sweep: match.result D
← lab · AUC 0.531 (weak signal) · ran 7/4/2026
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? |
|---|
| market.p_over25 | 0.0113 | ↓ -0.085 | ⏳ yes |
| home__team.elo_momentum_l5 | 0.0076 | ↓ -0.004 | no |
| home__team.cards_avg_l5 | 0.0071 | ↑ +0.006 | no |
| away__team.venue_ppg_l5 | 0.0063 | ↑ +0.002 | no |
| away__team.goal_diff_avg_l5 | 0.0060 | ↑ +0.002 | no |
| market.ah_line | 0.0034 | ↑ +0.043 | ⏳ yes |
| match.mismatch | 0.0027 | ↓ -0.079 | ⏳ yes |
| away__team.corners_against_avg_l5 | 0.0027 | ↓ -0.005 | no |
| away__team.corners_for_avg_l5 | 0.0023 | ↑ +0.001 | no |
| home__team.matches_since_clean_sheet | 0.0019 | ↑ +0.001 | no |
| match.humidity_pct | 0.0015 | ↓ -0.001 | no |
| home__team.elo | 0.0013 | ↓ -0.061 | ⏳ 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": "sweep: match.result D",
"sport": "football",
"target": {
"equals": "D",
"metric": "match.result"
},
"features": "all"
}