Discussion:
Six-primed
(too old to reply)
Tim Chow
2020-06-20 15:51:28 UTC
Permalink
XGID=-B-BaCCBB--a---bcbbbbA----:1:1:1:64:0:0:0:0:10

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-+
| O O O O | | O O X |
| O O O O | | O O |
| O | | |
| | | |
| | | |
| |BAR| |
| | | |
| | | |
| | | X X | +---+
| X X | | X X X X | | 2 |
| O X X | | X X O X X | +---+
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 92 O: 134 X-O: 0-0
Cube: 2, X own cube
X to play 64

---
Tim Chow
b***@gmail.com
2020-06-21 00:46:36 UTC
Permalink
Post by Tim Chow
XGID=-B-BaCCBB--a---bcbbbbA----:1:1:1:64:0:0:0:0:10
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-+
| O O O O | | O O X |
| O O O O | | O O |
| O | | |
| | | |
| | | |
| |BAR| |
| | | |
| | | |
| | | X X | +---+
| X X | | X X X X | | 2 |
| O X X | | X X O X X | +---+
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 92 O: 134 X-O: 0-0
Cube: 2, X own cube
X to play 64
---
Tim Chow
Why would we do anything other than lift the 7pt?

Stick
Tim Chow
2020-06-22 14:13:23 UTC
Permalink
XGID=-B-BaCCBB--a---bcbbbbA----:1:1:1:64:0:0:0:0:10

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-+
| O O O O | | O O X |
| O O O O | | O O |
| O | | |
| | | |
| | | |
| |BAR| |
| | | |
| | | |
| | | X X | +---+
| X X | | X X X X | | 2 |
| O X X | | X X O X X | +---+
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 92 O: 134 X-O: 0-0
Cube: 2, X own cube
X to play 64

I've found that in positions in which one is six-primed, the DMP play and
the money play often differ. For example, in a position like this one,
where X has just one straggler trapped, and O's board is weak, the DMP
strategy is typically to not worry about leaving blots, since getting another
checker sent back helps X anchor and put up a better defense. On the other
hand, for money, such a strategy often leads to too many gammon losses,
and X is better off just accepting his fate and trying not to get a second
checker sent back.

Here, the DMP play is 8/4* 8/2, but is it the money play? XG 3-ply doesn't
think so, and plays safe. However, the rollout says that 8/4* 8/2 is
mandatory.

Not sure Stick's rhetorical question is an oblique way of saying that one
should just make the DMP play here, but as it turns out, making the DMP play
is indeed what the rollout says to do.

1. Rollout¹ 8/2 8/4* eq:-0.799
Player: 16.48% (G:1.55% B:0.03%)
Opponent: 83.52% (G:18.73% B:0.83%)
Confidence: ±0.009 (-0.808..-0.790) - [100.0%]

2. Rollout¹ 8/4* 7/1 eq:-0.881 (-0.082)
Player: 13.17% (G:1.50% B:0.02%)
Opponent: 86.83% (G:18.82% B:0.95%)
Confidence: ±0.008 (-0.889..-0.873) - [0.0%]

3. Rollout¹ 7/3 7/1 eq:-0.902 (-0.103)
Player: 9.07% (G:1.04% B:0.01%)
Opponent: 90.93% (G:11.73% B:0.42%)
Confidence: ±0.007 (-0.910..-0.895) - [0.0%]

4. Rollout¹ 8/2 6/2 eq:-0.904 (-0.106)
Player: 9.25% (G:0.79% B:0.01%)
Opponent: 90.75% (G:12.04% B:0.37%)
Confidence: ±0.007 (-0.911..-0.898) - [0.0%]

5. Rollout¹ 7/1 6/2 eq:-0.906 (-0.107)
Player: 9.27% (G:1.03% B:0.01%)
Opponent: 90.73% (G:12.51% B:0.41%)
Confidence: ±0.007 (-0.912..-0.899) - [0.0%]

6. Rollout¹ 8/2 7/3 eq:-0.914 (-0.116)
Player: 9.20% (G:0.94% B:0.01%)
Opponent: 90.80% (G:13.02% B:0.50%)
Confidence: ±0.007 (-0.921..-0.908) - [0.0%]

7. Rollout¹ 8/2 5/1 eq:-0.920 (-0.121)
Player: 8.84% (G:0.89% B:0.01%)
Opponent: 91.16% (G:12.85% B:0.41%)
Confidence: ±0.007 (-0.927..-0.914) - [0.0%]

