          
 
        
 | | White is Player 2
score: 0
pip: 44 | | 13 point match | pip: 52
score: 0
Blue is Player 1 |
|
| XGID=-BECBB-------A-----bacbbe-:0:0:1:51:0:0:0:13:10 |
| Blue to play 51 |
1. | Rollout1 | 13/8 2/1 | eq: -0,589 |
| Player:
Opponent: | 31,33% (G:0,00% B:0,00%)
68,67% (G:0,00% B:0,00%) | Conf.: ± 0,003 (-0,592...-0,586) - [100,0%]
Duration: 1 minute 03 seconds |
|
| 2. | Rollout1 | 13/8 5/4 | eq: -0,610 (-0,021) |
| Player:
Opponent: | 30,48% (G:0,00% B:0,00%)
69,52% (G:0,00% B:0,00%) | Conf.: ± 0,003 (-0,613...-0,607) - [0,0%]
Duration: 48,9 seconds |
|
| 3. | Rollout1 | 13/7 | eq: -0,611 (-0,022) |
| Player:
Opponent: | 30,58% (G:0,00% B:0,00%)
69,42% (G:0,00% B:0,00%) | Conf.: ± 0,003 (-0,613...-0,608) - [0,0%]
Duration: 45,9 seconds |
|
| |
1 2592 Games rolled with Variance Reduction.
Moves: 3-ply, cube decisions: XG Roller
|
eXtreme Gammon Version: 2.19.208.pre-release, MET: Kazaross XG2
OTB I figured that 2/1 raise the wastage too much, so the ace is somehow lost. It seems confirmed by Keith Count I typically use (with some modifications) in these medium races.
So I was choosing between 8/7 and 5/4. I liked the structure and 44 after 5/4 more, so I discarded 8/7 because the only advantage over 5/4 I could think of is that it guards against one more ace, but only if it's rolled on the next turn. Didn't seem impactful, especially when I'm loosing the race, and rolling couple of aces would lose me the game most of the time anyway.
Lesson for me is that it's just plain overthinking. Just smoothen the structure by unloading highest stack, period. Don't climb checkers if You have stacks.
Below are two variants to show when 2/1 gets worse. Opp is adjusted by couple of pips too, to put the race in similar relative position. Small stack on higher point is more important to unload than bigger stack on low pt.
Variant A
Compared to the original position, player on roll is adjusted 3/5.
5/4 and 2/1 are too close to call here.
| | White is Player 2
score: 0
pip: 50 | | 13 point match | pip: 54
score: 0
Blue is Player 1 |
|
| XGID=-BEBBC-------A-----bbccbc-:0:0:1:51:0:0:0:13:10 |
| Blue to play 51 |
| 1. | Rollout1 | 13/8 5/4 | eq: -0,553 |
| Player:
Opponent: | 32,22% (G:0,00% B:0,00%)
67,78% (G:0,00% B:0,00%) | Conf.: ± 0,003 (-0,556...-0,550) - [72,9%]
Duration: 1 minute 26 seconds |
|
| 2. | Rollout1 | 13/8 2/1 | eq: -0,554 (-0,001) |
| Player:
Opponent: | 32,41% (G:0,00% B:0,00%)
67,59% (G:0,00% B:0,00%) | Conf.: ± 0,003 (-0,557...-0,552) - [27,1%]
Duration: 1 minute 31 seconds |
|
| 3. | Rollout1 | 13/7 | eq: -0,564 (-0,011) |
| Player:
Opponent: | 31,98% (G:0,00% B:0,00%)
68,02% (G:0,00% B:0,00%) | Conf.: ± 0,003 (-0,567...-0,561) - [0,0%]
Duration: 1 minute 05 seconds |
|
| |
1 2592 Games rolled with Variance Reduction.
Moves: 3-ply, cube decisions: XG Roller
|
eXtreme Gammon Version: 2.19.208.pre-release, MET: Kazaross XG2
Variant B
Compared to Variant A, player on roll is adjusted 2/6.
5/4 gets ahead.
Interesting note is that in none of these positions 8/7 is better than 5/4.
 
| | White is Player 2
score: 0
pip: 53 | | 13 point match | pip: 58
score: 0
Blue is Player 1 |
|
| XGID=-BDBBCA------A-----bbdcbb-:0:0:1:51:0:0:0:13:10 |
| Blue to play 51 |
| 1. | Rollout1 | 13/8 5/4 | eq: -0,554 |
| Player:
Opponent: | 32,14% (G:0,00% B:0,00%)
67,86% (G:0,00% B:0,00%) | Conf.: ± 0,004 (-0,558...-0,550) - [100,0%]
Duration: 1 minute 01 second |
|
| 2. | Rollout1 | 13/7 | eq: -0,566 (-0,012) |
| Player:
Opponent: | 31,81% (G:0,00% B:0,00%)
68,19% (G:0,00% B:0,00%) | Conf.: ± 0,004 (-0,570...-0,562) - [0,0%]
Duration: 44,1 seconds |
|
| 3. | Rollout1 | 13/8 2/1 | eq: -0,571 (-0,017) |
| Player:
Opponent: | 31,83% (G:0,00% B:0,00%)
68,17% (G:0,00% B:0,00%) | Conf.: ± 0,004 (-0,575...-0,567) - [0,0%]
Duration: 59,7 seconds |
|
| |
1 1296 Games rolled with Variance Reduction.
Moves: 3-ply, cube decisions: XG Roller
|
eXtreme Gammon Version: 2.19.208.pre-release, MET: Kazaross XG2