Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Money game, 43 to play

1 view
Skip to first unread message

Tim Chow

unread,
May 23, 2012, 5:52:37 PM5/23/12
to
XGID=a--BBBBBB-AA-A--------bba-:1:1:1:43:0:0:3:0:10

X:Player 1 O:Player 2
Score is X:0 O:0. Money session
+13-14-15-16-17-18------19-20-21-22-23-24-+
| X | | O O O |
| | | O O |
| | | |
| | | |
| | | |
| |BAR| |
| | O | |
| | | |
| | | | +---+
| X X | | X X X X | | 2 |
| X X X X | | X X X X | +---+
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pipcount X: 100 O: 36 X-O: 0-0
Cube : 2, X own cube
X to play 43

---
Tim Chow

Paul

unread,
May 24, 2012, 4:03:03 AM5/24/12
to
X needs to maintain the prime (as a non-world-class human sees it) but
also desperately needs to hit another blot.
X is fortunate that it's difficult for O to clean up to a blot-free
position.
10/6 11/8 over the board to prepare to attack and leave blots on the
lower points.
QF being what it is, I won't be terribly surprised if the blot-leaving
and return-hitting is more urgent than the above, and 8/1 is in fact
the best play. But I'll go with 10/6 11/8.

Play bold with good dice.
Play bold if they're not.
Cos it's all about the blot.

Paul Epstein





Walt

unread,
May 24, 2012, 9:32:10 AM5/24/12
to
The next objective is to plant a blot on the ace or deuce point and
force O to hit it so we can pick up a second checker. If we just close
the board we're only about 40% to win. But we need to keep the prime,
otherwise O can enter and leap.

QF suggests blotting now with 8/1, but this gives O a chance to enter
and leap while hitting with a roll of 62. I don't like it, but it may
well be correct.

My play is 11/8 10/6 aiming two builders at the open points. I
should be able to slot one of those points next roll, and unless O roll
11 or 22 there will still be a blot in her home board for me to aim at.

//Walt

Bradley K. Sherman

unread,
May 24, 2012, 9:38:42 AM5/24/12
to
Tim Chow <tchow...@yahoo.com> wrote:
> +13-14-15-16-17-18------19-20-21-22-23-24-+
> | X | | O O O |
> | | | O O |
> | | | |
> | | | |
> | | | |
> | |BAR| |
> | | O | |
> | | | |
> | | | | +---+
> | X X | | X X X X | | 2 |
> | X X X X | | X X X X | +---+
> +12-11-10--9--8--7-------6--5--4--3--2--1-+
>Pipcount X: 100 O: 36, Cube:2, X to play 43

The goal is to pick up the second checker by having
O hit X on entry. I could see breaking the 8 point
(leaving a five-pt prime) at some point, but for
now I'm looking to leave blots on tha 1&2 points
ASAP, so I'll play
10/6, 11/8

--bks

Paul

unread,
May 24, 2012, 9:54:19 AM5/24/12
to
On May 24, 2:32 pm, Walt <walt_ask...@yahoo.com> wrote:
> On 5/23/2012 5:52 PM, Tim Chow wrote:
>
>
>
>
>
>
>
>
>
> > XGID=a--BBBBBB-AA-A--------bba-:1:1:1:43:0:0:3:0:10
>
> > X:Player 1   O:Player 2
> > Score is X:0 O:0. Money session
> >  +13-14-15-16-17-18------19-20-21-22-23-24-+
> >  | X                |   |          O  O  O |
> >  |                  |   |          O  O    |
> >  |                  |   |                  |
> >  |                  |   |                  |
> >  |                  |   |                  |
> >  |                  |BAR|                  |
> >  |                  | O |                  |
> >  |                  |   |                  |
> >  |                  |   |                  | +---+
> >  |             X  X |   | X  X  X  X       | | 2 |
> >  |    X  X     X  X |   | X  X  X  X       | +---+
> >  +12-11-10--9--8--7-------6--5--4--3--2--1-+
> > Pipcount  X: 100  O: 36 X-O: 0-0
> > Cube  : 2, X own cube
> > X to play 43
>
...
>
> QF suggests blotting now with 8/1, but this gives O a chance to enter
> and leap while hitting with a roll of 62.  I don't like it, but it may
> well be correct.
>
...

