According to Iancho (who has more info about everyone's PR than anyone else), Fernando Braconi of Italy plays with one of the two best PRs on the planet.
Like Matt, however, his bear-in technique is not always perfect. I present my analysis below the position and rollout result posted here. Fernando played 8/6 8/3.
|
| |||
| XGID=-BCBBBCA---------babcabbb-:1:-1:-1:52:7:7:0:11:10 | ||||
 to play 52 | ||||
is KillingKid | 1. | Rollout1 | 8/3 7/5 | eq: -0.201 | |||
| ||||||
| 2. | Rollout1 | 8/6 8/3 | eq: -0.213 (-0.012) | |||
| ||||||
| 1 10368 Games rolled with Variance Reduction. Moves and cube decisions: 3 ply | ||||||
eXtreme Gammon Version: 1.21, MET: Rockwell-Kazaross
8/6 8/3 may look smoother because it evens out the number of checkers on the 6pt and 5pt, putting three checkers on each.
However, 8/3 7/5 is better for a subtle reason. Black's 4s plays better next roll after 8/3 7/5. With 44, Black can rip off two instead of one. With other 4s, Black plays 8/4 and then takes one off, leaving him with two checkers on his 4pt instead of only one on his 4pt after 8/6 8/3. Better to end up with two checkers on your 4pt instead of only one in this position, to avoid missing with 4s later.
Additionally, Black's 55 plays better after 8/3 7/5.
Admittedly, this error is tiny. I wish the most equity I ever gave up with a checker play was 0.012.