| |
BGonline.org Forums
Jeff Ward's 'Magic Triangle'
Posted By: ah_clem In Response To: Jeff Ward's 'Magic Triangle' (Chuck Bower)
Date: Friday, 11 March 2016, at 4:06 p.m.
With two checkers in the homeboard, the math simply how many roll bear both off for each of the 36 possible positions. It's binary - either you bear off both next roll or you don't.
If you consider positions where one of the checkers is outside the homeboard, then only doubles bear off in one roll and you may be choosing the position that makes it most likely to bear off in two rolls (or more if both are outside the homeboard). That's a much more difficult problem.
Alternatively, bearing in both checkers may be better than having a distance closest to e as the following counter-example shows. Here, a distance or 0 or 6 is better than a distance of 2 or 4.
White is Player 2
score: 0
pip: 41 point match pip: 17
score: 0
Blue is Player 1XGID=--------AA--------------d-:0:0:1:32:0:0:0:1:10 Blue to play 32
1. Rollout1 8/3 eq: -0.815
Player:
Opponent:9.26% (G:0.00% B:0.00%)
90.74% (G:0.00% B:0.00%)Conf.: ± 0.000 (-0.815...-0.815) - [100.0%]
Duration: 0.1 second2. Rollout1 9/6 8/6 eq: -0.815
Player:
Opponent:9.26% (G:0.00% B:0.00%)
90.74% (G:0.00% B:0.00%)Conf.: ± 0.000 (-0.815...-0.815) - [0.0%]
Duration: 0.6 second3. Rollout1 9/7 8/5 eq: -0.861 (-0.046)
Player:
Opponent:6.94% (G:0.00% B:0.00%)
93.06% (G:0.00% B:0.00%)Conf.: ± 0.000 (-0.861...-0.861) - [0.0%]
Duration: 0.2 second4. Rollout1 9/4 eq: -0.861 (-0.046)
Player:
Opponent:6.94% (G:0.00% B:0.00%)
93.06% (G:0.00% B:0.00%)Conf.: ± 0.000 (-0.861...-0.861) - [0.0%]
Duration: 0.1 second1 1296 Games rolled with Variance Reduction.
Moves: 3-ply, cube decisions: XG RollereXtreme Gammon Version: 2.10, MET: Kazaross XG2
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.