While it turns out that according to a standard rollout there's not much to it as long as you pick one of the top 4 or 5 plays, the bot's opinion could change with a higher level rollout. (not only playing strength but move filter increase in a position such as this) I'm not that interested in it to devote any CPU time to it.
Turns out 7/3* 3/1* doesn't make it past XG's 2 ply move filter and appears as a sizeable blunder until you manually force it through. Even then on XGR+ it's behind almost .050 and for ++ it's still almost .040 back. This seemed the natural play to me at the time, or at least the play I knew how to follow up the easiest. The plus side is that barring doubles the opponent won't be anchoring any time soon and we have ten men in the zone so a reasonable chance of carrying out this blitz and maybe even scooping up all six of the opponent's checkers in the process. Downside is the opponent does have an okay board meaning leaving a direct shot and getting hit could seriously stall our attack or even turn it around if we find a fan to us being the one with potentially nine! checkers on the bar.
It's also interesting to see how the other plays rank because I can't imagine a human playing 16* 6 even though it is currently tied as the best play.
         
 
          
 | | White is Player 2
score: 0
pip: 154 | Unlimited Game
Jacoby Beaver | pip: 149
score: 0
Blue is Player 1 |
|
| XGID=aa-aBBBBAAA-b-AaaaAc-bAbA-:1:-1:1:42:0:0:3:0:10 |
| Blue to play 42 |
| 1. | Rollout1 | 7/3* 3/1* | eq: +1.036 |
| Player:
Opponent: | 76.03% (G:56.64% B:16.47%)
23.97% (G:9.79% B:2.89%) | Conf.: ± 0.009 (+1.027...+1.045) - [56.1%]
Duration: 2 minutes 53 seconds |
|
| 2. | Rollout1 | 18/16* 10/6 | eq: +1.035 (-0.001) |
| Player:
Opponent: | 77.69% (G:51.53% B:12.47%)
22.31% (G:5.62% B:0.80%) | Conf.: ± 0.007 (+1.028...+1.042) - [41.8%]
Duration: 2 minutes 44 seconds |
|
| 3. | Rollout1 | 24/20 18/16* | eq: +1.025 (-0.011) |
| Player:
Opponent: | 78.67% (G:49.85% B:10.49%)
21.33% (G:5.30% B:0.91%) | Conf.: ± 0.007 (+1.019...+1.032) - [0.4%]
Duration: 3 minutes 05 seconds |
|
| 4. | Rollout2 | 18/16* 14/10 | eq: +1.022 (-0.014) |
| Player:
Opponent: | 78.23% (G:49.82% B:10.52%)
21.77% (G:4.65% B:0.55%) | Conf.: ± 0.013 (+1.009...+1.035) - [1.8%]
Duration: 55.7 seconds |
|
| 5. | Rollout1 | 18/16* 7/3* | eq: +1.008 (-0.028) |
| Player:
Opponent: | 76.39% (G:53.08% B:15.22%)
23.61% (G:8.27% B:2.44%) | Conf.: ± 0.009 (+0.999...+1.017) - [0.0%]
Duration: 2 minutes 48 seconds |
|
| 6. | Rollout2 | 18/16* 9/5 | eq: +1.004 (-0.032) |
| Player:
Opponent: | 77.34% (G:50.60% B:10.43%)
22.66% (G:5.48% B:0.79%) | Conf.: ± 0.013 (+0.991...+1.017) - [0.0%]
Duration: 53.2 seconds |
|
| 7. | Rollout2 | 22/16* | eq: +0.989 (-0.047) |
| Player:
Opponent: | 78.29% (G:48.39% B:9.86%)
21.71% (G:6.00% B:1.13%) | Conf.: ± 0.013 (+0.975...+1.002) - [0.0%]
Duration: 38.1 seconds |
|
| |
1 5184 Games rolled with Variance Reduction.
Moves: 3-ply, cube decisions: XG Roller
2 1296 Games rolled with Variance Reduction.
Moves: 3-ply, cube decisions: XG Roller
|