Thanks for your views. The take looks clear, indeed, mainly due to an open 2pt. in Blue's homeboard, which is already slotted by White. He will have some time to anchor up there, which will result in a prime vs prime battle with a positional advantage on his side (Blue has 2 dead checkers on the ace, and owns a 4-prime vs White's gapped 5-prime).
                       |
| ||||
| XGID=aBaBBBB------aA-cAbbcbB-A-:0:0:1:00:0:0:0:0:10 | |||||
| Blue on roll, cube action? | |||||
| Analyzed in Rollout | No double | Double/Take |
| Player Winning Chances: | 59,01% (G:31,78% B:1,13%) | 59,17% (G:31,47% B:1,06%) |
| Opponent Winning Chances: | 40,99% (G:10,54% B:0,55%) | 40,83% (G:10,48% B:0,52%) |
| Cubeless Equities | +0,398 | +0,798 |
| Cubeful Equities | ||
| No double: | +0,534 | ±0,010 (+0,524..+0,543) |
| Double/Take: | +0,469 (-0,064) | ±0,015 (+0,454..+0,484) |
| Double/Pass: | +1,000 (+0,466) | |
| Best Cube action: No double / Take | ||
| Percentage of wrong pass needed to make the double decision right: 10,8% | ||
| Rollout details | ||
| 2592 Games rolled with Variance Reduction. Moves: 3-ply, cube decisions: XG Roller | ||
| Double Decision confidence: | 100,0% | |
| Take Decision confidence: | 100,0% | |
| Duration: 21 minutes 14 seconds | ||
eXtreme Gammon Version: 2.19.208.pre-release
Acc. to JOH's Law we need 9 net market loosers to double. We loose the market after 22 rolls (11,22,44,55,12,13,14,15,16,52,53,54,56), followed by opp's dances, resulting in net ~15.5 market loosers (no anti-jokers). That seems like more than enough to double, and I did. But we're not doing "OK" on all other rolls. Almost whenever opp enters one checker, our lead mostly evaporates. In this particular problem, this downside is big enough to make the decision a no double.
I think a good way to look at this class of problems is seeing them as a pseudo prime v prime with opp's bar checkers virtually on the "0" point. Just like in PvP, the race doesn't matter, the timing is everything, escaping and closing the gaps are the highest priorities, etc.
Variant A - swithing Blue's gap to the ace
Compared to the original position, bottom player has been adjusted 1/2(2), and top player - 23/24. |
| ||||
| XGID=aaBBBBB------aA-cAbbcbB-A-:0:0:1:00:0:0:0:0:10 | |||||
| Blue on roll, cube action? | |||||
| Analyzed in Rollout | No double | Double/Take |
| Player Winning Chances: | 67,00% (G:36,02% B:2,17%) | 67,11% (G:35,99% B:2,16%) |
| Opponent Winning Chances: | 33,00% (G:7,59% B:0,40%) | 32,89% (G:7,65% B:0,39%) |
| Cubeless Equities | +0,642 | +1,287 |
| Cubeful Equities | ||
| No double: | +0,818 (-0,182) | ±0,007 (+0,811..+0,824) |
| Double/Take: | +1,018 (+0,018) | ±0,011 (+1,008..+1,029) |
| Double/Pass: | +1,000 | |
| Best Cube action: Double / Pass | ||
| Rollout details | ||
| 5184 Games rolled with Variance Reduction. Moves: 3-ply, cube decisions: XG Roller | ||
| Double Decision confidence: | 100,0% | |
| Take Decision confidence: | 100,0% | |
| Duration: 43 minutes 17 seconds | ||
eXtreme Gammon Version: 2.19.208.pre-release
Anchoring up on the ace doesn't give White the positional advantage anymore, since he's behind a stronger prime and Blue doesn't have any dead checkers. So even if he survives the attack, he might very well loose the resulting prime vs prime battle in the end.
Variant B - consolidating White's prime
Compared to Variant A, bottom player has been adjusted 1/2(2), 17/16, and top player - 23/24, 9/8(3). |
| ||||
| XGID=aaBBBBB------aA-AcbbcbB-A-:0:0:1:00:0:0:0:0:10 | |||||
| Blue on roll, cube action? | |||||
| Analyzed in Rollout | No double | Double/Take |
| Player Winning Chances: | 59,88% (G:29,81% B:1,88%) | 59,89% (G:30,10% B:1,88%) |
| Opponent Winning Chances: | 40,12% (G:8,98% B:0,42%) | 40,11% (G:9,11% B:0,42%) |
| Cubeless Equities | +0,420 | +0,845 |
| Cubeful Equities | ||
| No double: | +0,542 | ±0,009 (+0,532..+0,551) |
| Double/Take: | +0,513 (-0,028) | ±0,013 (+0,500..+0,527) |
| Double/Pass: | +1,000 (+0,458) | |
| Best Cube action: No double / Take | ||
| Percentage of wrong pass needed to make the double decision right: 5,5% | ||
| Rollout details | ||
| 2592 Games rolled with Variance Reduction. Moves: 3-ply, cube decisions: XG Roller | ||
| Double Decision confidence: | 100,0% | |
| Take Decision confidence: | 100,0% | |
| Duration: 21 minutes 12 seconds | ||
eXtreme Gammon Version: 2.19.208.pre-release