OK what's happening here is that with all due respect 1)
Your RO is too short as the varience is too high. I still wonder why with the speed of XG and any reasonably computer, anyone posts 3 ply RO's less than 1296 games.
                        |
| ||||
| XGID=a--BBBB-----AAA-b-B-cBdcb-:0:0:1:00:0:0:3:0:10 | |||||
| Blue on roll, cube action? | |||||
| Analyzed in Rollout | No double | Double/Take |
| Player Winning Chances: | 75.373% (G:12.808% B:0.387%) | 76.003% (G:13.721% B:0.489%) |
| Opponent Winning Chances: | 24.627% (G:9.026% B:0.090%) | 23.997% (G:8.523% B:0.074%) |
| Cubeless Equities | +0.5482 | +1.1524 |
| Cubeful Equities | ||
| No double: | +0.9205 (-0.0795) | ±0.0058 (+0.9147..+0.9263) |
| Double/Take: | +1.0320 (+0.0320) | ±0.0074 (+1.0246..+1.0394) |
| Double/Pass: | +1.0000 | |
| Best Cube action: Double / Pass | ||
| Rollout details | ||
| 5184 Games rolled with Variance Reduction. Dice Seed: 46746902 Moves: 3-ply, cube decisions: XG Roller | ||
| Double Decision confidence: | 100.0% | |
| Take Decision confidence: | 100.0% | |
| Duration: 30 minutes 18 seconds | ||
eXtreme Gammon Version: 2.03
2) All current bots fail to see the priming potential and outfield control here.
When XGR++ is significantly different that XGR+, normally the RO will also continue to move in that direction.
|
| ||||
| XGID=a--BBBB-----AAA-b-B-cBdcb-:0:0:1:00:0:0:3:0:10 | |||||
| Blue on roll, cube action? | |||||
| Analyzed in XG Roller++ | No double | Double/Take |
| Player Winning Chances: | 75.103% (G:12.306% B:0.362%) | 75.064% (G:12.235% B:0.419%) |
| Opponent Winning Chances: | 24.897% (G:9.110% B:0.085%) | 24.936% (G:8.686% B:0.074%) |
| Cubeless Equities | +0.5368 | +1.0804 |
| Cubeful Equities | ||
| No double: | +0.8806 (-0.0407) | |
| Double/Take: | +0.9213 | |
| Double/Pass: | +1.0000 (+0.0787) | |
| Best Cube action: Double / Take | ||
eXtreme Gammon Version: 2.03