Reading this article I found this to be an interesting problem. Glancing at the problem I hit, studying it, since it was an article, I wasn't sure but in the end decided I'd still hit. The rollout is right on the cusp of stat. sig. but close enough for me.
Everything everyone has noted is true. Blue has the stronger board, White has a blot on the ace point if you are hit back. What wasn't mentioned is Blue's difficulty in winning the position if he is hit back compared to how easily he should win it if he plays the safe & sound bar/21 13/10. White's made 5 pt. and Blue's complete lack of outfield control (or points) means if Blue is hit back he'll have one hell of a time bringing this game home. For more analysis check out Trice's article, I'm not in the mood to write, just posting it because I realized I forgot to post it!
| 134 |                          |
| 99 | |
| Board image courtesy of GO-Figure | |
| # | Ply | Move | Equity | |
| 1 | R | bar/21 13/10 | 0.492 | |
| 0.3% 28.1% 69.0% 31.0% 5.6% 0.2% | ||||
| Live cube rollout: 0.484 | ||||
| 95% confidence interval: - money cubeless eq.: 0.606 ±0.008, - live cube: 0.484 ±0.011. Rollout settings: Full rollout, 1944 games (equiv. 77695 games), played 3-ply (precise), cube 3-ply, settlement 0.550 at 16 pts, random seed, with race database. | ||||
| 2 | R | bar/22 6/2* | 0.472 (-0.020) | |
| 0.3% 32.3% 67.5% 32.5% 8.0% 0.4% | ||||
| Live cube rollout: 0.466 | ||||
| 95% confidence interval: - money cubeless eq.: 0.594 ±0.009, - live cube: 0.466 ±0.013. Rollout settings: Full rollout, 1944 games (equiv. 75862 games), played 3-ply (precise), cube 3-ply, settlement 0.550 at 16 pts, random seed, with race database. | ||||