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

BGonline.org Forums

win probability for a PR difference and match length

Posted By: Henrik Bukkjaer
Date: Friday, 21 September 2012, at 7:41 a.m.

In Response To: win probability for a PR difference and match length (Rick Janowski)

Back in 87 I think it was Michael Sørensen who implemented the DBgF rating system (based on Elo). If I recall correctly, there's a slight difference between that and the FIBS one. DBgF system use these formulas:

First, calculate the chance of an upset:
P_upset = 1/(10^(D*SQRT(N)/2000)+1)

Then calculate the delta amount of rating points to adjust after the match:
delta=N*P, where
P = P_upset, if the highest rated player wins.
P = (1-P_upset), if the lowest rated player wins.

------------------

@OP, you have to consider the time restraint. Swiss is much less time efficient, since all matches have to complete in a round, before you can plan the next round. You also need to allow time for planning rounds and announcing match-ups. Finally you have to allow time at the end for a tie-break match, since players can end up having en equal number of wins. If you make some other sort of tiebreak system (points derived), then you end up with other kinds of issues you'd probably prefer to be without (for an important championship 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.