Cube 117 - From off the bar II

[sixty_something]

If the board image does not appear in the question above, you may view it in the attached PNG file below.

one vote per user .. results of the poll are shown after you have voted

GNUbg says the correct actions are

Spoiler

This is the same position as in Cube 116, the only difference is the match score, now 0-3, rather than 0-2. I also reset the score to 0-4 and 0-5 getting a similar result - "Too good to cube." Since RED with one off already is in no danger of being gammoned for the match, the different evaluation must be the result of a possible redouble by WHITE to match point, if accepted. While this may just be an evaluation anomoly, I found the different results surprising and worthy of a pair of problems.

Cube analysis (RED to play)
0-ply cubeless equity  +0.605 (Money:  +0.598)
  0.695 0.205 0.003 - 0.305 0.000 0.000

Cubeful equities:
1. No double            +2.120
2. Double, take         +2.356  ( +0.235)
3. Double, pass         +1.000  ( -1.120)
Proper cube action: Too good to double, pass (82.6%)

[close]

you may enter the ID's below into GNUbg or Snowie to evaluate your answer:

Position ID: 250BgkC7twEAIA   Match ID: MAHgAAAAGAAA
               
Artists who seek perfection in everything are those who cannot attain it in anything.
-- Eugene Delacroix French Romantic painter (1798 - 1863)

[playBunny]

Hmmm, I get a different analysis using the match and position ids given.

Spoiler

Cube analysis
2-ply cubeless equity  +0.635 (Money:  +0.604)
   69.2%  21.7%   0.3% -  30.8%   0.0%   0.0%

Cubeful equities:
1. Double, pass         +1.000
2. Double, take          +1.129  ( +0.129)
3. No double             +0.842  ( -0.158)
Proper cube action: Double, pass

(This is 2-ply but the 0-ply result agrees with it being double/pass)

[close]

[sixty_something]

Hmmmm, indeed. When I entered the Position/Match ID's I got a similar 2-ply result to yours .. like I said, this may be an evaluation anolmoly .. here are the results I obtained at 2-ply, 0-ply, and after a Rollout with GNUbg

Spoiler

All suggest Double, pass as the preferred play. Not the different result I obtained on my first evaluation.

Cube analysis
2-ply cubeless equity  +0.630 (Money:  +0.604)
  0.692 0.217 0.003 - 0.308 0.000 0.000

Cubeful equities:
1. Double, pass         +1.000
2. Double, take         +1.110  ( +0.110)
3. No double            +0.838  ( -0.162)
Proper cube action: Double, pass


Cube analysis
0-ply cubeless equity  +0.623 (Money:  +0.598)
  0.695 0.205 0.003 - 0.305 0.000 0.000
Cubeful equities:
1. Double, pass         +1.000
2. Double, take         +1.091  ( +0.091)
3. No double            +0.879  ( -0.121)
Proper cube action: Double, pass


Cube analysis
Rollout cubeless equity  +0.665 (Money:  +0.640)

Cubeful equities:
1. Double, pass         +1.000
2. Double, take         +1.322  ( +0.322)
3. No double            +0.848  ( -0.152)
Proper cube action: Double, pass

Rollout details:
Centered 1-cube:
  0.714 0.209 0.003 - 0.286 0.000 0.000 CL  +0.665 CF  +0.848
[0.001 0.001 0.000 - 0.001 0.000 0.000 CL   0.002 CF   0.003]
Player Me owns 2-cube:
  0.731 0.215 0.003 - 0.269 0.000 0.000 CL  +1.475 CF  +1.322
[0.001 0.001 0.000 - 0.001 0.000 0.000 CL   0.005 CF   0.007]
Truncated cubeful rollout (depth 11) with var.redn.
1296 games, Mersenne Twister dice gen. with seed 808408639 and quasi-random dice
Play: 0-ply cubeful prune [expert]
Cube: 0-ply cubeful prune [expert]

[close]