BGonline.org Forums

31P-64 data, 42P-64, 53P-64

Posted By: Nack Ballard
Date: Tuesday, 11 May 2010, at 7:24 p.m.

In Response To: 31P - 64 to play leading 2a/8a (Daniel Murphy)

Here is data for 31P-64:

Money
Snowie .. [R S14 P27] 20k
GnuBG .. [R P19 S21] 5k

Other scores
GnuBG .. d[R S3.2] 10k
GnuBG .. g[S P0.1] 5k

Key: R = Run (24/14), S = Split (24/18 13/9), P = Point (8/2 6/2). Error sizes are in thousandths. Score is given before the brackets (d = DMP, g = GG, else assume money), and number of trials (in thousands, rounded down) are given after the brackets. For example, "d[R S3.2] 10k" means that at DMP, R is best and S is -.0032 after 10000+ trials.

R wasn't in Gnu's GG rollout, but Snowie has it at about -.025.

I think it is only at GG that 31P-64P is reasonable, though personally I never play it. I might try S at an even score against a much weaker opponent, but against most opps I play R unless I'm at GG (gammon go) or Gg (gammon good) or -4-2 or worse, in which case I play S.

At the actual score (-2-8), R is right by an even bigger margin than for money. I'd guess that S is an error of .040 and P is an error of .050.

Click here for more on 31P-64.

What about

(a) 42P-64 for money
(b) 53P-64 at GS (or at -2-8, is your answer different?)
(c) 53P-64 at DMP

Anyone?

Nack

Messages In This Thread

• 31P - 64 to play leading 2a/8a
  Daniel Murphy -- Tuesday, 11 May 2010, at 4:38 p.m.
  • 31P - 64 to play leading 2a/8a
    Chase -- Tuesday, 11 May 2010, at 4:59 p.m.
    • 31P - 64 to play leading 2a/8a
      John O'Hagan -- Tuesday, 11 May 2010, at 5:39 p.m.
      • 31P - 64 to play leading 2a/8a
        Seth -- Tuesday, 11 May 2010, at 6:38 p.m.
  • 31P - 64 to play leading 2a/8a
    neilkaz -- Tuesday, 11 May 2010, at 5:18 p.m.
  • 31P-64 data, 42P-64, 53P-64
    Nack Ballard -- Tuesday, 11 May 2010, at 7:24 p.m.
    • 31P-64 data, 42P-64, 53P-64
      Stick -- Tuesday, 11 May 2010, at 7:49 p.m.
      • 31P-64 data, 42P-64, 53P-64
        Nack Ballard -- Tuesday, 11 May 2010, at 8:43 p.m.