Walt,

My analysis was extremely similar to yours and my play is the same.
However, it should be borne in mind that 62 is only a problem if X
fails to hit back on the next shake. The combined parlay that O leaps
and hits with 62 and X fails to hit back immediately is 1.2%

Paul Epstein

badgolferman

unread,
May 24, 2012, 9:55:44 AM5/24/12
to
Same as everyone else, 10/6, 11/8.

Paul

unread,
May 24, 2012, 4:04:05 PM5/24/12
to
On May 23, 10:52 pm, Tim Chow <tchow12...@yahoo.com> wrote:
After further thought, 8/1 is absolutely clear. Over the board I
would instinctively keep the prime. So it felt somehow dishonest to
move to the QF-friendly 8/1.

But 8/1 is clearly correct after further consideration.

The point is that when X slots the ace X's probability of picking up
the vital second blot improves massively.
The main downside only has a 1.2% probability (as discussed
elsethread).

If X slots the ace, X needs O to come in and hit (50% approx as O has
a long time to do this) and X needs to rehit (11/36).

But what if X doesn't slot the ace? Then X has less time in which to
get hit, and X needs extra luck to hit O when O enters.

When you go through the parlays, you see that refusing to slot the ace
makes it so much harder to hit the other blot.

8/1 is absolutely essential and it's a massive massive blunder not to
play it.

Good problem Tim!


Tim Chow

unread,
May 25, 2012, 8:41:50 PM5/25/12
to
Paul nailed it on his second try. When O has this many checkers off
and has a blot in her table to be picked up, getting that second
checker is imperative. The fact that O's blot is on her 1pt means
that she can't hit and clean up the blot at the same time.

For a variant position, I've given O two more checkers.

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

I didn't roll out 11/8 10/6 but I'm confident that that play isn't
significantly better than 13/9 10/7 or 13/6.

1. Rollout: 8/1 eq:+0.312
Player : 56.30% (G:0.00% B:0.00%)
Opponent: 43.70% (G:5.98% B:0.68%)
Confidence: ±0.011 (+0.301<E<+0.323)

2. Rollout: 13/9 10/7 eq:+0.269 (-0.043)
Player : 53.98% (G:0.00% B:0.00%)
Opponent: 46.02% (G:3.69% B:0.42%)
Confidence: ±0.010 (+0.259<E<+0.279)

3. Rollout: 13/6 eq:+0.267 (-0.045)
Player : 53.84% (G:0.00% B:0.00%)
Opponent: 46.16% (G:3.84% B:0.44%)
Confidence: ±0.010 (+0.257<E<+0.277)

1210 Games rolled with Variance Reduction.
Dice Seed: 2
Moves and cube decisions: 4 ply

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

XGID=a--BBBBBB-AA-A--------cca-:1:1:1:43:0:0:3:0:10

X:Player 1 O:Player 2
Score is X:0 O:0. Money session
+13-14-15-16-17-18------19-20-21-22-23-24-+
| X | | O O O |
| | | O O |
| | | O O |
| | | |
| | | |
| |BAR| |
| | O | |
| | | |
| | | | +---+
| X X | | X X X X | | 2 |
| X X X X | | X X X X | +---+
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pipcount X: 100 O: 41 X-O: 0-0
Cube : 2, X own cube
X to play 43

1. Rollout: 13/6 eq:+0.493
Player : 62.55% (G:0.00% B:0.00%)
Opponent: 37.45% (G:1.25% B:0.06%)
Confidence: ±0.005 (+0.488<E<+0.498)

2. Rollout: 13/9 10/7 eq:+0.489 (-0.003)
Player : 62.44% (G:0.00% B:0.00%)
Opponent: 37.56% (G:1.29% B:0.06%)
Confidence: ±0.005 (+0.484<E<+0.494)

3. Rollout: 13/9 11/8 eq:+0.486 (-0.007)
Player : 62.23% (G:0.00% B:0.00%)
Opponent: 37.77% (G:1.14% B:0.05%)
Confidence: ±0.005 (+0.481<E<+0.491)

4. Rollout: 8/1 eq:+0.479 (-0.014)
Player : 62.54% (G:0.00% B:0.00%)
Opponent: 37.46% (G:2.95% B:0.18%)
Confidence: ±0.005 (+0.474<E<+0.484)

5184 Games rolled with Variance Reduction.
Dice Seed: 2
Moves and cube decisions: 3 ply

eXtreme Gammon Version: 1.21

---
Tim Chow
0 new messages