| |
BGonline.org Forums
RO of your 38-43 position
Posted By: Ian Shaw In Response To: RO of your 38-43 position (neilkaz)
Date: Saturday, 6 February 2010, at 9:19 p.m.
Neil gives a good example of a maximally efficient position for both sides that results in a miniscule take one pip deeper than predicted by the table. However, it only takes a small amount of wastage to push into pass territory.
The optimal bearoff with 38 pips (on the left) wastes 6.51. White's position on the right wastes 6.85 pips - a difference of 0.34 pips. This certainly can be classed as an "efficient position", and fits in with the table.
I think this illustrates my contention that the table is intended to apply to low-wastage positions in general. (As with all generalizations, there are exceptions.)
gnubg 43
ian 38 Position ID: 6B4AANAeAAAAAA Match ID: cAkAAAAAAAAA
Cube decision 0-ply cubeless equity +0.5485 0.7743 0.0000 0.0000 - 0.2257 0.0000 0.0000 Cubeful equities: 1. Double, take +0.9944 2. Double, pass +1.0000 +0.0056 3. No double +0.8056 -0.1888 Proper cube action: Double, take
gnubg 43
ian 38 Position ID: ZB8AANAeAAAAAA Match ID: cAkOAAAAAAAA
Cube decision 2-ply cubeless equity +0.5728 0.7864 0.0000 0.0000 - 0.2136 0.0000 0.0000 Cubeful equities: 1. Double, pass +1.0000 2. Double, take +1.0522 +0.0522 3. No double +0.8344 -0.1656 Proper cube action: Double, pass
The results here are from the gnubg two-sided database (the 0- or 2-ply distinction is false). I don't see any need for a rollout when we have pre-computed precise data available.
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.