A while ago, I asked for positions in money play where ND and D/T are
both optimal.

It looks like I may very well have hit on a position in an OTB game that
answers my own question! Although XG sees a difference by 0.001 or
0.002, it's quite likely to be some type of roundoff pasta, meaning that there's no difference at all.

I was very surprised at the take though. I remember being taught that
a closeout of two checkers on the bar is double and close take even
when the opponent has removed all 13 other checkers.
XG's remaining checkers on the deuce point, makes XG's bearoff much
slower. It seems that what I was told (in the early days of modern bg
theory) is just wrong.

XGID=bBBBCBB-------A--A-----b--:2:1:1:00:2:2:3:0:10
X:Daniel O:eXtremeGammon

Score is X:2 O:2. Unlimited Game, Jacoby Beaver
+13-14-15-16-17-18------19-20-21-22-23-24-+
| X X | | O |
| | | O |
| | | |
| | | |
| | | |
| |BAR| |
| | O | |
| | O | |
| | | X | +---+
| | | X X X X X X | | 4 |
| | | X X X X X X | +---+
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 77 O: 54 X-O: 2-2
Cube: 4, X own cube
X on roll, cube action

Analyzed in 4-ply
Player Winning Chances: 78.43% (G:0.00% B:0.00%)
Opponent Winning Chances: 21.57% (G:0.02% B:0.00%)

Cubeless Equities: No Double=+0.568, Double=+1.137

Cubeful Equities:
No redouble: +0.984 (-0.002)
Redouble/Take: +0.985
Redouble/Pass: +1.000 (+0.015)

Best Cube action: Redouble / Take

eXtreme Gammon Version: 2.10

On 5/15/2022 7:54 AM, peps...@gmail.com wrote:
> A while ago, I asked for positions in money play where ND and D/T are
> both optimal.
>
> It looks like I may very well have hit on a position in an OTB game that
> answers my own question! Although XG sees a difference by 0.001 or
> 0.002, it's quite likely to be some type of roundoff pasta, meaning that there's no difference at all.
>
> I was very surprised at the take though. I remember being taught that
> a closeout of two checkers on the bar is double and close take even
> when the opponent has removed all 13 other checkers.
> XG's remaining checkers on the deuce point, makes XG's bearoff much
> slower. It seems that what I was told (in the early days of modern bg
> theory) is just wrong.

This position rolls out to a pass. But should we trust the rollout?

XGID=bBBBCBB-------A--A-----b--:1:1:1:00:0:0:0:0:10

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 |
| | | |
| | | |
| | | |
| |BAR| |
| | O | |
| | O | |
| | | X | +---+
| | | X X X X X X | | 2 |
| | | X X X X X X | +---+
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 77 O: 54 X-O: 0-0
Cube: 2, X own cube
X on roll, cube action

Analyzed in Rollout
No redouble
Player Winning Chances: 78.92% (G:0.00% B:0.00%)
Opponent Winning Chances: 21.08% (G:0.01% B:0.00%)
Redouble/Take
Player Winning Chances: 79.15% (G:0.00% B:0.00%)
Opponent Winning Chances: 20.85% (G:0.00% B:0.00%)

Cubeful Equities:
No redouble: +0.998 (-0.002)
Redouble/Take: +1.029 (+0.029)
Redouble/Pass: +1.000

Best Cube action: Redouble / Pass

Rollout:
1296 Games rolled with Variance Reduction.
Dice Seed: 271828
Moves: 4-ply, cube decisions: XG Roller+
Search interval: Large
Confidence No Double: ± 0.002 (+0.995..+1.000)
Confidence Double: ± 0.009 (+1.020..+1.038)

