| |
BGonline.org Forums
xxP-54, progression of R, S, D
Posted By: Nack Ballard In Response To: xxP-54, progression of R, S, D (Perry Gartner)
Date: Sunday, 25 July 2010, at 12:24 a.m.
3. Making the 3 seems to lead to more attacking than priming so the results, althugh close are still counter to the theory. However I wonder if at match scores equal to money, 0-0 to 7 for example, the gammons on one wouldn't reverse the result? Or is the premise of the 3 point tending towards an attacking GP overstated? How would you explain it and did I convince someone to do a rollout?
Good, thought-provoking questions, Perry.
Occasionally, I roll out positions at 0-0 to 7 on GnuBG by accident (because my parameters are arbitrarily set for a 7-point match and I forget to click on the proper settings for one of the standard four scores When this happens (a lot less often nowadays), I call the result "quasi-money" and append the file name with "Q" instead of M, D, S or G. (I use capitals for file names and lower case -- m, d, s, g -- when writing/inserting them before nacbracs.)
I haven't made a point of comparing Q file results to those of M (pure money), though someday I might make a project of searching for them to see what sort of patterns might emerge. For the time being, I blindly assume that Q follows M (indeed when I refer to M/nothing it is intended to include Q under its umbrella for the most part) and M is the best reference "score", so that's where I focus CPU time. (That said, I don't want to discourage anyone else from doing an intentional -7-7 rollout.)
As for the concept of following up the making of the 3pt with an attack, I think the two common ways of making the 3pt are sometimes confused (not saying that you did), and that can give rise to erroneous principles. If the 3pt is made with 55 (double 5s), there are two extra checkers in the zone compared to making the 3pt with 53 -- the 6pt/8pt distribution is five/three compared to four/two. It tends to be suicidal for Opp to split vs 55 (better to stay put, avoid the blitz and win the priming battle) but very reasonable to split vs 53.
Consider the last three positions just covered, where Opp has made an inside point (4pt, 3pt, 2pt). For ease of comparison, I'll omit R (Run, 24/15) and isolate S (Split, 24/20 13/8) versus D (Down, 13/9 13/8):
42P-54 [S D40]
53P-54 [S D7]
64P-54 [D S26]There is a difference of .033 between the 4pt and 3pt (S/D) margins. And there is an identical .033 difference between the 3pt and 2pt margins. The typical classifications of high point (for 5pt and 4pt) and low point for (3pt and 2pt) or more for convenience than anything else -- there is no sudden "wrinkle" or drop-off between the 4pt and 3pt.
Wrinkles and even hiccups/reversals exist, but they are usually explained by quirks of specific rolls. For example, it is normally more reasonable to run past the 18pt when facing the 5pt instead of the 4pt (which also means gaining 2 pips in the race) -- and that principle holds up for 31P-6x vs 42P-6x where x is 2, 4 or 5; and yet we have 31P-63 [S R16] vs 42P-63 [S R4]. It is useful to notice that after running with 63 in the former position you are blocked from linking the back checkers with 54 on the following roll.
Nack
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.