Re: [Bug-gnubg] Current categorization of doubling mistakes (new thread)

[Joern Thyssen]

On Tue, Oct 07, 2003 at 08:26:20PM +0200, Jim Segrave wrote
> 
> This leads to a new question on what must be becoming Joern's least
> favourite subject...
> 
> I understand the idea of early/late doubling (I don't apply it well,
> but I do understand it). Now, when gnubg analyses a game where I was 
> creeping up on an optimal double (or never saw it was possible to
> double or whatever), how does gnubg decide if not doubling was an
> error or correct. Let's say 
> 
> DP = 75% CP = 81% TG = 84%
> 
> If I'm at 77% and don't double, is that an error of say .030? I assume
> gnubg isn't examining market losers to decide if you should postpone
> doubling? 

Well, it depends on your evaluation settings.

For all plies gnubg is comparing the cubeful equity of doubling now with
the cubeful equity of not doubling.

If you use 0-ply gnubg can obviously not detect market losers, so it
doubles somewhere in the market window. Precisely where it doubles
depend on the match score and position class (race/contact/bearoff).

For 1-ply gnubg will indirectly see your market losers, but it'll not
see your opponents market gainers. At 2-ply it'll see market losing
sequences, and you'll effectively be comparing the cubeful equity of
doubing now versus doubling on the next roll.

Jørn

pgpelkqRI__7E.pgp
Description: PGP signature