Only two people replied so I can't know how or if anyone else was thinking but problems like this should be approached from a fairly straightforward mathematical estimation breakdown. (if you have the tools to do so)
What we should have done here is this:
- Asked ourselves how often do we hit. (obv 20/36)
- Out of those 20 games how often do we win a gammon? This question on the surface may seem daunting but it really isn't especially in this position. Some numbers outright we can say *always win a gammon like [55 32]. For most if not all of the other numbers we can also simplify in some sense such as "if I hit and he fans we almost always win a gammon"
- If we fan, do we get cubed out immediately? (A resounding no)
- How often do we win when we dance since we don't get cubed out immediately?
That's the beginning gist of it then you toss all those into the blender and see what comes out on the other side as far as doubling/taking/passing. Looking at a position like this and going on gut feel isn't good enough.
                          |
| ||||
| XGID=-BB-CCB-aa--b-----BbacabbA:0:0:1:00:0:0:3:0:10 | |||||
| Blue on roll, cube action? | |||||
| Analyzed in Rollout | No double | Double/Take |
| Player Winning Chances: | 63.48% (G:41.53% B:2.56%) | 63.86% (G:44.48% B:3.90%) |
| Opponent Winning Chances: | 36.52% (G:9.11% B:0.12%) | 36.14% (G:9.79% B:0.18%) |
| Cubeless Equities | +0.618 | +1.323 |
| Cubeful Equities | ||
| No double: | +0.320 (-0.680) | ±0.007 (+0.313..+0.327) |
| Double/Take: | +1.079 (+0.079) | ±0.012 (+1.067..+1.092) |
| Double/Pass: | +1.000 | |
| Best Cube action: Double / Pass | ||
| Rollout details | ||
| 1296 Games rolled with Variance Reduction. Dice Seed: 24638962 Moves: 3-ply, cube decisions: XG Roller | ||
| Double Decision confidence: | 100.0% | |
| Take Decision confidence: | 100.0% | |
| Duration: 1 minute 13 seconds | ||