In my analysis above I assumed that white would re-whip with any of the rolls that expose a double-direct shot. Looking at the cube action after one of the "bad rolls" it's only a rewhip if blue's blot is on the 16 - it's not a rewhip if the blot is on the 17 or 18. And the only times that happens is when blue shakes a 31 or a 34.
I don't understand why it's not a recube with the checker on the 17 or 18. White is more than 50% GWC and is looks like a last roll situation to me. I suppose that blue still wins a few after being hit, and white is only a tad above 50% with the blot on the 17 or 18. When missed, blue only has one bad roll so the re-re-whip seems correct.
Am I missing something here?
                        |
| ||||
| XGID=-IBC-ab-A---------abbcbb--:3:-1:-1:00:0:0:3:0:10 | |||||
| Blue on roll, cube action? | |||||
| Analyzed in Rollout | No redouble | Redouble/Take |
| Player Winning Chances: | 52.14% (G:2.51% B:0.05%) | 54.27% (G:2.78% B:0.07%) |
| Opponent Winning Chances: | 47.86% (G:5.50% B:0.00%) | 45.73% (G:5.49% B:0.00%) |
| Cubeless Equities | +0.013 | +0.118 |
| Cubeful Equities | ||
| No redouble: | +0.128 | ±0.002 (+0.127..+0.130) |
| Redouble/Take: | +0.084 (-0.044) | ±0.005 (+0.079..+0.089) |
| Redouble/Pass: | +1.000 (+0.872) | |
| Best Cube action: No redouble / Take | ||
| Percentage of wrong pass needed to make the double decision right: 4.6% | ||
| Rollout details | ||
| 1296 Games rolled with Variance Reduction. Moves: 3-ply, cube decisions: XG Roller | ||
| Double Decision confidence: | 100.0% | |
| Take Decision confidence: | 100.0% | |
| Duration: 4.6 seconds | ||
eXtreme Gammon Version: 2.10