Discussion:
Cube Action?
(too old to reply)
BlueDice
2020-06-18 16:52:55 UTC
Permalink
XGID=-abBB-C-A---dE--Ac-e----A-:0:0:1:00:0:0:0:0:10
X:Player 1 O:Player 2

Score is X:0 O:0. Unlimited Game
+13-14-15-16-17-18------19-20-21-22-23-24-+
| X X O | | O X |
| X O | | O |
| X O | | O |
| X | | O |
| X | | O |
| |BAR| |
| | | |
| O | | |
| O | | X |
| O | | X X X O |
| O X | | X X X O O |
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 145 O: 176 X-O: 0-0
Cube: 1
X on roll, cube action
--
BD
Tim Chow
2020-06-18 17:51:15 UTC
Permalink
Post by BlueDice
XGID=-abBB-C-A---dE--Ac-e----A-:0:0:1:00:0:0:0:0:10
X:Player 1 O:Player 2
Score is X:0 O:0. Unlimited Game
+13-14-15-16-17-18------19-20-21-22-23-24-+
| X X O | | O X |
| X O | | O |
| X O | | O |
| X | | O |
| X | | O |
| |BAR| |
| | | |
| O | | |
| O | | X |
| O | | X X X O |
| O X | | X X X O O |
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 145 O: 176 X-O: 0-0
Cube: 1
X on roll, cube action
It seems like a pretty clear take to me, even though O is 30 pips down and
has no new points. X doesn't have a very strong offensive structure yet.
I'm inclined to double, though.

---
Tim Chow
BlueDice
2020-06-20 08:47:59 UTC
Permalink
Post by BlueDice
XGID=-abBB-C-A---dE--Ac-e----A-:0:0:1:00:0:0:0:0:10
X:Player 1 O:Player 2
Score is X:0 O:0. Unlimited Game
+13-14-15-16-17-18------19-20-21-22-23-24-+
| X X O | | O X |
| X O | | O |
| X O | | O |
| X | | O |
| X | | O |
| |BAR| |
| | | |
| O | | |
| O | | X |
| O | | X X X O |
| O X | | X X X O O |
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 145 O: 176 X-O: 0-0
Cube: 1
X on roll, cube action
--
BD
XGR++ says borderline ND/T


Analyzed in XG Roller++
Player Winning Chances: 65.18% (G:23.34% B:1.38%)
Opponent Winning Chances: 34.82% (G:7.14% B:0.28%)

Cubeless Equities: No Double=+0.477, Double=+0.956

Cubeful Equities:
No double: +0.732
Double/Take: +0.706 (-0.026)
Double/Pass: +1.000 (+0.268)

Best Cube action: No double / Take
Percentage of wrong pass needed to make the double decision right: 8.2%
--
BD
b***@gmail.com
2020-06-21 00:43:39 UTC
Permalink
Post by BlueDice
Post by BlueDice
XGID=-abBB-C-A---dE--Ac-e----A-:0:0:1:00:0:0:0:0:10
X:Player 1 O:Player 2
Score is X:0 O:0. Unlimited Game
+13-14-15-16-17-18------19-20-21-22-23-24-+
| X X O | | O X |
| X O | | O |
| X O | | O |
| X | | O |
| X | | O |
| |BAR| |
| | | |
| O | | |
| O | | X |
| O | | X X X O |
| O X | | X X X O O |
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 145 O: 176 X-O: 0-0
Cube: 1
X on roll, cube action
--
BD
XGR++ says borderline ND/T
Analyzed in XG Roller++
Player Winning Chances: 65.18% (G:23.34% B:1.38%)
Opponent Winning Chances: 34.82% (G:7.14% B:0.28%)
Cubeless Equities: No Double=+0.477, Double=+0.956
No double: +0.732
Double/Take: +0.706 (-0.026)
Double/Pass: +1.000 (+0.268)
Best Cube action: No double / Take
Percentage of wrong pass needed to make the double decision right: 8.2%
--
BD
I'm not sure where everyone's definition of *borderline* is for a cube action but mine is much lower than .026. (I'd say .010 or less) FWIW a rollout has this as a ND by at least .050. Tbh I don't hate the cube v. the right humans because you will get some passes.

Stick
Paul Epstein
2020-06-21 09:29:10 UTC
Permalink
Post by b***@gmail.com
Post by BlueDice
Post by BlueDice
XGID=-abBB-C-A---dE--Ac-e----A-:0:0:1:00:0:0:0:0:10
X:Player 1 O:Player 2
Score is X:0 O:0. Unlimited Game
+13-14-15-16-17-18------19-20-21-22-23-24-+
| X X O | | O X |
| X O | | O |
| X O | | O |
| X | | O |
| X | | O |
| |BAR| |
| | | |
| O | | |
| O | | X |
| O | | X X X O |
| O X | | X X X O O |
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 145 O: 176 X-O: 0-0
Cube: 1
X on roll, cube action
--
BD
XGR++ says borderline ND/T
Analyzed in XG Roller++
Player Winning Chances: 65.18% (G:23.34% B:1.38%)
Opponent Winning Chances: 34.82% (G:7.14% B:0.28%)
Cubeless Equities: No Double=+0.477, Double=+0.956
No double: +0.732
Double/Take: +0.706 (-0.026)
Double/Pass: +1.000 (+0.268)
Best Cube action: No double / Take
Percentage of wrong pass needed to make the double decision right: 8.2%
--
BD
I'm not sure where everyone's definition of *borderline* is for a cube action but mine is much lower than .026. (I'd say .010 or less) FWIW a rollout has this as a ND by at least .050. Tbh I don't hate the cube v. the right humans because you will get some passes.
I think a sensible definition would be "Low enough so that any human,
no matter how good, should not be concerned if their decision went the other
way."

This definition means that my definition of "borderline" varies with
the difficulty of the position. But hey, that's a bug, not a feature.
So please don't sue me if my definition is bad.
Also, you already sued me over some previous analysis.

I agree (with everyone except Stick) that, for a position that is not
ultra-simple, 0.026 is indeed borderline.

However, suppose your only two checkers are on your deuce point, and
my only two checkers are on my three point and two point and you own the
cube. If I wrongly take your double, I wouldn't dismiss my bad take
as "borderline". It's a pretty simple position that should be known.
The loss from taking is 1/81 which is in between the 0.026 and 0.01
thresholds.

Suppose we observe world-class (human) play in contact positions
and consider all doubling decisions where the difference between the right and wrong action is <= 0.026.
I don't think the proportion of correct decisions would be significantly
50%. Since I don't think humans can tell the difference between a
0.026 cube error and a correct cube action, I think it's borderline.

Haven't you heard Madonna's song: "You keep on pushing my cube over the
borderline"? This was addressed at strong bg players.

Paul
Paul Epstein
2020-06-21 09:34:58 UTC
Permalink
Post by Paul Epstein
Suppose we observe world-class (human) play in contact positions
and consider all doubling decisions where the difference between the right and wrong action is <= 0.026.
I don't think the proportion of correct decisions would be significantly
50%. Since I don't think humans can tell the difference between a
0.026 cube error and a correct cube action, I think it's borderline.
I wrote "significantly 50%" when I meant "significantly > 50%".
What makes this correction particularly awkward is that google groups
puts in ">" tags to denote quotations, thus making it seem that I never
made a mistake in the first place.
If only all my errors were overlooked by this type of coincidence.

Paul

Loading...