Just the two rather obvious options this time.
Spoiler
The four point and the bar point are roughly equal in value, but making the bar covers the blot and makes a solid four prime. That should be better than making the four.
Spoiler
Given the vote, I'm surprised that GNU 3-ply decided it was wafer-thin and gave the edge to making the four. Although the 6s to hit and cover are duplicated, a third man back would be a real downer!
rollout
Spoiler
Interestingly, XG thinks making the bar point is a near blunder (.075) Post-hoc, the only thing I can offer up by way of explanation is that we're gong to need to get our runners out of there and when we do they will be exposed to attack. This will likely result in an exchange of hits and having a 3-point board better positions us to that eventuality.
XGID=-b---BCAD---cC--ab-bb-baB-:0:0:1:42:2:0:0:5:3
X:The Herd O:COsborne
Score is X:2 O:0 5 pt.(s) match.
+13-14-15-16-17-18------19-20-21-22-23-24-+
| X O O | | O O O O X |
| X O | | O O O X |
| X | | |
| | | |
| | | |
| |BAR| |
| | | |
| X | | |
| O X | | X |
| O X | | X X O |
| O X X | | X X O |
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 154 O: 142 X-O: 2-0/5
Cube: 1
X to play 42
1. Rollout¹ 8/4 6/4 eq:+0.191
Player: 58.03% (G:19.74% B:2.44%)
Opponent: 41.97% (G:14.99% B:1.19%)
Confidence: ±0.013 (+0.178..+0.205) - [100.0%]
Duration: 10 minutes 05 seconds
2. Rollout¹ 13/7 eq:+0.117 (-0.075)
Player: 56.43% (G:18.84% B:1.99%)
Opponent: 43.57% (G:15.15% B:0.98%)
Confidence: ±0.016 (+0.101..+0.133) - [0.0%]
Duration: 10 minutes 03 seconds
¹ 1296 Games rolled with Variance Reduction.
Moves: 3-ply, cube decisions: XG Roller
eXtreme Gammon Version: 2.10, MET: Kazaross XG2