As I already mentioned elsewhere in this thread, I spent a lot of time on this play OtB. The rollout is still far from stat. sig. and in watching it over its 10k trials it has been like a roller coaster ride of what's best between the top two plays. (I believe 6/2 was in the running in the early stages but don't quote me on that)
I hated 6/2 and dismissed it rather quickly. Putting a checker out somewhat out of play when there are other reasonable alternatives is never my preference. So I'm down to coming out or slotting it up. What decided it for me are one of the things you'll never seen in the bot rollouts, avoiding plays where you can get wiped off your board and any rich dummy opponent will know how to do it. Holding the anchor though and duplicating aces in itself is a good play and creates an entire other issue for your opponent. They have to make plays now, if they do wind up escaping they have to figure out when to cube. In a short match even against a good opponent I will shy away from a play where the gammon loses significantly spike if I think the plays are close. The bot gammon values don't convert to the human v. human gammon values.
           
    
         
 |  is Ray
score: 1
pip: 140 | | 9 point match | pip: 149
score: 0
 is Stick |
|
| XGID=---a-CDaC---aCaabb-d-bA--A:0:0:1:43:0:1:0:9:10 |
| to play 43 |
| 1. | Rollout1 | Bar/18 | eq: -0.259 |
| Player:
Opponent: | 43.59% (G:17.52% B:3.29%)
56.41% (G:24.11% B:1.07%) | Conf: ± 0.008 (-0.267...-0.251)
Duration: 2 hours 40 minutes |
|
| 2. | Rollout1 | Bar/22 8/4 | eq: -0.265 (-0.006) |
| Player:
Opponent: | 42.10% (G:12.96% B:0.69%)
57.90% (G:12.33% B:0.77%) | Conf: ± 0.008 (-0.273...-0.257)
Duration: 3 hours 18 minutes |
|
| 3. | Rollout1 | Bar/22 6/2 | eq: -0.267 (-0.008) |
| Player:
Opponent: | 42.12% (G:12.83% B:0.64%)
57.88% (G:11.88% B:0.69%) | Conf: ± 0.007 (-0.274...-0.260)
Duration: 2 hours 58 minutes |
|
| |
1 10368 Games rolled with Variance Reduction.
Dice Seed: 36438452
Moves and cube decisions: 3 ply
|