44 to play

[Timothy Chow]

XGID=----a-EbD---bD--Bbacbb----:0:0:1:44:0:0:0:0:10

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 | | O O O |
| X X O | | O O O |
| X | | O |
| X | | |
| | | |
| |BAR| |
| | | X |
| X | | X |
| X | | X |
| O X O | | X |
| O X O | | X O |
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 146 O: 142 X-O: 0-0
Cube: 1
X to play 44

---
Tim Chow

[atila...@gmail.com]

11 Mayıs 2022 Çarşamba tarihinde saat 16:24:46 UTC+3 itibarıyla Tim Chow şunları yazdı:

At first glance, I focused on three moves. But all three are wrong. Let's face it, it's a provocative position to attack.

Meanwhile, if the cube had been turned, the errors would be as low as 20 per thousand.

I guess the key is the oppenent's strong homeboard and accordingly which game plan X will choose? Race, attack or prime?

To understand this, I shifted the opponent's two checkers from 21 to 23 and gave her a weaker homeboard structure that was prone to attack.

Both 13/9(4) and 13/9(2), 8/4*(2) keep all three game plans alive. But weakening the opponent's homeboard is the only thing that will compensate for the possibility of leaving a blot after 13/9(2), 8/4*(2)

Variant
XGID=----a-EbD---bD--Bbacb--b--:0:0:1:44:0:0:3:0:10

X:Player 1 O:Player 2

Score is X:0 O:0. Unlimited Game, Jacoby Beaver

+13-14-15-16-17-18------19-20-21-22-23-24-+
| X X O O | | O O O |
| X X O | | O O O |
| X | | O |
| X | | |
| | | |
| |BAR| |
| | | X |
| X | | X |
| X | | X |
| O X O | | X |
| O X O | | X O |
+12-11-10--9--8--7-------6--5--4--3--2--1-+

Pip count X: 146 O: 138 X-O: 0-0

Cube: 1
X to play 44

1. Rollout¹ 13/9(4) eq:+0,256
Player: 57,65% (G:3,91% B:0,12%)
Opponent: 42,35% (G:2,91% B:0,09%)
Confidence: ±0,007 (+0,249..+0,263) - [87,4%]
Duration: 26 minutes 15 seconds

2. Rollout¹ 13/9(2) 8/4*(2) eq:+0,249 (-0,006)
Player: 57,25% (G:9,30% B:0,39%)
Opponent: 42,75% (G:8,66% B:0,23%)
Confidence: ±0,008 (+0,241..+0,257) - [12,6%]
Duration: 24 minutes 29 seconds

3. Rollout² 8/4*(2) 6/2(2) eq:+0,189 (-0,066)
Player: 55,21% (G:10,97% B:0,52%)
Opponent: 44,79% (G:10,27% B:0,28%)
Confidence: ±0,020 (+0,169..+0,210) - [0,0%]
Duration: 3 minutes 00 second

4. Rollout² 13/5(2) eq:+0,177 (-0,078)
Player: 54,96% (G:6,26% B:0,27%)
Opponent: 45,04% (G:6,21% B:0,15%)
Confidence: ±0,019 (+0,158..+0,197) - [0,0%]
Duration: 3 minutes 18 seconds

5. 3-ply 13/9(2) 6/2(2) eq:+0,143 (-0,113)
Player: 54,95% (G:6,90% B:0,19%)
Opponent: 45,05% (G:7,35% B:0,07%)


¹ 2592 Games rolled with Variance Reduction.
Moves: 3-ply, cube decisions: XG Roller

² 324 Games rolled with Variance Reduction.
Moves: 3-ply, cube decisions: XG Roller


eXtreme Gammon Version: 2.10

[Timothy Chow]

It's tempting to PoH here, or perhaps make the 5pt. At gammon go,
XG does play 13/9(2) 8/4*(2). However, there's another strategy
available, which can be hard to see if you're not used to looking
for it: 13/9(4)! The idea is that X is ahead in the race and so
should keep open the racing game plan. Attacking is committal,
leaving X somewhat stripped, and vulnerable to getting hit back
within the next few turns. 13/9(4) does not rule out attacking
and priming in the future if the dice allow it, but carries less
risk, and gives X the best chance of winning just by racing.

I don't understand atilamalcok's comment that if the cube were
turned then the errors would be as low as 20 per thousand, because
that's not what XG tells me. But I like the variant, which both
decreases X's racing advantage and weakens O's counterattacking
chances.

1. Rollout¹ 13/9(4) eq:+0.345
Player: 60.52% (G:3.99% B:0.10%)
Opponent: 39.48% (G:2.60% B:0.08%)
Confidence: ±0.009 (+0.335..+0.354) - [100.0%]

2. Rollout¹ 13/5(2) eq:+0.205 (-0.140)
Player: 56.20% (G:5.55% B:0.22%)
Opponent: 43.80% (G:5.94% B:0.18%)
Confidence: ±0.011 (+0.194..+0.215) - [0.0%]

3. Rollout¹ 13/9(2) 8/4*(2) eq:+0.201 (-0.144)
Player: 55.86% (G:10.48% B:0.44%)
Opponent: 44.14% (G:10.70% B:0.33%)
Confidence: ±0.013 (+0.188..+0.213) - [0.0%]

4. Rollout¹ 13/9(2) 6/2(2) eq:+0.172 (-0.173)
Player: 54.96% (G:8.15% B:0.25%)
Opponent: 45.04% (G:9.31% B:0.23%)
Confidence: ±0.011 (+0.161..+0.183) - [0.0%]

5. Rollout¹ 8/4*(2) 6/2(2) eq:+0.144 (-0.201)
Player: 54.08% (G:10.69% B:0.41%)
Opponent: 45.92% (G:11.21% B:0.32%)
Confidence: ±0.012 (+0.132..+0.156) - [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