Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
MTG Gameplay on MTG Domain Ramp (in-distribution opponent pool)
Loading...
20.6
Win Rate
PPO
11.864
14.132
16.4
18.668
May 7, 2026
Win Rate
95% CI Value
Difference (pp)
p-value (Bootstrap)
p-value (Holm)
Updated 26d ago
Evaluation Results
Method
Method
Links
Win Rate
95% CI Value
Difference (pp)
p-value (Bootstrap)
p-value (Holm)
PPO
ns=2
2026.05
20.6
9.4
-
-
-
CGFA-PPO
ns=2
2026.05
12.2
5
8.3
0.0001
0.0005
Feedback
Search any
task
Search any
task
