Re: [Bug-gnubg] small suggestion and questions

[Jonathan Kinsey]

Christian Anthon wrote:
> 
>     Perhaps another approach is to ignore "easy" chequer plays, i.e. similar
>     to cube decisions, only count the moves that can potentially be
>     misplayed (perhaps were the worst move is more than a certain equity
>     from the best move)?  This would automatically include forced plays.
> 
> 
> 
> The problem with this approach is that it is difficult to say which moves
> are hard and wich are easy. A middle-game move might be hard even
> though the difference between the top-two plays are 0.1 and and
> end-game play may be easy even though the difference is only 0.001.

You can classify some moves as easy - e.g. assume a move has 4 choices
and the difference between best and worst equity is 0.002, clearly this
move can't be miss-played.

There might be quite a few moves with very little equity to lose (in a
race for example) - this currently 'inflates' you're overall rating.

I'm not sure this is a good idea (then I'm not sure
isCloseCubedecision() is a good idea either).

Jon

signature.asc
Description: OpenPGP digital signature