BGonline.org Forums

31P-xxP-51

Posted By: Nack Ballard
Date: Tuesday, 11 May 2010, at 5:11 a.m.

In Response To: 31P-xxP-51 (mtuhtan)

Thank you for the rollouts, Miran.

The data for these four related positions is shown below. The first nacbrac in each row is XG, the second is GnuBG:

31P-61P-51 [S N27] 10k, [S N22] 5k
31P-31P-51 [S N4 R26] 20k, [S N6.6 R20] 15k 5kR
31P-42P-51 [S N23] 10k, [S N28] 5k
31P-53P-51 [N S4] 20k, [N S3.9] 10k
31P-65R-51 [S N1] 20k, [S N8.3] 5k

Averaging (simplifying), I get these margins (in thousandths of a point) for S (Split, 24/23 13/8) vs N (Near, 13/8 6/5):

31P-61P-51 [S N25]
31P-31P-51 [S N5]
31P-42P-51 [S N25]
31P-53P-51 [N S4]

For reference, the four positions are diagrammed below, with comments beneath:

31P-61P-51 [S N25]

31P-31P-51 [S N5]

31P-42P-51 [S N25]

31P-53P-51 [N S4]

Normally, one would expect that the higher/primier White's point, the more motivated Blue is to split. The relationship between [S N25] in the first position (where White owns the 7pt) and [N S4] in the fourth position (where White owns the 3pt) therefore makes sense. But there's something unusual going on with the 5pt and 4pt margins -- [S N5] and [S N25], respectively.

Two rolls stand out to me as major contributors to this hiccup: 44 and 61.

Blue has made his 5pt, blocking double 4s. White cannot play B (Both, 24/20 13/9, as she normally would like to); she must play on her own side of the board. If Blue splits 24/23 against the 4pt, it's no problem because White won't point on her anyway with 44 (she'll play P -- make her own 5pt).

Conversely, if Blue splits against the 5pt, he'll get drilled with A (Attack, 8/4(2) 6/2*(2)), whereas if Blue plays 51N (13/8 6/5), White has to settle for instead of the considerably milder O (Outer, 13/9(2) 8/4(2)).

The deal with White's 61 is that Blue's splitting prevents White from making her 7pt (the cost of leaving the direct shot on her 8pt is too great). Her best play if she owns her 4pt is 13/6 (according to Snowie eval). Owning her 5pt, though, she can improve on 13/6 greatly by playing 24/18 6/5.

I doubt that the effects of 44 and 61 alone explain the entire "hiccup" between 31P-31P-51 [S N5] and 31P-42P-51 [S N25], which is .020 plus maybe .010 that it "should" be the other way). Someday, I'll examine this set of positions in more depth.

Nack

Messages In This Thread

• 31P-xxP-51
  mtuhtan -- Friday, 7 May 2010, at 2:31 p.m.
  • 31P-65R-51
    mtuhtan -- Friday, 7 May 2010, at 2:49 p.m.
    • 31P-65R-51 - Rollout
      mtuhtan -- Monday, 10 May 2010, at 3:02 p.m.
  • 31P-xxP-51
    Casper van der Tak -- Friday, 7 May 2010, at 4:23 p.m.
    • 31P-xxP-51
      mtuhtan -- Sunday, 9 May 2010, at 8:46 p.m.
  • 31P-xxP-51
    Leonardo Jerkovic -- Sunday, 9 May 2010, at 10:01 p.m.
  • 31P-61P-51 - Rollout
    mtuhtan -- Monday, 10 May 2010, at 2:58 p.m.
  • 31P-31P-51 - Rollout
    mtuhtan -- Monday, 10 May 2010, at 2:59 p.m.
  • 31P-42P-51 - Rollout
    mtuhtan -- Monday, 10 May 2010, at 3:00 p.m.
  • 31P-53P-51 - Rollout
    mtuhtan -- Monday, 10 May 2010, at 3:01 p.m.
  • 31P-xxP-51
    Nack Ballard -- Tuesday, 11 May 2010, at 5:11 a.m.