bug-gnubg
[Top][All Lists]

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

 From: Ian Shaw Subject: RE: [Bug-gnubg] Snowie 4 vs. GNU 0.13 Date: Mon, 9 Jun 2003 13:10:24 +0100

```
> -----Original Message-----
> From: Øystein Johansen
> Sent: 09 June 2003 12:47
> To: Nardy Pillards; Joern Thyssen
> Cc: GnuBg Bug (E-mail)
> Subject: Re: [Bug-gnubg] Snowie 4 vs. GNU 0.13
>
>
> std.dev = sqrt( n * p * q )
>
> p = Estimated probability of GNU winning based on the trials.
> q = 1 - p
> n = number of trials.
>
> std.dev = sqrt( 100 * 0.56 * 0.44) = 4.964
>
> -Øystein
>
I did it the long way in Excel, making a column of 56 1s & 44 0s. I came up
with 0.4964,  the same value as Øystein but for a power of 10. Excel's built in
formula made it 0.4989. Are you sure you should have included n = 100 in there?

Being rather ignorant of stats, I also wondered if you should be scoring +1 and
-1. This would give a mean of 0.12, in a ppg kind of way. The Std Dev is about
0.99 if you do it this way. I suppose it's the difference between equity and
winning chances.

--Ian

```

reply via email to