|
BGonline.org Forums
OLM Mo 01/02/12
Posted By: Chuck Bower In Response To: OLM Mo 01/02/12 (Jason Lee)
Date: Tuesday, 3 January 2012, at 6:51 p.m.
Here's what I see (and I did use pencil+paper AND looked up the match equities which I don't have memorized for this length match):
A) Assuming half of our cubeless wins are G's and 20% of our cubeless losses are G's, opp needs around 17% cubeless GWC to take this.
B) I see three common scenarios for White: 1) enter and hit; 2) anchor up; 3) get treed until after we start bearing off. Case 1 (the most infrequent of these) is clearly White's best as he will likely be leading the game. Case 2 gives White (ballpark) enough GWC to take. Case 3 gives White no more than a few % chances to win. Let's say case 1 occurs 10% of the time and the other two are equally divided 45-45. Assigning some cubeless GWC's, 10%*(2/3) + 45%*(1/5) + 45%*(1/22.5) ~ 18% or a bit over the drop/take line (i.e. a take).
C) This is close enough that it must be a redouble with the position's volatility.
D) Did I mis-represent anything? Probably, and maybe several things. :) But the most obvious to me is the overoptimistic (for White) assumption that anchoring justifies White's take. A high anchor could still be bad if we get all our checkers around (or nearly around) before White gets all his checkers onto the anchor. And at that point his board still needs a lot of work.
Bottom line good news is that we don't need to decide on the take and I feel like my calculation more likely favored White, so...
Redouble
|
BGonline.org Forums is maintained by Stick with WebBBS 5.12.