[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums

The FIBS rating formula : why sqrt(n) ?

Posted By: Maik Stiebler
Date: Thursday, 15 September 2011, at 3:22 p.m.

In Response To: The FIBS rating formula : why sqrt(n) ? (Maik Stiebler)

Let me try to be less cryptic: I think a reasonable estimator of the winning percentage in the cubeless and gammonless model would be p(N, rating difference)=f(rating difference * sqrt(N)), where f is some (close-to) normal cumulative distribution function. This p, I think, would decrease like exp(-N) for large N. The FIBS formula is not a bad approximation, if you disregard the tails; and nobody really cares if p is 1e-10 or 1e-20 anyway.

Messages In This Thread

 

Post Response

Your Name:
Your E-Mail Address:
Subject:
Message:

If necessary, enter your password below:

Password:

 

 

[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums is maintained by Stick with WebBBS 5.12.