I misevaluated this position and still have trouble seeing that hopping out is clearly correct. Snowie got it wrong on 3 ply also so I'm in decent company I guess. Kit, Neil, & GNU all preferred hoping out. My reasoning was that even though staying on the 2 pt. left more shots, those shots left returns that were directs whereas popping out, even though it does force him to break his midpoint, brings another builder into range and I would get no direct return. (though I'd get a few indirects) I also preferred the distribution and the possibllity of a forward attack after 13/8. It puts a spare on my 8 pt. which is right where I'd like it if I get to attack those back two blots. My play clearly wins more gammons but it simply loses too many Gs and wins a small % less than bar/18.
| 115 |                       |
| 145 | |
| Board image courtesy of GO-Figure | |
| # | Ply | Move | Equity | |
| 1 | R | bar/18 | -0.477 | |
| 0.3% 6.5% 39.8% 60.2% 24.3% 0.2% | ||||
| Live cube rollout: -0.450 | ||||
| 95% confidence interval: - money cubeless eq.: -0.382 ±0.012, - live cube: -0.450 ±0.022. Rollout settings: Full rollout, 1296 games (equiv. 38044 games), played 3-ply (precise), cube 3-ply, settlement 0.550 at 4 pts, random seed, with race database. | ||||
| 2 | R | bar/23 13/8 | -0.535 (-0.058) | |
| 0.6% 8.5% 39.0% 61.0% 27.6% 0.2% | ||||
| Live cube rollout: -0.519 | ||||
| 95% confidence interval: - money cubeless eq.: -0.408 ±0.011, - live cube: -0.519 ±0.026. Rollout settings: Full rollout, 1296 games (equiv. 46952 games), played 3-ply (precise), cube 3-ply, settlement 0.550 at 4 pts, random seed, with race database. | ||||