Re: [Bug-gnubg] bias in the probability to win a gammon?

[Ian Shaw]

Can you tell whether Jacoby rule was enabled or not. 

You played without the cube: if Jacoby was still enabled the bot may not have 
been taking any account of gammon losses. 

This might affect the overall numbers of gammons won and lost (since they don't 
count), but I would expect it to be the same for both sides.  Unles there's 
something odd in the programming, such as the bot using it's own settings for 
itself, which respects Jacoby, but the quick finish for the human player 
doesn't use the Jacoby setting.


-- Ian Shaw 



From: address@hidden [mailto:address@hidden On Behalf Of Mrs Bibi e Bibo
Sent: 30 April 2014 21:23
To: address@hidden
Subject: [Bug-gnubg] bias in the probability to win a gammon?

Hi,
I did a simple experiment in which I let gnubg play against itself 1800 games 
in $ mode (unlimited) and without cube (actually, I played the first 20 games). 
This is an important point. The machine was set as computer against human, in 
settings>players and I had to click "end game" every single game. I noticed 
that the advantage of gnubg steadily increased.  Intrigued, I did some 
statistical analysis and found  a statistically significant difference in the 
probability of winning a gammon between gnubg and the human player.  Apparently 
gnubg has a probability nearly 3% higher. I have attached a PDF file with my 
results. I find difficult to believe that gnubg cheats but I wanted to share 
with you these results,  perhaps someone knows better. Thanks!
FS