Mary had this 5-3 to play and she couldn't stomach making the ace point and made a fine play by making the 3 pt. instead. I was undecided at the time but heeding my own advice "if something is so ugly you can't swallow it, find another play" I hope I'd have made the 3 pt. also. It's close, and the rollout isn't yet conclusive but it tells us enough that at worst the plays will be close and I believe Mary's play is correct. By making the 3 pt. you leave twos to hit and you still have the other 2 blots floating around in White's home board. You will in all likelyhood be getting attacked next turn in one form or another, so making the 3 pt. clearly loses more gammons. If you survive though, esp. when you aren't hit on your 4 pt., if you make that point you now have a broken six prime and White will have a very tough time wiggling out of there. Any small change in this position could make making the ace point correct, such as taking one of the spares off the 8 pt. (it is currently overstacked) and moving it to the midpoint.
| 130 |                           |
| 148 | |
| Board image courtesy of GO-Figure | |
| # | Ply | Move | Equity | |
| 1 | R | 8/3 6/3 | -0.311 | |
| 1.1% 15.1% 46.0% 54.0% 33.5% 1.0% | ||||
| Live cube rollout: -0.331 | ||||
| 95% confidence interval: - money cubeless eq.: -0.263 ±0.014, - live cube: -0.331 ±0.023. Rollout settings: Full rollout, 1296 games (equiv. 42809 games), played 3-ply (precise), cube 3-ply, settlement 0.550 at 16 pts, random seed, with race database. | ||||
| 2 | R | 6/1 4/1 | -0.329 (-0.018) | |
| 0.6% 14.0% 44.2% 55.8% 28.8% 0.6% | ||||
| Live cube rollout: -0.294 | ||||
| 95% confidence interval: - money cubeless eq.: -0.263 ±0.014, - live cube: -0.294 ±0.026. Rollout settings: Full rollout, 1296 games (equiv. 41599 games), played 3-ply (precise), cube 3-ply, settlement 0.550 at 16 pts, random seed, with race database. | ||||