| |
BGonline.org Forums
Rollout: 4-2 to play
Posted By: Daniel Murphy In Response To: Rollout: 4-2 to play (Timothy Chow)
Date: Monday, 3 May 2010, at 2:36 a.m.
Tim, I think you rightly point that the spare on the 9 point is well placed, but it's also true that an additional spare on the 13 point is not well placed, and 15/13 is weak, accomplishing nothing besides clearing the blot. That 15 point checker must be safetied, but it would really also like to make a crossover or do something useful.
If Blue were ahead in the race (by f.ex. moving the deep checkers back, as you describe), 15/9 would in my mind be the "next!" play. Since Blue is behind, 13/9 13/11 to block seems better. I looked at the three top 4-2 plays, followed by an average White number, and Blue to play 6-5. Blue's checkers do look better placed if Blue has played 15/11 13/11 ... White 5-3 ... Blue 6-5, but the bot says the equity difference is small. All three of the 4-3 5-3 6-5 sequences make the game about 50/50. I think the big swing from blocking White's 4-4 for just a roll (since Blue is behind) is probably a more important reason for preferring 15/11 13/11 over the other plays. Of the second and third place plays, I can't really see any reason why 15/13 9/5 should be better than 15/9. Maybe it is, but your rollout has a difference between those plays of only 0.006 and confidence intervals of SDs of ~0.006, too. You correctly point out that White is very unlikely to break the 18 point blotting any time soon, so there's no need to make the 5 point now.
In the third position (the third you discuss, that is, the second diagram) Blue leads by 19 pips after the roll. I'd play 13/8 fast, surely not 8/5 6/4 -- as in the first position, the spare on the 13 point has nothing better to do than move somewhere else. White won't be volunteering a blot from the 3 checkers on his 18 point, so there's no reason to be worried about making a board behind that anchor -- Blue should just bring the checkers home. I prefer 13/8 by more than the rollout's 0.003 (noting that the CIs are ~0.006). I imagine many sequences tranpose, and guess that there must be very few sequences where having a spare to move from the 10 point, leaving the point made, is better than simply clearing the point (or not being able to clear the point).
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.