Again, this wasn't a decision I personally faced OTB; I was playing XG Roller+ and it had to play the snake eyes. It played bar/24 6/5 2/1*(2).
However, when I ran the match through World Class analysis, XGR+ charged itself with a whopper here!
This rather odd circumstance happened because XG 3-ply likes 6/5, and at least on the settings I use, XG Roller+ is apparently satisfied with the 3-ply evaluation and doesn't bother running the play through a stronger analysis level. However, if you force both 6/5 and 24/23 through XGR+ then XGR+ thinks that 6/5 is a whopper.
GNU 2-ply slightly prefers 24/23 though it doesn't think 6/5 is that far behind.
The rollout below says that 6/5 is a serious mistake but not in whopper territory. I personally would have played 6/5, underestimating the importance of diversifying escaping numbers. I haven't checked what happens if you clean up White's outfield blots but I suspect that that will make the decision closer; part of the motivation for escaping quickly is to vacuum up more blots.
Sorry I didn't roll out any other plays but 2/1*(2) looks clear here, especially with White's vulnerable outfield blots.
                        |
| ||||
| XGID=-aBB-BD-B--aa-a--c-ccbB--A:0:0:1:11:0:0:1:0:10 | |||||
| Blue to play 11 | |||||
| 1. | Rollout1 | Bar/23 2/1*(2) | eq: +0.315 | |||
| ||||||
| 2. | Rollout1 | Bar/24 6/5 2/1*(2) | eq: +0.253 (-0.062) | |||
| ||||||
| 1 2592 Games rolled with Variance Reduction. Dice Seed: 2 First 3 moves and cube decisions: 4 ply Remaining moves and cube decisions: 3 ply | ||||||
eXtreme Gammon Version: 1.21