HOMINID references a PDF file of the BGO OLM match 3 from January 2011, and mentions two positions.
Cube action (PDF p.21). HOMINID's comment:
I love it what a great idea, expert players actually giving opinions and thoughts on best plays. I page through and arrive at a cube situation. I fancy myself on cube decisions, let's see what they have to say. A unanimous decision from all twelve, double take. I guess that at their level it just needs looking at no big deal but here's the thing, not one not one proffered a game winning % chance. Whichever way you look at it, it's always good to know what they are in that they are pivotal to cube decisions. Let's do it. 20 seconds later I arrived at 72%. Let's whack it into xg. I should not get that close but on occasions I do. The difference here is that xg does not seem to include a value for Gammons in its game winning chances, I do. The frustrating point is that we are led to believe that such players are masters of game winning chances mwc and Windows of opportunity etc when it comes to the cube.
Comment: The match participants are not all experts. Any BGO reader can participate in the on-line match. I think most OLM participants don't usually give GWC estimates on cube decisions. I don't know how often they make them. Cube decisions don't always need very precise estimates. But it's certainly good practice to make them. I don't understand the comment that "xg does not seem to include a value for Gammons in its game winning chances, I do."
OLM Match 3, Game 1, Score 0-0/15
| XG | 109 | ||
|---|---|---|---|
                           | |||
| BGO | 119 | ||
| Position ID: 1D0HAxCzmYMpAA Match ID: MAHgAQAAAAAE | |||
• XG doubles
Jason's 11 January 2011 rollout:
| Analyzed in Rollout | |
| No Double | |
| Player Winning Chances: | 69.64% (G: 13.24% B: 0.30%) |
| Opponent Winning Chances: | 30.36% (G: 5.32% B: 0.19%) |
| Double/Take | |
| Player Winning Chances: | 69.63% (G: 13.64% B: 0.31%) |
| Opponent Winning Chances: | 30.37% (G: 5.19% B: 0.19%) |
| Cubeless Equities | |
| No Double: | +0.475 |
| Double: | +0.975 |
| Cubeful Equities | |
| No Double: | +0.733 (-0.050) |
| Double/Take: | +0.783 |
| Double/Drop: | +1.000 (+0.217) |
| Best Cube action: Double / Take | |
| Rollout details | |
| 7776 Games rolled with Variance Reduction. Moves and cube decisions: 3 ply | |
| Confidence No Double: | ± 0.006 (+0.727...+0.739) |
| Confidence Double: | ± 0.007 (+0.776...+0.790) |
| Double Decision confidence: | 100.0% |
| Take Decision confidence: | 100.0% |
| Duration: 1 hour 55 minutes | |
eXtreme Gammon Version: 1.21, MET: Rockwell-Kazaross
Bear-in (PDF p.27). See the PDF for much discussion. HOMINID's comment:
The next interesting point for me is bearing in which of course is virtually the same as bearing off and there's a rule of thumb. Opinions are divided three choices, the truth is all these choices may be marginal but there is one best play . A 2 x 2 to play an instant decision. 12-6 10-8 and xg is in agreement because it gives you the most opportunities of getting your boys home. I found it strange that credence was given to optimising what was deemed a bad roll at the expense of making the best play.
Comment: I don't understand the last sentence above. In any case, the gist of the discussion at the time was that the BGO team was a favorite to get off the gammon, should thus cater to bad rolls, and 12/6 10/8 and 10/6 11/7 were probably close plays. The vote was 5 for 11/7 10/6, 4 for 12/6 10/8, 2 for 12/6 11/9. In my rollout below, 10/6 11/7 is just slightly better than 12/6 10/8. Jason's 13 January 2011 XG rollout agreed; it's here: http://www.bgonline.org/forums/webbbs_config.pl?noframes;read=86561
OLM Match 3, Game 1, Score 0-0/15
| XG | 16 | ||
|---|---|---|---|
      | |||
| BGO | 87 | ||
| Position ID: bgEAAO32UAEAAA Match ID: UQnpAQAAAAAE | |||
| # | Ply | Move | Equity | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | R | 11/7 10/6 | -1.1533 | |||||||||||||||||||
| ||||||||||||||||||||||
| Full cubeful rollout with var.redn. | ||||||||||||||||||||||
| 2592 games, Mersenne Twister dice gen. with seed 722728712 and quasi-random dice | ||||||||||||||||||||||
| Play: supremo 2-ply cubeful prune [world class] | ||||||||||||||||||||||
| keep the first 0 0-ply moves and up to 16 more moves within equity 0.32 | ||||||||||||||||||||||
| Skip pruning for 1-ply moves. | ||||||||||||||||||||||
| Cube: 2-ply cubeful prune [world class] | ||||||||||||||||||||||
| • | 2 | R | 12/6 10/8 | -1.1540 ( -0.0007) | ||||||||||||||||||
| ||||||||||||||||||||||
| Full cubeful rollout with var.redn. | ||||||||||||||||||||||
| 2592 games, Mersenne Twister dice gen. with seed 722728712 and quasi-random dice | ||||||||||||||||||||||
| Play: supremo 2-ply cubeful prune [world class] | ||||||||||||||||||||||
| keep the first 0 0-ply moves and up to 16 more moves within equity 0.32 | ||||||||||||||||||||||
| Skip pruning for 1-ply moves. | ||||||||||||||||||||||
| Cube: 2-ply cubeful prune [world class] | ||||||||||||||||||||||
| 3 | R | 12/6 11/9 | -1.1640 ( -0.0107) | |||||||||||||||||||
| ||||||||||||||||||||||
| Full cubeful rollout with var.redn. | ||||||||||||||||||||||
| 2592 games, Mersenne Twister dice gen. with seed 722728712 and quasi-random dice | ||||||||||||||||||||||
| Play: supremo 2-ply cubeful prune [world class] | ||||||||||||||||||||||
| keep the first 0 0-ply moves and up to 16 more moves within equity 0.32 | ||||||||||||||||||||||
| Skip pruning for 1-ply moves. | ||||||||||||||||||||||
| Cube: 2-ply cubeful prune [world class] | ||||||||||||||||||||||
| 4 | R | 11/9 10/6 3/1 | -1.1642 ( -0.0109) | |||||||||||||||||||
| ||||||||||||||||||||||
| Full cubeful rollout with var.redn. | ||||||||||||||||||||||
| 2592 games, Mersenne Twister dice gen. with seed 722728712 and quasi-random dice | ||||||||||||||||||||||
| Play: supremo 2-ply cubeful prune [world class] | ||||||||||||||||||||||
| keep the first 0 0-ply moves and up to 16 more moves within equity 0.32 | ||||||||||||||||||||||
| Skip pruning for 1-ply moves. | ||||||||||||||||||||||
| Cube: 2-ply cubeful prune [world class] | ||||||||||||||||||||||
| 5 | R | 12/6 3/1 | -1.1707 ( -0.0174) | |||||||||||||||||||
| ||||||||||||||||||||||
| Full cubeful rollout with var.redn. | ||||||||||||||||||||||
| 791 games, Mersenne Twister dice gen. with seed 722728712 and quasi-random dice | ||||||||||||||||||||||
| Play: supremo 2-ply cubeful prune [world class] | ||||||||||||||||||||||
| keep the first 0 0-ply moves and up to 16 more moves within equity 0.32 | ||||||||||||||||||||||
| Skip pruning for 1-ply moves. | ||||||||||||||||||||||
| Cube: 2-ply cubeful prune [world class] | ||||||||||||||||||||||