Re: [Bug-gnubg] new contact net

[Joern Thyssen]

On Tue, Jun 10, 2003 at 09:25:59AM +0100, Ian Shaw wrote
> 
> 
> > -----Original Message-----
> > From: Joseph Heled
> > 
> > I am updating the development version for a new contact net 
> > (version 0.14)
> > 
> > The net is here -
> > http://pages.quicksilver.net.nz/pepe/ngb/gnubg.weights.14.gz
> > 
> > For 500000 7pt matches, the new net is 50.59% - 49.41% against the 
> > current one.
> > 
> > -Joseph
> > 
> That's a lot of matches. Using our hard-won knowledge of yesterday, I
> calculate the SD as 0.0001%, so the 99%CI is about 0.0003%. (I'd say
> that's significant enough not to bother with the luck analysis, Jorn.)

Hmm, I get a std.dev of 0.1% ( 0.5/sqrt(500,000) ), and a 95% CI of +/-0.2%. 

Anyway, I've suggested to Joseph it might be worthwhile to do the luck
analysis, since this will reduce the number of matches needed. 

In the 100-match sample I analysed the std.deviation was reduced with a
factor of 4. I estimate that the luck analysis will double the time need
to play a match, so this will save a factor of two. Alternatively, it
may be possible to do 1,000,000 matches for the same computational
effort.

Jørn