ChessUSA.com TalkChess.com
Hosted by Your Move Chess & Games
 
 FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   RegisterRegister 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

Advantage for White; Bayeselo (to Rémi Coulom)
Post new topic    TalkChess.com Forum Index -> Computer Chess Club: Programming and Technical Discussions Flat
View previous topic :: View next topic  
Author Message
Edmund Moshammer



Joined: 03 Dec 2007
Posts: 598
Location: St. Gallen, Switzerland
PostPost subject: Re: Advantage for White; Bayeselo (to Rémi Coulom)    Posted: Sun Mar 04, 2012 4:13 pm Reply to topic Reply with quote
hgm wrote:
It is interesting that the Gaussian seems to give a better fit, despite the fact that the ratings were derived using the logistic. (Now hope the logistic doesn't give a better fit on ratings derived with the Gaussian...) It could be that for this large data set based mostly on low delta-Elo data the obtained ratings are not very sensitive to the model used.

What worries me is that the empirical data seems steeper than the curve of the model from which they were derived. That means that Elo differences you have to put into the logistic formula to get the true score percentage are larger than those spit out by BayesElo. In other words, BayesElo systematically underestimates rating differences, compressing the rating scale.

I wonder if this is an artifact caused by the prior. Do you calculate the ratings for this data set yourself? If so, could you recalculate them using a smaller prior (e.g. 0.1 in stead of the standard 2.0)?


I cannot easily recalculate the Elos; I have taken the data from CCRL.

Taking into account the elo-delta of the players my model predicts the game outcome on average 0.15 percent-points better.

Code:
Elo(x) = 1 / (1 + 10^(x/400)) 
Elo^-1(x) = 400 *LOG10(x / (1-x))
P_draw_given_elodelta = Elo(-delta-whiteadvantage) * (1 - Elo(-delta-whiteadvantage)) * 40/25
Elo_draw = Elo^-1(P_draw_given_elodelta) + 0.096 * eloavg + 75
Back to top
View user's profile Send private message Visit poster's website
Display posts from previous:   
Subject Author Date/Time
Advantage for White; Bayeselo (to Rémi Coulom) Edmund Moshammer Sat Mar 03, 2012 11:22 pm
      Re: Advantage for White; Bayeselo (to Rémi Coulom) H.G.Muller Sun Mar 04, 2012 6:56 am
            Re: Advantage for White; Bayeselo (to Rémi Coulom) Edmund Moshammer Sun Mar 04, 2012 8:49 am
                  Re: Advantage for White; Bayeselo (to Rémi Coulom) H.G.Muller Sun Mar 04, 2012 10:44 am
                        Re: Advantage for White; Bayeselo (to Rémi Coulom) Lucas Braesch Sun Mar 04, 2012 11:06 am
                              Re: Advantage for White; Bayeselo (to Rémi Coulom) H.G.Muller Sun Mar 04, 2012 11:29 am
                        Re: Advantage for White; Bayeselo (to Rémi Coulom) Edmund Moshammer Sun Mar 04, 2012 2:42 pm
                              Re: Advantage for White; Bayeselo (to Rémi Coulom) Edmund Moshammer Sun Mar 04, 2012 2:52 pm
                              Re: Advantage for White; Bayeselo (to Rémi Coulom) H.G.Muller Sun Mar 04, 2012 3:26 pm
                                    Re: Advantage for White; Bayeselo (to Rémi Coulom) Edmund Moshammer Sun Mar 04, 2012 4:13 pm
                                          Re: Advantage for White; Bayeselo (to Rémi Coulom) H.G.Muller Sun Mar 04, 2012 4:33 pm
                                                Re: Advantage for White; Bayeselo (to Rémi Coulom) Edmund Moshammer Sun Mar 04, 2012 5:18 pm
                                    Re: Advantage for White; Bayeselo (to Rémi Coulom) Adam Hair Sat Jun 30, 2012 11:00 pm
                                          Re: Advantage for White; Bayeselo (to Rémi Coulom) Adam Hair Mon Jul 02, 2012 1:25 pm
                              Re: Advantage for White; Bayeselo (to Rémi Coulom) Adam Hair Sun Jul 01, 2012 3:48 am
                                    Re: Advantage for White; Bayeselo (to Rémi Coulom) Adam Hair Sun Jul 01, 2012 8:47 pm
                                          Re: Advantage for White; Bayeselo (to Rémi Coulom) Edmund Moshammer Sun Jul 01, 2012 9:55 pm
                                                Re: Advantage for White; Bayeselo (to Rémi Coulom) Adam Hair Mon Jul 02, 2012 12:37 am
                                                      Re: Advantage for White; Bayeselo (to Rémi Coulom) Adam Hair Mon Jul 02, 2012 1:27 pm
                        Re: Advantage for White; Bayeselo (to Rémi Coulom) Marco Costalba Sun Jul 01, 2012 7:17 am
      Re: Advantage for White; Bayeselo (to Rémi Coulom) Lucas Braesch Sun Mar 04, 2012 8:53 am
      Re: Advantage for White; Bayeselo (to Rémi Coulom) Daniel Shawul Sun Mar 04, 2012 12:01 pm
      Re: Advantage for White; Bayeselo (to Rémi Coulom) Rémi Coulom Sun Mar 04, 2012 1:29 pm
            Re: Advantage for White; Bayeselo (to Rémi Coulom) H.G.Muller Sun Mar 04, 2012 2:20 pm
                  Re: Advantage for White; Bayeselo (to Rémi Coulom) Rémi Coulom Sun Mar 04, 2012 3:52 pm
                        Re: Advantage for White; Bayeselo (to Rémi Coulom) H.G.Muller Sun Mar 04, 2012 4:12 pm
                              Re: Advantage for White; Bayeselo (to Rémi Coulom) Rémi Coulom Sun Mar 04, 2012 5:13 pm
                                    Re: Advantage for White; Bayeselo (to Rémi Coulom) H.G.Muller Sun Mar 04, 2012 5:27 pm
                                          Re: Advantage for White; Bayeselo (to Rémi Coulom) Rémi Coulom Sun Mar 04, 2012 5:32 pm
      Re: Advantage for White; Bayeselo (to Rémi Coulom) Daniel Shawul Wed Sep 12, 2012 12:18 am
Post new topic    TalkChess.com Forum Index -> Computer Chess Club: Programming and Technical Discussions

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum




Powered by phpBB © 2001, 2005 phpBB Group
Enhanced with Moby Threads