what drives home wins?
← lab · AUC 0.673 (real signal) · ran 7/3/2026
What this is: Asks which pre-match factors drive one specific outcome, using a walk-forward model and permutation importance.
| Factor | Importance | Direction | Survives all eras? |
|---|
| home__team.elo | 0.0512 | ↑ +0.262 | ⏳ yes |
| away__team.elo | 0.0208 | ↓ -0.242 | ⏳ yes |
| match.rainfall_mm | 0.0039 | ↓ -0.006 | no |
| home__team.goals_against_avg_l5 | 0.0029 | ↓ -0.137 | ⏳ yes |
| away__team.corners_against_avg_l5 | 0.0028 | ↑ +0.140 | ⏳ yes |
| away__team.form_points_l5 | 0.0024 | ↓ -0.155 | ⏳ yes |
| home__team.shots_for_avg_l5 | 0.0019 | ↑ +0.179 | ⏳ yes |
| home__team.goals_for_avg_l5 | 0.0016 | ↑ +0.154 | ⏳ yes |
| away__team.shots_for_avg_l5 | 0.0015 | ↓ -0.170 | ⏳ yes |
| home__team.corners_against_avg_l5 | 0.0006 | ↓ -0.143 | ⏳ yes |
| away__team.goals_for_avg_l5 | 0.0002 | ↓ -0.149 | ⏳ yes |
| away__team.corners_for_avg_l5 | -0.0006 | ↓ -0.123 | ⏳ yes |
Reading the columnswhat each number actually means
| AUC | predictability: 0.50 = coin flip, ~0.70 = ceiling for sports |
| Importance | how much the model leans on this factor (permutation importance) |
| Direction | sign of the raw correlation with the outcome |
| Survives all eras | effect points the same way in every historical era |
Spec · the reproducible recipe
{
"name": "what drives home wins?",
"sport": "football",
"target": {
"equals": "H",
"metric": "match.result"
},
"features": "all"
}