[Computer-go] cgos 19x19 gets interesting

David Fotland fotland at smart-games.com
Tue Oct 12 07:59:37 PDT 2010


National rating systems like the AGA uses have a more fine grained rating
system, based on tournament results.  It's not exactly ELO, because the AGA
uses 100 rating points = 1 handicap rank, and of course it would a huge
coincidence if 100 Elo happened to correspond to the win rate for one
handicap stone at all strengths.

David

> -----Original Message-----
> From: computer-go-bounces at dvandva.org [mailto:computer-go-
> bounces at dvandva.org] On Behalf Of Michael Williams
> Sent: Tuesday, October 12, 2010 4:46 AM
> To: computer-go at dvandva.org
> Subject: Re: [Computer-go] cgos 19x19 gets interesting
> 
> That's equivalent to ELO.
> 
> 
> On Tue, Oct 12, 2010 at 3:25 AM, Stefan Kaitschick
> <Stefan.Kaitschick at hamburg.de> wrote:
> >  difference in rank = number of handicap stones
> > is very convinient ofcourse.
> > An alternative way to look at this is to say:
> > if player A beats player B 70% of the time in an even game(with komi)
> player
> > A is 1 rank higher.
> > (Ranking systems make this kind of presumption. Maybe somebody on the
list
> > can supply the value KGS uses)
> > This would accommodate an almost infinite number of ranks within the
span
> of
> > the last stone towards perfection.
> >
> > Stefan
> >
> >> For amateurs, rank is defined as the number of handicap stones for an
> even
> >> game.  For pro's it's somewhat arbitrary, depending on good results in
> the
> >> rating tournaments over many games.  Of course Elo is a European chess
> >> concept and is not used for traditional games in Asia.  A 9 dan pro is
> >> perhaps two or three handicap stones stronger than a 1 dan pro, so
there
> >> are
> >> more than one pro rank per handicap stone.
> >>
> >> David
> >>
> >>> -----Original Message-----
> >>> From: computer-go-bounces at dvandva.org [mailto:computer-go-
> >>> bounces at dvandva.org] On Behalf Of Michael Williams
> >>> Sent: Monday, October 11, 2010 9:41 PM
> >>> To: computer-go at dvandva.org
> >>> Subject: Re: [Computer-go] cgos 19x19 gets interesting
> >>>
> >>> On Mon, Oct 11, 2010 at 8:37 PM, Mark Boon<tesujisoftware at gmail.com>
> >>
> >> wrote:
> >>>>
> >>>> If you accept that the number of 'dan' ranks for a handicap stone
> >>>> increases, maybe the difference in rank between god and a 9p may
> >>>> actually be very, very high as the number of ranks accounting for the
> >>>> last stone could be dozens or more.
> >>>
> >>> Clearly being able to given a stone and have an even match is not what
> >>> is used to define rank.  Is ELO used to define rank?  If not, then
> >>> isn't it a largely arbitrary measure?
> >>> _______________________________________________
> >>> Computer-go mailing list
> >>> Computer-go at dvandva.org
> >>> http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
> >>
> >> _______________________________________________
> >> Computer-go mailing list
> >> Computer-go at dvandva.org
> >> http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
> >>
> >
> > _______________________________________________
> > Computer-go mailing list
> > Computer-go at dvandva.org
> > http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
> >
> _______________________________________________
> Computer-go mailing list
> Computer-go at dvandva.org
> http://dvandva.org/cgi-bin/mailman/listinfo/computer-go




More information about the Computer-go mailing list