So I 'knew' the right play was 22, 1 not only because it's the natural play of rolling our dilly builder off to make a point but also because it screamed DMP to me. However, as I'm often wondering, I wondered if maybe it was right against a weaker player to make the anchor and stodge it up with 23, 3?
I thought if the plays were close enough that was the way I'd go. This way, the much weaker player wouldn't have a simple way of blowing me off the board with what I'd consider rather routine decisions of attacking me when he can. I had no idea the plays would be so close. Also, gammons lost (in a match) are much more expensive than what the bot's numbers would reflect. I showed this to Steve Sax and he made the correct play too of course, but saw no merit in my potential Falafel-esque 'genius' play.
                           |
| ||||
| XGID=-AaCaaC-B---bD---a-cbb-AbA:1:1:1:32:0:0:3:0:10 | |||||
| Blue to play 32 | |||||
| 1. | Rollout1 | Bar/22 3/1 | eq: -0.410 | |||
| ||||||
| 2. | Rollout1 | Bar/23 6/3 | eq: -0.416 (-0.006) | |||
| ||||||
| 3. | Rollout2 | Bar/23 13/10 | eq: -0.458 (-0.048) | |||
| ||||||
| 1 10368 Games rolled with Variance Reduction. Moves: 3-ply, cube decisions: XG Roller 2 1296 Games rolled with Variance Reduction. Moves: 3-ply, cube decisions: XG Roller | ||||||