8. Rollout¹ 7/1 5/1 eq:-0.934 (-0.135)
Player: 9.02% (G:1.06% B:0.01%)
Opponent: 90.98% (G:14.49% B:0.56%)
Confidence: ±0.007 (-0.941..-0.927) - [0.0%]

¹ 1296 Games rolled with Variance Reduction.
Dice Seed: 271828
Moves: 3-ply, cube decisions: XG Roller

eXtreme Gammon Version: 2.19.207.pre-release

---
Tim Chow
b***@gmail.com
2020-06-25 00:37:36 UTC
Permalink
Post by Tim Chow
XGID=-B-BaCCBB--a---bcbbbbA----:1:1:1:64:0:0:0:0:10
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-+
| O O O O | | O O X |
| O O O O | | O O |
| O | | |
| | | |
| | | |
| |BAR| |
| | | |
| | | |
| | | X X | +---+
| X X | | X X X X | | 2 |
| O X X | | X X O X X | +---+
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 92 O: 134 X-O: 0-0
Cube: 2, X own cube
X to play 64
I've found that in positions in which one is six-primed, the DMP play and
the money play often differ. For example, in a position like this one,
where X has just one straggler trapped, and O's board is weak, the DMP
strategy is typically to not worry about leaving blots, since getting another
checker sent back helps X anchor and put up a better defense. On the other
hand, for money, such a strategy often leads to too many gammon losses,
and X is better off just accepting his fate and trying not to get a second
checker sent back.
Here, the DMP play is 8/4* 8/2, but is it the money play? XG 3-ply doesn't
think so, and plays safe. However, the rollout says that 8/4* 8/2 is
mandatory.
Not sure Stick's rhetorical question is an oblique way of saying that one
should just make the DMP play here, but as it turns out, making the DMP play
is indeed what the rollout says to do.
1. Rollout¹ 8/2 8/4* eq:-0.799
Player: 16.48% (G:1.55% B:0.03%)
Opponent: 83.52% (G:18.73% B:0.83%)
Confidence: ±0.009 (-0.808..-0.790) - [100.0%]
2. Rollout¹ 8/4* 7/1 eq:-0.881 (-0.082)
Player: 13.17% (G:1.50% B:0.02%)
Opponent: 86.83% (G:18.82% B:0.95%)
Confidence: ±0.008 (-0.889..-0.873) - [0.0%]
3. Rollout¹ 7/3 7/1 eq:-0.902 (-0.103)
Player: 9.07% (G:1.04% B:0.01%)
Opponent: 90.93% (G:11.73% B:0.42%)
Confidence: ±0.007 (-0.910..-0.895) - [0.0%]
4. Rollout¹ 8/2 6/2 eq:-0.904 (-0.106)
Player: 9.25% (G:0.79% B:0.01%)
Opponent: 90.75% (G:12.04% B:0.37%)
Confidence: ±0.007 (-0.911..-0.898) - [0.0%]
5. Rollout¹ 7/1 6/2 eq:-0.906 (-0.107)
Player: 9.27% (G:1.03% B:0.01%)
Opponent: 90.73% (G:12.51% B:0.41%)
Confidence: ±0.007 (-0.912..-0.899) - [0.0%]
6. Rollout¹ 8/2 7/3 eq:-0.914 (-0.116)
Player: 9.20% (G:0.94% B:0.01%)
Opponent: 90.80% (G:13.02% B:0.50%)
Confidence: ±0.007 (-0.921..-0.908) - [0.0%]
7. Rollout¹ 8/2 5/1 eq:-0.920 (-0.121)
Player: 8.84% (G:0.89% B:0.01%)
Opponent: 91.16% (G:12.85% B:0.41%)
Confidence: ±0.007 (-0.927..-0.914) - [0.0%]
8. Rollout¹ 7/1 5/1 eq:-0.934 (-0.135)
Player: 9.02% (G:1.06% B:0.01%)
Opponent: 90.98% (G:14.49% B:0.56%)
Confidence: ±0.007 (-0.941..-0.927) - [0.0%]
¹ 1296 Games rolled with Variance Reduction.
Dice Seed: 271828
Moves: 3-ply, cube decisions: XG Roller
eXtreme Gammon Version: 2.19.207.pre-release
---
Tim Chow
For a play like this it's essential that the opponent's prime is mainly external so when we're hit we're able to anchor up not just at the edge of the prime, but that edge of the prime is significantly advanced so that when we do see daylight later in the game we have a fair chance of spring our checkers. In other words, if you keep the 6 prime but shift points from the back of the prime to the front of the prime 8/4* 8/2 should get significantly worse/wrong.

Stick

Loading...