I watched computer play several games at DMP and it does some strange stuff. The values you got for DMP seem to accurately reflect how often XG does win. At money it wins more becuse the fear of the gammon prevents some of these plays
      
 
         
 | | White is Player 2
score: 0
pip: 21 | | 1 point match | pip: 139
score: 0
Blue is Player 1 |
|
| XGID=----aBBBDB--A------A--A---:0:0:1:51:0:0:0:1:10 |
| Blue to play 51 |
| 1. | 4-ply | 22/17 6/5 | eq: -0.8814 |
| Player:
Opponent: | 5.93% (G:0.00% B:0.00%)
94.07% (G:37.16% B:21.07%) | |
|
| 2. | 4-ply | 19/14 6/5 | eq: -0.8821 (-0.0007) |
| Player:
Opponent: | 5.90% (G:0.00% B:0.00%)
94.10% (G:40.11% B:22.50%) | |
|
| 3. | 4-ply | 12/7 6/5 | eq: -0.8825 (-0.0011) |
| Player:
Opponent: | 5.88% (G:0.00% B:0.00%)
94.12% (G:41.68% B:24.32%) | |
|
| 4. | 4-ply | 22/21 6/1 | eq: -0.8989 (-0.0175) |
| Player:
Opponent: | 5.06% (G:0.00% B:0.00%)
94.94% (G:46.80% B:27.43%) | |
|
| 5. | 3-ply | 22/17 5/4* | eq: -0.8597 (+0.0217) |
| Player:
Opponent: | 7.01% (G:0.00% B:0.00%)
92.99% (G:31.11% B:17.77%) | |
|
eXtreme Gammon Version: 2.10, MET: Kazaross XG2
| | White is Player 2
score: 0
pip: 21 | | Unlimited Game | pip: 139
score: 0
Blue is Player 1 |
|
| XGID=----aBBBDB--A------A--A---:1:1:1:51:0:0:0:0:10 |
| Blue to play 51 |
| 1. | 4-ply | 22/17 19/18 | eq: -1.1049 |
| Player:
Opponent: | 5.66% (G:0.00% B:0.00%)
94.34% (G:19.74% B:4.06%) | |
|
| 2. | 4-ply | 19/18 9/4* | eq: -1.1096 (-0.0047) |
| Player:
Opponent: | 5.73% (G:0.00% B:0.00%)
94.27% (G:16.29% B:8.15%) | |
|
| 3. | 4-ply | 22/21 9/4* | eq: -1.1173 (-0.0124) |
| Player:
Opponent: | 5.61% (G:0.00% B:0.00%)
94.39% (G:16.16% B:8.77%) | |
|
| 4. | 4-ply | 9/4* 8/7 | eq: -1.1233 (-0.0184) |
| Player:
Opponent: | 5.55% (G:0.00% B:0.00%)
94.45% (G:16.39% B:9.00%) | |
|
| 5. | 4-ply | 22/16 | eq: -1.1390 (-0.0341) |
| Player:
Opponent: | 5.79% (G:0.00% B:0.00%)
94.21% (G:20.05% B:7.47%) | |
|
eXtreme Gammon Version: 2.10