eXtreme Gammon Version: 2.19.207.pre-release

---
Tim Chow

On Sunday, May 15, 2022 at 8:07:36 PM UTC+1, Tim Chow wrote:
> On 5/15/2022 7:54 AM, peps...@gmail.com wrote:
> > A while ago, I asked for positions in money play where ND and D/T are
> > both optimal.
> >
> > It looks like I may very well have hit on a position in an OTB game that
> > answers my own question! Although XG sees a difference by 0.001 or
> > 0.002, it's quite likely to be some type of roundoff pasta, meaning that there's no difference at all.
> >
> > I was very surprised at the take though. I remember being taught that
> > a closeout of two checkers on the bar is double and close take even
> > when the opponent has removed all 13 other checkers.
> > XG's remaining checkers on the deuce point, makes XG's bearoff much
> > slower. It seems that what I was told (in the early days of modern bg
> > theory) is just wrong.
> This position rolls out to a pass. But should we trust the rollout?
>
>
> XGID=bBBBCBB-------A--A-----b--:1:1:1:00:0:0:0:0:10
>
> 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 |
> | | | |
> | | | |
> | | | |
> | |BAR| |
> | | O | |
> | | O | |
> | | | X | +---+
> | | | X X X X X X | | 2 |
> | | | X X X X X X | +---+
> +12-11-10--9--8--7-------6--5--4--3--2--1-+
> Pip count X: 77 O: 54 X-O: 0-0
> Cube: 2, X own cube
> X on roll, cube action
> Analyzed in Rollout
> No redouble
> Player Winning Chances: 78.92% (G:0.00% B:0.00%)
> Opponent Winning Chances: 21.08% (G:0.01% B:0.00%)
> Redouble/Take
> Player Winning Chances: 79.15% (G:0.00% B:0.00%)
> Opponent Winning Chances: 20.85% (G:0.00% B:0.00%)
>
> Cubeful Equities:
> No redouble: +0.998 (-0.002)
> Redouble/Take: +1.029 (+0.029)
> Redouble/Pass: +1.000
>
> Best Cube action: Redouble / Pass
>
> Rollout:
> 1296 Games rolled with Variance Reduction.
> Dice Seed: 271828
> Moves: 4-ply, cube decisions: XG Roller+
> Search interval: Large
> Confidence No Double: ± 0.002 (+0.995..+1.000)
> Confidence Double: ± 0.009 (+1.020..+1.038)

Yes, I think we should trust the rollout.
I trust Rolly Rollout about as much as I trust Roland Browning.
Thanks for rolling it out.

Paul

On Sunday, May 15, 2022 at 8:07:36 PM UTC+1, Tim Chow wrote:
> On 5/15/2022 7:54 AM, peps...@gmail.com wrote:
> > A while ago, I asked for positions in money play where ND and D/T are
> > both optimal.
> >
> > It looks like I may very well have hit on a position in an OTB game that
> > answers my own question! Although XG sees a difference by 0.001 or
> > 0.002, it's quite likely to be some type of roundoff pasta, meaning that there's no difference at all.
> >
> > I was very surprised at the take though. I remember being taught that
> > a closeout of two checkers on the bar is double and close take even
> > when the opponent has removed all 13 other checkers.
> > XG's remaining checkers on the deuce point, makes XG's bearoff much
> > slower. It seems that what I was told (in the early days of modern bg
> > theory) is just wrong.
> This position rolls out to a pass. But should we trust the rollout?
>
>
> XGID=bBBBCBB-------A--A-----b--:1:1:1:00:0:0:0:0:10
>
> 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 |
> | | | |
> | | | |
> | | | |
> | |BAR| |
> | | O | |
> | | O | |
> | | | X | +---+
> | | | X X X X X X | | 2 |
> | | | X X X X X X | +---+
> +12-11-10--9--8--7-------6--5--4--3--2--1-+
> Pip count X: 77 O: 54 X-O: 0-0
> Cube: 2, X own cube
> X on roll, cube action
> Analyzed in Rollout
> No redouble
> Player Winning Chances: 78.92% (G:0.00% B:0.00%)
> Opponent Winning Chances: 21.08% (G:0.01% B:0.00%)
> Redouble/Take
> Player Winning Chances: 79.15% (G:0.00% B:0.00%)
> Opponent Winning Chances: 20.85% (G:0.00% B:0.00%)
>
> Cubeful Equities:
> No redouble: +0.998 (-0.002)
> Redouble/Take: +1.029 (+0.029)
> Redouble/Pass: +1.000
>
> Best Cube action: Redouble / Pass
>
> Rollout:
> 1296 Games rolled with Variance Reduction.
> Dice Seed: 271828
> Moves: 4-ply, cube decisions: XG Roller+
> Search interval: Large
> Confidence No Double: ± 0.002 (+0.995..+1.000)
> Confidence Double: ± 0.009 (+1.020..+1.038)
>
> eXtreme Gammon Version: 2.19.207.pre-release
>
> ---
> Tim Chow

What should not be taken literally is the verdict that not doubling loses 0..002 equity.
Of course not. If you wait a roll and double, it will still be a pass.
But XG's deductions of 0.001 and 0.002 are sometimes bogus. We talked about this
on another thread.

Paul