All rating model assume tha If A cab B B easily and B can beat C easily then A can beat C easily as well. As we all know this is ususally true but no always. I know I cannot handle AyaMC in KGS although rating of KgS indicate that I should be favorite. I assume that these sort wrong skill set against C problems could be a big part of explanation. Petri
2009/10/9 Brian Sheppard <[email protected]> > >One must be very careful about proclaiming wild transitivity issues. I'm > >not saying it's not an issue, there is some going on with every program on > >CGOS, but with less than 500 games between any two players you are going > >to get error margins of +/- 30-50 ELO or something like that. > > Actually we are certain that significant differences are being observed. If > we pool the Pachi and Pebbles data, then the null hypothesis is that > Valkyria defeats both programs by 79%. The observed data differs by at > least > 3.5 standard deviations. > > Note that we are talking about 150 rating points. > > > > > _______________________________________________ > computer-go mailing list > [email protected] > http://www.computer-go.org/mailman/listinfo/computer-go/ > -- Petri Pitkänen e-mail: [email protected]
_______________________________________________ computer-go mailing list [email protected] http://www.computer-go.org/mailman/listinfo/computer-go/
