I've originally evaluated it as being close to a cash. Turns out it's not even a redouble nor an initial double.
                       |
| ||||
| XGID=aAB-ABB-BA---Ba-abbdAbb-A-:1:1:1:00:0:0:0:0:10 | |||||
| Blue on roll, cube action? | |||||
| Analyzed in Rollout | No redouble | Redouble/Take |
| Player Winning Chances: | 57,63% (G:31,84% B:1,45%) | 58,15% (G:32,72% B:1,31%) |
| Opponent Winning Chances: | 42,37% (G:15,88% B:1,24%) | 41,85% (G:15,90% B:1,36%) |
| Cubeless Equities | +0,314 | +0,661 |
| Cubeful Equities | ||
| No redouble: | +0,517 | ±0,016 (+0,501..+0,532) |
| Redouble/Take: | +0,393 (-0,124) | ±0,024 (+0,369..+0,416) |
| Redouble/Pass: | +1,000 (+0,483) | |
| Best Cube action: No redouble / Take | ||
| Percentage of wrong pass needed to make the double decision right: 16,9% | ||
| Rollout details | ||
| 648 Games rolled with Variance Reduction. Moves: 3-ply, cube decisions: XG Roller | ||
| Double Decision confidence: | 100,0% | |
| Take Decision confidence: | 100,0% | |
| Duration: 6 minutes 39 seconds | ||
eXtreme Gammon Version: 2.19.208.pre-release
| Analyzed in Rollout | No double | Double/Take |
| Player Winning Chances: | 57,59% (G:31,71% B:1,38%) | 58,04% (G:32,72% B:1,33%) |
| Opponent Winning Chances: | 42,41% (G:15,75% B:1,20%) | 41,96% (G:15,94% B:1,36%) |
| Cubeless Equities | +0,313 | +0,657 |
| Cubeful Equities | ||
| No double: | +0,431 | ±0,012 (+0,419..+0,442) |
| Double/Take: | +0,388 (-0,043) | ±0,016 (+0,372..+0,404) |
| Double/Pass: | +1,000 (+0,569) | |
| Best Cube action: No double / Take | ||
| Percentage of wrong pass needed to make the double decision right: 6,5% | ||
| 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: 10 minutes 37 seconds | ||
eXtreme Gammon Version: 2.19.208.pre-release
One way of looking at such dynamic position is to look at some close, more stable variant. Let's move our 4pt blot to 1pt, making a clean 4pt board.
|
| ||||
| XGID=aBB--BB-BA---Ba-abbdAbb-A-:1:1:1:00:0:0:0:0:10 | |||||
| Blue on roll, cube action? | |||||
We threaten to put second guy up vs our stronger, 4pt board. Even if we don't hit, opp will still dance 56% of the time. Obviously a very strong double. In fact, it's already a pass.
| Analyzed in Rollout | No redouble | Redouble/Take |
| Player Winning Chances: | 66,24% (G:40,17% B:1,62%) | 66,62% (G:41,73% B:1,58%) |
| Opponent Winning Chances: | 33,76% (G:9,89% B:0,81%) | 33,38% (G:10,37% B:0,95%) |
| Cubeless Equities | +0,636 | +1,305 |
| Cubeful Equities | ||
| No redouble: | +0,821 (-0,179) | ±0,010 (+0,811..+0,830) |
| Redouble/Take: | +1,101 (+0,101) | ±0,019 (+1,081..+1,120) |
| Redouble/Pass: | +1,000 | |
| Best Cube action: Redouble / Pass | ||
| Rollout details | ||
| 648 Games rolled with Variance Reduction. Moves: 3-ply, cube decisions: XG Roller | ||
| Double Decision confidence: | 100,0% | |
| Take Decision confidence: | 100,0% | |
| Duration: 4 minutes 17 seconds | ||
eXtreme Gammon Version: 2.19.208.pre-release
Knowing that, we might evaluate the original position as being 1 sound step behind it, since our board is still a mess, so should result in an easy take, and maybe a small to medium-sized double (at least I'd do that, but we already know it's off too, since it's not even a double). For better viewing, let me repost the original position, followed by a complete table of our market loosing sequences (per XGR++ evaluation):
|
| ||||
| XGID=aAB-ABB-BA---Ba-abbdAbb-A-:1:1:1:00:0:0:0:0:10 | |||||
| Blue on roll, cube action? | |||||
eXtreme Gammon Version: 2.19.208.pre-release
| play | rolls | no. of rolls (A) | opp's response | response coefficient (B) | A x B |
|---|---|---|---|---|---|
| make 4pt board | 11, 21, 31, 41, 32, 52, 33, 53 | 14 | dance | 16/36 | 6.2 |
| make 4pt board, 1 slot | 51, 22 | 3 | dance | 16/36 | 1.3 |
| make 5pt board, 1 slot | 55 | 1 | dance | 25/36 | 0.7 |
| hit 1, make 5pt board | 44 | 1 | anything but 33 joker | 35/36 | 1.0 |
| hit 1, make 4pt board | 43 | 2 | dance or enter 1 | 32/36 | 1.8 |
| hit 1, make 4pt board | 63 | 2 | dance or enter 1 on 3pt | 24/36 | 1.3 |
| hit 1, make 4pt board, 1 slot | 54, 64, 66 | 5 | dance or enter 1 (miss) | 24/36 | 3.3 |
| hit 1, make 4pt board, 1 slot | 65 | 2 | dance | 16/36 | 0.9 |
| hit 2, keep 3pt board, 2 slots | 42 | 2 | dance or enter 1 (miss) | 15/36 | 0.8 |
| 32 | 17.3 |
17.3 "rolls" is quite a big number. Only after 4 rolls (61, 62), played acc. to XGR++, we can't loose our market (although technically, we could choose a more bold plays with these rolls than XGR++ does, which would result in market loss when opp dances).
Trying to apply O'Hagan's Law here would be quite misleading. You got 17 market losers and no anti-jokers.
So why isn't it a double?
I think the answer is hidden in the quality of opp's possible responses, elevated by his strong frontal structure, dreadfully shadowing our back checker(s). We both hit and clean homeboard slots with only 5 rolls (43, 44, 63). Most of the time though we either hit and leave a blot in our home, or make some points but don't hit. He can then hit our unsecured blot from the bar or he can enter and lift something after our misses. These responses puts him into equity range of avg +0.600 (after we build but don't hit) or avg. -0.060 (after we hit but leave slots).