This result should be clear from Paul's earlier money rollout. With no gammon bonus 13/9(4) is a clear winner over 13/5*(2). I haven't heard good explanations. It seems odd to strand the back checkers. The best explanation I can come up with is that the play is so robust that you have many turns to roll doubles and advance the back checkers. But I find this less than convincing.
I can't imagine anyone finding this over the board.
         
 
        
 | | White is Player 2
score: 5
pip: 151 | | 7 point match | pip: 144
score: 5
Blue is Player 1 |
|
| XGID=---B-AC-Cb--dB-B-c-fA-A---:1:1:-1:44:5:5:0:7:10 |
| Blue to play 44 |
| 1. | Rollout1 | 13/9(4) | eq: +0.277 |
| Player:
Opponent: | 63.85% (G:14.47% B:0.52%)
36.15% (G:6.07% B:0.46%) | Conf.: ± 0.005 (+0.272...+0.282) - [100.0%]
Duration: 38 minutes 19 seconds |
|
| 2. | Rollout1 | 13/5*(2) | eq: +0.224 (-0.053) |
| Player:
Opponent: | 61.20% (G:25.70% B:2.04%)
38.80% (G:8.51% B:0.76%) | Conf.: ± 0.006 (+0.218...+0.230) - [0.0%]
Duration: 37 minutes 47 seconds |
|
| 3. | Rollout1 | 13/9(2) 6/2(2) | eq: +0.199 (-0.078) |
| Player:
Opponent: | 59.94% (G:18.78% B:0.72%)
40.06% (G:9.95% B:0.66%) | Conf.: ± 0.005 (+0.194...+0.204) - [0.0%]
Duration: 35 minutes 37 seconds |
|
| |
1 1296 Games rolled with Variance Reduction.
Moves: 3-ply, cube decisions: XG Roller
|
eXtreme Gammon Version: 2.10, MET: Kazaross XG2