one who played
the mixed field. (However, playing the same person more than twice
makes you ineligible for the bonus formula.)
--
Mike Nolan
Copyright 2003 - 2009 gamesreviews.net Software forum  PC Hardware reviews