As Stick so graciously suggested, it is a double by .1 before the roll (given the score).
But I'm surprised by the differing best plays when the cube is centered and when it is turned.
First, centered (the original position):
           
 
           
 | | White is Player 2
score: 4
pip: 141 | | 7 point match | pip: 122
score: 3
Blue is Player 1 |
|
| XGID=-BaA--CCB---cBb--cBdaa----:0:0:1:54:3:4:0:7:10 |
| Blue to play 54 |
| 1. | Rollout1 | 13/8 7/3 | eq: +0.833 |
| Player:
Opponent: | 59.92% (G:21.96% B:0.25%)
40.08% (G:9.02% B:0.42%) | Conf.: ± 0.051 (+0.783...+0.884) - [100.0%]
Duration: 1 minute 44 seconds |
|
| 2. | Rollout1 | 7/3 7/2* | eq: +0.688 (-0.145) |
| Player:
Opponent: | 60.03% (G:24.05% B:0.35%)
39.97% (G:9.41% B:1.65%) | Conf.: ± 0.032 (+0.656...+0.720) - [0.0%]
Duration: 1 minute 45 seconds |
|
| 3. | Rollout1 | 7/2* 6/2 | eq: +0.552 (-0.282) |
| Player:
Opponent: | 57.41% (G:22.08% B:0.32%)
42.59% (G:10.53% B:0.68%) | Conf.: ± 0.037 (+0.515...+0.588) - [0.0%]
Duration: 1 minute 41 seconds |
|
| 4. | 2-ply | 18/13 7/3 | eq: +0.643 (-0.191) |
| Player:
Opponent: | 57.46% (G:21.27% B:0.38%)
42.54% (G:9.85% B:0.32%) | |
|
| 5. | 1-ply | 8/3 6/2* | eq: +0.488 (-0.346) |
| Player:
Opponent: | 56.47% (G:20.15% B:0.29%)
43.53% (G:8.30% B:0.42%) | |
|
| |
1 108 Games rolled with Variance Reduction.
Moves: 3-ply, cube decisions: XG Roller
|
eXtreme Gammon Version: 2.01, MET: Kazaross XG2
XG preserves structure with 13/8 7/3.
But if I had cubed:
| | White is Player 2
score: 4
pip: 141 | | 7 point match | pip: 122
score: 3
Blue is Player 1 |
|
| XGID=-BaA--CCB---cBb--cBdaa----:1:-1:1:54:3:4:0:7:10 |
| Blue to play 54 |
| 1. | Rollout1 | 7/3 7/2* | eq: +0.514 |
| Player:
Opponent: | 59.95% (G:24.85% B:0.27%)
40.05% (G:8.86% B:2.18%) | Conf.: ± 0.022 (+0.492...+0.536) - [99.9%]
Duration: 2 minutes 45 seconds |
|
| 2. | Rollout1 | 13/8 7/3 | eq: +0.463 (-0.050) |
| Player:
Opponent: | 59.88% (G:22.27% B:0.14%)
40.12% (G:9.19% B:0.31%) | Conf.: ± 0.024 (+0.440...+0.487) - [0.1%]
Duration: 2 minutes 24 seconds |
|
| 3. | Rollout1 | 7/2* 6/2 | eq: +0.379 (-0.135) |
| Player:
Opponent: | 57.58% (G:21.22% B:0.46%)
42.42% (G:9.68% B:0.56%) | Conf.: ± 0.023 (+0.356...+0.402) - [0.0%]
Duration: 2 minutes 44 seconds |
|
| 4. | 3-ply | 8/3 6/2* | eq: +0.331 (-0.183) |
| Player:
Opponent: | 57.15% (G:20.29% B:0.23%)
42.85% (G:8.58% B:0.38%) | |
|
| 5. | 3-ply | 18/13 7/3 | eq: +0.330 (-0.184) |
| Player:
Opponent: | 57.85% (G:19.64% B:0.32%)
42.15% (G:9.92% B:0.35%) | |
|
| |
1 108 Games rolled with Variance Reduction.
Moves: 3-ply, cube decisions: XG Roller
|
eXtreme Gammon Version: 2.01, MET: Kazaross XG2
Now, attack with 7/3 7/2. Why the change? Both plays leave a direct shot, so I don't understand why the first play leads to more efficient cubes.
Also, why is 7/3 7/2, leaving an extra blot on the bar, so much better than 7/2 6/2 (regardless of cube position)?