sweep: match.total_goals >= 2
← lab · AUC 0.534 (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? |
|---|
| home__t3__team.rest_days | 0.0005 | ↑ +0.002 | no |
| away__team.congestion_21d | 0.0005 | ↓ -0.028 | ⏳ yes |
| home__team.conversion_l5 | 0.0005 | ↑ +0.006 | no |
| ratio__team.ht_lead_rate_l20 | 0.0004 | ↑ +0.020 | no |
| away__t3__team.shot_diff_avg_l5 | 0.0004 | ↑ +0.010 | no |
| market.p_draw | 0.0004 | ↓ -0.083 | ⏳ yes |
| away__team.shots_for_avg_l5 | 0.0004 | ↑ +0.041 | ⏳ yes |
| market.p_over25 | 0.0003 | ↑ +0.141 | ⏳ yes |
| home__t3__team.conversion_l5 | 0.0003 | ↓ -0.002 | no |
| away__team.season_ppg | 0.0003 | ↑ +0.008 | no |
| ratio__team.goals_volatility_l10 | 0.0002 | ↑ +0.013 | no |
| ratio__team.goals_for_avg_l5 | 0.0002 | ↑ +0.013 | 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.total_goals >= 2",
"sport": "football",
"target": {
"op": ">=",
"value": 2,
"metric": "match.total_goals"
},
"features": "all"
}