Re: [Bug-gnubg] Snowie 4 vs. GNU 0.13

[Joern Thyssen]

On Mon, Jun 09, 2003 at 04:11:15PM +0100, Ian Shaw wrote

> Perhaps the numbers are right and it's out interprettion that's wrong.
> 
> The numbers say that the average reault is 0.56 in favour of GnuBg,
> +-0.98 95%CI, assuming a normal distribution. It's certainly true that
> the results fitted in the range 1.54 to -0.42, but the distribution
> wasn't a normal one.
> 
> I think Rod is perhaps right, and we need stats from several series of
> matches.
> 
> OTOH, maybe not. Check out these threads from rec.games.backgammon:
> 
> http://www.bkgm.com/rgb/rgb.cgi?view+709 

Aaah, it seems I'm missing a factor of sqrt(n) in my formulae.

It should be

0.56 +/- 0.98/sqrt(n)

instead.

That is,

0.56 +/- 0.1

The luck adjusted result is

0.500 +/- 0.014

Jørn