| |
BGonline.org Forums
64P-64P-54 at this score
Posted By: Nack Ballard In Response To: 64P-64P-54 at this score (neilkaz)
Date: Wednesday, 17 February 2010, at 3:40 a.m.
Original question: What is the correct play at a -13-5 score?
Bonus question 1) What is the correct play for money?
Bonus question 2) Was Opp's 64P reply correct at this score?
Let's start with Bonus question 2, the only easy one. A good reference position is 53P-64, because the choice of playing P at DMP and GS is essentially tied. (Not having recalled the exact margins, I have now looked them up and here they are for reference: d[S P4 R11] 10k and s[P R4] 15k.)
Knowing that, I am sure it is correct to make the 2pt with 64P-64 at all scores. The deeper Opp's point (and her 2pt is deeper than her 3pt), the better 64P becomes as a reply.
Bonus question 1 pertains to 64P-64P-54 for money. Let's start by examining two related reference positions:
(1) Opening 54: Averaging Snowie and Gnu rollout results, S beats D by .017, as I recall (or close enough).
(2) 64P-54: Averaging bots again, but this time I believe D beats S by a bit under .03, call it .028.
64P-64P-54 will undoubtedly be somewhere in between. Using a straight average of (1) and (2), D would win by around .005. but on the surface, at least, the position will act a lot more like (1) than (2), because the relative race is the same as (1).
One counterargument is that it is bad to split when one of Opp's main game plans is a blitz/semi-blitz (he already has a second point in his board, which is deep, and he is more willing to break his 8pt to hit because his 8pt and 2pt are six apart). However, the fact that you also have a second point in your board ameliorates the efficiency of that blitz plan.
Another counterargument is that Opp cannot split with 1s (they're blocked), so it is worse than usual to give him aces to hit on his 5pt.
I'm pretty sure the equal-race argument outweighs the two counterarguments. Does it override them by more than a ratio of 28 to 17? I'll guess yes and go with S, but without much conviction; I expect it to be very close one way or the other.
[Another factor I considered is that bringing the 4 down adds support to a potential blitz that could tend to cooperate with owning the 2pt. OTOH, it is less synergetic in terms of priming value. So, I'm choosing to discount all that.]
Finally, I'll address your original poser: 64P-64P-54 at -13-5 (13 away 5 away):
For money, I expect D to win more gammons, while I expect S (by virtue of walking into a two-point board blitz) to lose more gammons, but that the sum of those two differences will not quite add up to S's additional wins over D.
Applying that to the match score is tricky. Unless checker play effects significantly skew the money win/gammon breakdown (which I doubt, given the cube is on 1), it seems to me you just want to make the more volatile play when trailing -13-5. But which is the more volatile play? It would seem to depend on whether D adds more won gammons or S adds more lost gammons (i.e., in the normal money breakdown). I don't know which is true, so I'll guess that S barely beats D at the match score as well.
Again, I don't have a strong opinion about which play is better than the other (nor do I much care, unless the margin is more than slight); only that they are close, both for money and at the lopsided match score, and probably pretty much any match score. The only exception I see offhand is that at DMP or DMP-like scores I feel more definitely favorable towards S (based on what I said earlier about the win/gammon breakdown).
Nack
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.