[Bug-gnubg] Re: Beaver/Raccoon in Match play

[Timothy Y. Chow]

Hades wrote:
> Hmm ok I see. Too bad, I thought there is an option to enable that 
> feature. Can someone tell me how often the developers update particular 
> things or if you can contact them?

The developers follow this email list closely so you're doing the right 
thing already.  They update things on a continuing basis.

Christian Anthon wrote:
> Enabling beavers for matchplay probably won't happen as the rule
> change would have a substantial impact on the evaluation near the end
> of the match. That is, implementing it amounts to more than just
> enabling an option.

Yes, that's right.  Beavers in matches would *not* simply serve the 
function of "punishing" grossly incorrect doubles, as in money games.  
Beavers in matches would radically change the entire landscape of correct 
cube actions.

For example, suppose I'm leading 9 to 3 in an 11-point match, playing with 
standard rules (no beavers).  I might reach a position where it is 
perfectly correct to double.  However, if we change the rules so that 
beavers are allowed, then doubling *in that exact same position* would 
almost certainly be incorrect.  To see this, note that when I send the 
cube over when I have only 2 points to go, while my opponent has 8 points 
to go, then my opponent will automatically send the cube back when it's 
his turn to play.  He has nothing to lose by turning the cube to 4, 
because if he loses then he loses the match anyway, whether the cube is on 
2 or 4.  On the other hand, he stands to gain a lot if he wins.

The same reasoning implies that if beavers are allowed, then my opponent 
will first beaver, and *then* send the cube back at 8 when it's his turn, 
forcing us to play for the match in this one game.  So it will only be 
correct for me to double if my winning chances are so large that it's 
better to play this one game for the match than to play on with the cube 
in the center, leading 9 to 3.  This is a much more stringent criterion 
than the doubling criterion in standard match play at 2-away/8-away, when 
I only have to worry about losing the match in the current game if I get 
gammoned.

Tim