Concession stand, part 1

Tim Chow

unread,

Jun 25, 2012, 6:52:54 PM6/25/12

to

This time I'll post three related decisions, one at a time, before
giving bot results. Note that it's DMP.

The subject line is borrowed from a similar problem I posted some
months ago.

XGID=-BBBBAB-----bB----cdBc-ba-:0:0:1:62:0:0:0:1:10

X:Player 1 O:Player 2
Score is X:0 O:0 1 pt(s) match.
+13-14-15-16-17-18------19-20-21-22-23-24-+
| X O | | O X O O O |
| X O | | O X O O |
| O | | O O |
| | | O |
| | | |
| |BAR| |
| | | |
| | | |
| | | |
| O | | X X X X X |
| O | | X X X X X X |
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pipcount X: 103 O: 88 X-O: 0-0/1
Cube : 1
X to play 62

---
Tim Chow

Bradley K. Sherman

unread,

Jun 25, 2012, 7:52:26 PM6/25/12

to

Tim Chow <tchow...@yahoo.com> wrote:
> +13-14-15-16-17-18------19-20-21-22-23-24-+
> | X O | | O X O O O |
> | X O | | O X O O |
> | O | | O O |
> | | | O |
> | | | |
> | |BAR| |
> | | | |
> | | | |
> | | | |
> | O | | X X X X X |
> | O | | X X X X X X |
> +12-11-10--9--8--7-------6--5--4--3--2--1-+

>Pipcount X: 103 O: 88 X-O: 0-0/1, Cube: 1, X to play 62

13/5.

I like being set up for subsequent instant winners. The
smoothness of X's board makes up for much of the race
disparity, so I don't feel that strongly about this vs.
13/7,13/11.

--bks

Walt

unread,

Jun 26, 2012, 10:42:57 AM6/26/12

to

X is way behind in the race, so moving off the anchor has to be out.
That leaves the choice of filling in the homeboard (which leaves a shot)
or moving both checkers off the midpoint.

The Safe vs bold criteria say bold, so I'll play 13/5. make the full
board now while we have a chance, and if hit we have lots of
counterplay. In fact, I don't know that I would hit the midpoint blot
as O even if I could.

13/5.

--
//Walt

Tim Chow

unread,

Jun 30, 2012, 7:20:26 PM6/30/12

to

On Jun 25, 6:52 pm, I wrote:
> XGID=-BBBBAB-----bB----cdBc-ba-:0:0:1:62:0:0:0:1:10
>
> X:Player 1 O:Player 2
> Score is X:0 O:0 1 pt(s) match.
> +13-14-15-16-17-18------19-20-21-22-23-24-+
> | X O | | O X O O O |
> | X O | | O X O O |
> | O | | O O |
> | | | O |
> | | | |
> | |BAR| |
> | | | |
> | | | |
> | | | |
> | O | | X X X X X |
> | O | | X X X X X X |
> +12-11-10--9--8--7-------6--5--4--3--2--1-+
> Pipcount X: 103 O: 88 X-O: 0-0/1
> Cube : 1
> X to play 62

The choice here is between 13/5 and 13/11 13/7. 13/5 beautifully
closes our board, but it leaves a blot on our midpoint. Is that blot
an asset or a liability? O has a blot in her table, but she can hit
and cover with 61 51 31 11, and in these cases X will certainly regret
playing 13/5 rather than 13/11 13/7. The midpoint checker does
maintain contact with O's midpoint, but this contact is only
temporary, since X will often have to run that checker next turn
anyway. Are there any rolls by O that force O to break her midpoint
immediately? No. O has a spare on her bar point that she can use to
play 6's. In fact, she'll probably use that spare to play 6's even if
X plays 13/11 13/7, so keeping the blot on the midpoint doesn't even
force O to play 6's in a way that she wouldn't want to. The only real
advantage of 13/5 is that it closes X's board, but this is not enough
compensation for O's 7 hit-and-cover numbers, since X is a favorite to
close his board after 13/11 13/7 anyway. 13/11 13/7 is the best play
here.

If O didn't have that spare on her 7pt, then the decision would be
much closer, as the variant below illustrates.

=======
Rollout
=======

1. Rollout: 13/11 13/7 eq:-0.329
Player : 33.55% (G:1.16% B:0.01%)
Opponent: 66.45% (G:2.15% B:0.02%)
Confidence: ±0.003 (-0.332<E<-0.326)

2. Rollout: 13/5 eq:-0.387 (-0.058)
Player : 30.65% (G:1.25% B:0.01%)
Opponent: 69.35% (G:3.02% B:0.02%)
Confidence: ±0.003 (-0.390<E<-0.384)

1296 Games rolled with Variance Reduction.
Dice Seed: 2
Moves and cube decisions: 3 ply

=======
Variant
=======

XGID=-BBBBAB-----bB----beBc-ba-:0:0:1:62:0:0:0:1:10

X:Player 1 O:Player 2
Score is X:0 O:0 1 pt(s) match.
+13-14-15-16-17-18------19-20-21-22-23-24-+
| X O | | O X O O O |
| X O | | O X O O |
| | | O O |
| | | O |
| | | O |
| |BAR| |
| | | |
| | | |
| | | |
| O | | X X X X X |
| O | | X X X X X X |
+12-11-10--9--8--7-------6--5--4--3--2--1-+

Pipcount X: 103 O: 87 X-O: 0-0/1

Cube : 1
X to play 62

1. Rollout: 13/11 13/7 eq:-0.327
Player : 33.64% (G:1.75% B:0.01%)
Opponent: 66.36% (G:2.31% B:0.02%)
Confidence: ±0.003 (-0.330<E<-0.324)

2. Rollout: 13/5 eq:-0.334 (-0.007)
Player : 33.30% (G:2.59% B:0.05%)
Opponent: 66.70% (G:3.57% B:0.02%)
Confidence: ±0.003 (-0.337<E<-0.331)

1296 Games rolled with Variance Reduction.
Dice Seed: 2
Moves and cube decisions: 3 ply

eXtreme Gammon Version: 1.21, MET: Rockwell-Kazaross

---
Tim Chow