I did a rollout at 2592 trials and was very surprised at GNU's analysis of the position.
The way I see it, I have a clear double and white has one of those takes that you hate to take, but probably should. GNU marks my double as "very bad" and that white should take. I'm pleased that my opponent passed the cube, but I would like to gain a better understanding of the reasoning behind the analysis.
My main obstacle is obviously bringing my midpoint checkers in safely, and then I will have to clear out my outfield points before bearing off.
I have five spare checkers to dump on the 7 and 8 point and another on the 6 point. White will have to drop his midpoint very soon, making it easier for me to clear my midpoint, as I can leave an indirect shot if need be. I don't see clearing my midpoint as a big problem most of the time.
Once I do clear my midpoint, I certainly like my chances against the 4 point anchor. Furthermore, white's board rates to collapse if things go wrong during my bearoff.
I chose to cube at this point because for every roll where I do not safely bring down my midpoint, I have less and less reason to double. If I get a good double roll like 22, 44, 55, 66, or a 65/56, I missed my chance to double.
I rate to win this match, and I want to throw the cube. I'm willing to wait but I just don't see a better opportunity to double than right now.
Can someone help me understand what I am missing?