sweep: match.temp_c >= 10.7
← lab · AUC 0.785 (real 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? |
|---|
| match.stakes | 0.0689 | ↑ +0.035 | ⏳ yes |
| away__team.rest_days | 0.0155 | ↑ +0.152 | ⏳ yes |
| away__t3__team.season_ppg | 0.0109 | ↑ +0.009 | no |
| home__team.season_ppg | 0.0090 | ↑ +0.002 | no |
| home__t3__team.season_ppg | 0.0089 | ↓ -0.021 | no |
| home__team.rest_days | 0.0088 | ↑ +0.156 | ⏳ yes |
| home__t3__team.congestion_21d | 0.0086 | ↓ -0.026 | ⏳ yes |
| away__team.season_ppg | 0.0079 | ↑ +0.004 | no |
| home__team.shots_for_avg_l5 | 0.0043 | ↑ +0.064 | ⏳ yes |
| home__t3__team.pass_acc_avg_l5 | 0.0043 | ↑ +0.066 | ⏳ yes |
| away__t3__team.sot_for_avg_l5 | 0.0040 | ↑ +0.025 | ⏳ yes |
| diff__team.ht_lead_rate_l20 | 0.0038 | ↓ -0.002 | no |
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.temp_c >= 10.7",
"sport": "football",
"target": {
"op": ">=",
"value": 10.7,
"metric": "match.temp_c"
},
"features": "all"
}