| Books |
From Backgammon, the Cruelest Game, by Barclay Cooke and Jon Bradshaw
|
8
MATCH TO 15 14 |
|
Should White double to 2? |
In both tournaments and money games, black would tend to drop, but given the score — that is, 14 to any lesser even number — black should take every time.
The subtle logic here is that since white is at 8, he needs four consecutive victories to win the match — excluding double games and assuming that it is legal for him to double and that he does so in each successive game after the opening roll. But if he were at 9 instead of at 8, he would need only three victories to win the match.
Therefore, since being at 9 is virtually as good as being at 10, white should double, hoping to bluff black into dropping. In over 90 percent of such cases the bluff would work. If white accomplishes this, he has, in effect, won a game by default and increased his chances of winning the match because he now needs only three victories instead of four.
On the other hand, in this instance black should take the double, regardless of the hopelessness of his position — unless, of course, there were any danger of losing a gammon. Should black accept the double and lose, the score will now be 14–10 in his favor, which is barely worse than 14–9. The point being that the man who is behind is always trying to get from even to odd, and this is what black is trying to obstruct. Since the comparative values of 9 and 10 are almost the same, black accepts what appears to be an insane double (which in all money games he would drop) because if he loses this game he has lost nothing extra. and if he should win, he wins the match.
Rollout
 Tom Keith 2013 |
|
Match to 15 White 8, Black 14 Centered cube White on roll 1296 games with VR Checker play: 3-ply Cube play: XG Roller |
| Cube Action | Game | G | BG | Equity | |||||
| No double |
W L |
.8306 .1694 |
.0417 .0077 |
.0010 .0004 | +0.8107 | (0.0002) | |||
| Double | Take |
W L |
.8307 .1693 |
.0418 .0075 |
.0010 .0004 | +0.8109 | +0.8109 |
|
|
| Drop | +1.0000 | ||||||||
|
|
