One of the rare non obvious exceptions to 'make the dmp play'. The difference in gammons lost should feel staggering, even overwhelming to you though. The safe play it's fairly easy to see and estimate how often you lose a gammon as it's basically when you can't straight run off of it. With the bold play it's much, much harder to put a number to the estimate but even making a loosey-goosey estimation such as "we get hit nearly 70% of the time (25/36) and we lose a gammon the vast majority of those times" would be of some help.
It should be clear that if one of the inner board blots was picked up even to an awful spot hitting would be clearly correct.
         
 
       
 | | White is Player 2
score: 0
pip: 60 | Unlimited Game
Jacoby Beaver | pip: 120
score: 0
Blue is Player 1 |
|
| XGID=--BABAB-CA-A---a---bcdBbc-:1:1:1:61:0:0:3:0:10 |
| Blue to play 61 |
| 1. | Rollout1 | 11/5 8/7 | eq: -0.882 |
| Player:
Opponent: | 9.25% (G:0.10% B:0.00%)
90.75% (G:10.13% B:0.08%) | Conf.: ± 0.003 (-0.885...-0.879) - [100.0%]
Duration: 10.4 seconds |
|
| 2. | Rollout1 | 22/15* | eq: -0.971 (-0.089) |
| Player:
Opponent: | 27.49% (G:2.28% B:0.06%)
72.51% (G:58.32% B:6.76%) | Conf.: ± 0.007 (-0.978...-0.964) - [0.0%]
Duration: 14.5 seconds |
|
| |
1 1296 Games rolled with Variance Reduction.
Moves: 3-ply, cube decisions: XG Roller
|