what drives 10+ corners?
← lab · AUC 0.505 (no real pattern — honest result) · 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.corners_for_avg_l5 | 0.0050 | ↑ +0.040 | ⏳ yes |
| home__team.corners_against_avg_l5 | 0.0043 | ↑ +0.018 | no |
| away__team.elo | 0.0041 | ↓ -0.057 | ⏳ yes |
| away__team.corners_against_avg_l5 | 0.0035 | ↑ +0.044 | ⏳ yes |
| away__team.season_ppg | 0.0019 | ↓ -0.030 | ⏳ yes |
| away__team.shots_for_avg_l5 | 0.0017 | ↓ -0.036 | no |
| home__team.season_ppg | 0.0013 | ↑ +0.014 | no |
| match.humidity_pct | 0.0007 | ↑ +0.007 | no |
| away__team.rest_days | 0.0004 | ↓ -0.015 | no |
| home__team.goals_against_avg_l5 | 0.0001 | ↓ -0.016 | no |
| match.temp_c | -0.0007 | ↓ -0.003 | no |
| away__team.cards_avg_l5 | -0.0010 | ↓ -0.030 | ⏳ 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 10+ corners?",
"sport": "football",
"target": {
"op": ">=",
"value": 10,
"metric": "match.total_corners"
},
"features": "all"
}