Match to 7 points - Opponent 0, Player 2
| Opponent | 210 | ||
|---|---|---|---|
                          | |||
| Player | 126 | ||
| Position ID: sCMJ2BaDu8FDAA Match ID: QYnyAAAAEAAE | |||
I hesitate to offer advice in a complex backgame like this. My backgame skills speak for themselves. I put myself amid the lower ranks of advanced players, and I have the PR to prove it! With that caveat, let me point out that Blue is in bad shape here. The evaluations I posted put Blue's equity at –400 millipoints or more. The two checkers on his 1pt are badly misplaced, and cannot be used to help build an outside prime.
I agree with Keene's analysis. He identified three tactics for Blue: hitting loose on the 2pt, with a plan NOT to cover; slotting and eventually making the 9pt; and possibly hitting outside. Rolling his prime backwards should be Blue's primary plan.
White's biggest strength is the seven checkers he has in Blue's board. That means, however, that White will have trouble containing any checkers he might hit on the next turn or two. Blue should not fear getting hit too much.
At any time, White can run off his front anchor on Blue's 5pt. That means that containing White will be difficult. If White makes another anchor on Blue's 2pt, then he can run off his front anchor at his convenience. Blue should hit on the 2pt primarily hoping that White will hit back. Blue may find it useful to recycle both of the spares on his 6pt and 7pt. Hitting loose is the best way to accomplish that. If such a play also prevents White from making the 2pt, that is good as well.
After hitting on the 2pt, slotting the 9pt may be too loose. Nevertheless, 13/9 7/2* was my first reaction to this position. Hitting on the 12pt, 17/12* 6/2* is an attempt to keep White off balance while Blue tries to make another outside point. The goal of both plays is to make the 9pt.
If Blue is able to make a four-point prime, White may get squeezed. He would need to jump back out to outfield soon, or else he could be forced to weaken his inner board. To this end, hitting on the 12pt looks good, but if White jumps out quickly, the hit could backfire.
Two down, 13/9 13/8, is yet another play that attempts make the 9pt. If White hits, he may have to surrender one of his anchors. Getting sent back will help Blue's timing.
Because the top three plays all seem to be focused on making the 9pt, I tend to trust the bot evaluations of this position. I still have no idea, however, which play is best!
To improve the bot results, I would do a full rollout in XG using the standard 3-ply/XGR settings, but with a larger Search Interval. How large depends on your hardware. NeilKaz often uses Gigantic, sometimes in conjunction with 4-ply, but he does have more than one good machine for rollouts like this. I would include the top three plays identified in my evals, plus any other moves you think are interesting, such as 17/8.
Mike
P.S. Obviously there is a big difference in the b.rol results that I got and the ones obtained by sebalotek. I have not changed the design of b.rol, so, assuming that we both have that installed correctly, I am at a loss to explain the difference. I am using GnuBg 1.04.