BGonline.org Forums

to the edge or prime?

Posted By: David Rockwell
Date: Thursday, 19 June 2008, at 4:01 a.m.

In Response To: to the edge or prime? (bob koca)

I've put together some analysis looking at each of the rolls to see if this would help us see what is happening. The analysis uses Snowie 3 ply. Snowie thinks the best play coming to the edge of the prime is 24-22, 8-7(2), so this is what I've used.

Snowie lists 6-4(2) [choice #1] as being preferred over 24-22, 8-7(2) [choice #2] by .339. I've broken this amount down roll by roll. The rolls are grouped to faciliate discussion. I've listed the rolls contained in each group in paranthesis. Positive numbers favor choice #1. Negative numbers favor choice #2.

Group #1 - The blot on the eight point is hit after choice #2 (4-3, 5-2, 6-1) / .174 {.174 is the component of the .339 difference due to these six rolls}

Group #2 - Point on head (POH) on the 22 point after choice #2 (2-2, 4-4, 5-3, 5-4, 5-5) / .140

Group #3 - Making the 4 point after choice #1 blocks threes effectively (3-1, 3-2, 3-3) / .077

Group #4 - Rolls with smaller impact that I am not going to bother to discuss in detail (1-1, 2-1, 4-1, 4-2, 5-1, 6-2, 6-3, 6-5) / .035

Group #5 - The blot on the eleven point is hit after choice #1 (6-4) / -.046

Group #6 - boxes are blocked by the bar after choice #2 (6-6) / -.042

These six numbers add to .338, very close to the Snowie evaluation difference of .339.

Here's what I see in this analysis. The general priming and long term advantages of choice #1 are significant. The possibility of escaping the back checker next turn is embedded in these numbers and just doesn't help as much as the offensive improvement. This was pointed out in earlier posts. (Group #3 & Group #4) / .112

But, the bulk of the variance comes from the 13 big hitting rolls which are avoided by choice #1 (Group #1 & Group #2) / .314

And, there are only three of the opponent's rolls which punish choice #1 to offset the 13 rolls above (Group #5 & Group #6) / -.088

Now, there's a problem with my presentation of these numbers. Consider the 5-5 roll. Not only are we in less trouble if we played choice #1, we also have a better prime. So, conceptually, some of the .314 number for the hitting rolls really is due to the priming advantages of making the four point. Some of the .314 needs to be shifted into the .112 number. I'm not going to attempt to do this.

In summary 6-4(2) is preferred because it avoids immediate trouble and because of its priming advantages. Both considerations are significant.

I'm not sure whether or not I've presented these numbers in a way that can be easily understood. Let me know if further explanation is desired.

I just realized I should have repeated the diagram. Oh well.

Messages In This Thread

• to the edge or prime?
  bob koca -- Monday, 16 June 2008, at 3:53 p.m.
  • to the edge or prime?
    Achim -- Monday, 16 June 2008, at 9:28 p.m.
  • to the edge or prime?
    Stick -- Monday, 16 June 2008, at 10:56 p.m.
    • to the edge or prime?
      bob koca -- Tuesday, 17 June 2008, at 2:19 p.m.
      • to the edge or prime?
        neilkaz -- Tuesday, 17 June 2008, at 2:57 p.m.
      • to the edge or prime?
        Stick -- Tuesday, 17 June 2008, at 3:26 p.m.
      • to the edge or prime?
        David Rockwell -- Wednesday, 18 June 2008, at 4:30 p.m.
        • to the edge or prime?
          Stick -- Wednesday, 18 June 2008, at 4:31 p.m.
  • to the edge or prime?
    Dave Ranson -- Monday, 16 June 2008, at 11:05 p.m.
  • to the edge or prime?
    neilkaz -- Tuesday, 17 June 2008, at 12:29 a.m.
  • to the edge or prime?
    bob koca -- Wednesday, 18 June 2008, at 8:28 p.m.
  • to the edge or prime?
    David Rockwell -- Thursday, 19 June 2008, at 4:01 a.m.