On Saturday, December 10, 2022 at 8:56:46 AM UTC-5, Tim Chow wrote:
XGID=-BCBBbB-B------b---ccBb-c-:0:0:1:51:0:0:0:0:10
Score is X:0 O:0. Unlimited Game
+13-14-15-16-17-18------19-20-21-22-23-24-+
| O | | O O X O O |
| O | | O O X O O |
| | | O O O |
| | | |
| | | |
| |BAR| |
| | | |
| | | |
| | | X |
| X | | X O X X X X |
| X | | X O X X X X |
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 92 O: 102 X-O: 0-0
Cube: 1
X to play 51
Before trying to explain this one, see the variant below.
The fact that X has the stronger board and is running out of time
points toward breaking anchor, especially since the alternative
still volunteers a shot. Nevertheless, one still needs to take
into account O's attacking potential. In the variant, I have
moved a dead checker from O's 1pt to her 6pt. Although this of
course increases O's attacking potential, I wouldn't have expected
it to make as much difference as XG says it does.
I'm also not sure why 8/3 2/1 comes out slightly ahead of 8/7 8/3.
Probably one could figure it out by going through all of O's rolls.
But the more interesting question to me is how one could figure out
OTB to break anchor here but not in the variant. I have no good
answer to that.
1. Rollout¹ 21/16 2/1 eq:-0.357
Player: 39.77% (G:8.00% B:0.19%)
Opponent: 60.23% (G:15.63% B:0.36%)
Confidence: ±0.007 (-0.364..-0.350) - [100.0%]
2. Rollout¹ 8/3 2/1 eq:-0.477 (-0.120)
Player: 36.75% (G:5.08% B:0.07%)
Opponent: 63.25% (G:7.83% B:0.18%)
Confidence: ±0.009 (-0.486..-0.468) - [0.0%]
3. Rollout¹ 8/7 8/3 eq:-0.492 (-0.134)
Player: 36.58% (G:4.93% B:0.10%)
Opponent: 63.42% (G:8.38% B:0.19%)
Confidence: ±0.009 (-0.501..-0.483) - [0.0%]
4. Rollout¹ 8/2 eq:-0.499 (-0.142)
Player: 36.16% (G:4.86% B:0.07%)
Opponent: 63.84% (G:8.32% B:0.22%)
Confidence: ±0.009 (-0.508..-0.491) - [0.0%]
¹ 1296 Games rolled with Variance Reduction.
Dice Seed: 271828
Moves: 3-ply, cube decisions: XG Roller
eXtreme Gammon Version: 2.19.211.pre-release
-------
Variant
-------
XGID=-BCBBbB-B------b---dcBb-b-:0:0:1:51:0:0:0:0:10
Score is X:0 O:0. Unlimited Game
+13-14-15-16-17-18------19-20-21-22-23-24-+
| O | | O O X O O |
| O | | O O X O O |
| | | O O |
| | | O |
| | | |
| |BAR| |
| | | |
| | | |
| | | X |
| X | | X O X X X X |
| X | | X O X X X X |
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 92 O: 107 X-O: 0-0
Cube: 1
X to play 51
1. Rollout¹ 8/3 2/1 eq:-0.546
Player: 35.43% (G:4.85% B:0.08%)
Opponent: 64.57% (G:10.09% B:0.27%)
Confidence: ±0.010 (-0.556..-0.536) - [99.9%]
2. Rollout¹ 8/2 eq:-0.568 (-0.023)
Player: 34.85% (G:4.68% B:0.07%)
Opponent: 65.15% (G:10.48% B:0.29%)
Confidence: ±0.009 (-0.577..-0.560) - [0.0%]
3. Rollout¹ 21/16 2/1 eq:-0.571 (-0.025)
Player: 36.54% (G:8.21% B:0.19%)
Opponent: 63.46% (G:20.53% B:0.52%)
Confidence: ±0.010 (-0.581..-0.561) - [0.0%]
4. Rollout¹ 8/7 8/3 eq:-0.571 (-0.026)
Player: 35.31% (G:4.62% B:0.08%)
Opponent: 64.69% (G:10.70% B:0.29%)
Confidence: ±0.008 (-0.580..-0.563) - [0.0%]
¹ 1296 Games rolled with Variance Reduction.
Dice Seed: 271828
Moves: 3-ply, cube decisions: XG Roller
eXtreme Gammon Version: 2.19.211.pre-release
---
Tim Chow
The main take away should be that 16, 1 is correct. One can't play perfect backgammon but one can play pretty damn good backgammon. Making a small mistake as in the variant can at times not be avoided but making a blunder + usually is.
And on the surface it would appear to me that 3, 1 is better than 7, 3 because of how the opponent's two's would play on the upcoming roll. Basically we still have something to attack when the opponent rolls a one or a two if we stay on the 8pt. If we
move to the 7pt though then every roll not containing an ace either hits or hops two men past us completely.
Stick
--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)