| |
BGonline.org Forums
Rollout (I made the same mistake as Keene)
Posted By: Maik Stiebler In Response To: Rollout (I made the same mistake as Keene) (Ray Kershaw)
Date: Tuesday, 5 July 2016, at 8:56 a.m.
Because after the safe play, there is NO chance of doubling White out and their take will be easy, not close, most of the time. After paying now, there will be a good chance of doubling White out next time, and it will often be a quite efficient cash, with no huge market loss involved.
If we know that the plays would be quite close if White already held the cube, we could look at it this way: We know that the plays would still be close if it went double/take before the next roll every single time. But contrary to that assumption, next roll White can in some cases do better than that by dropping Blue's cube. This is Blue's market loss. So Blue should make the play that avoids market loss. But on the other side of the scale, Blue can do better than our assumption by not doubling in some cases. Thus, while those are obviously bad for Blue, if both plays are close if White already held the cube, Blue should make the play that leads to these situations.
After 12/1, there are 6 hits, each of which leave Blue not good enough to double, with doubling being an 1.070 whopper. So from these variations Blue is 6*1.070 / 36 = 0.177 better off with the centered cube after 12/1 than with White owning the cube on two. Add a few more millipoints for the no-doubles after 55 and 66. On the other hand, total market loss after 12/1 amounts to 1.6 / 36 or 0.045 (I checked with XG). Subtracting that from the gain from not doubling still leaves about 0.140. After 8/3 8/2 there is no market loss and some small gain from not doubling after high doubles, amounting to 0.568/36 = 0.016 according to XG. So 12/1 gains 0.124 relative to 8/3 8/2 with the centered cube as compared to White owning the cube on two, which is very close to what the rollouts say.
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.