sweep: match.total_fouls >= 23

← lab · AUC 0.600 (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.ref_fouls_avg	0.0164	↑ +0.221	⏳ yes
market.p_draw	0.0110	↑ +0.142	⏳ yes
match.ref_cards_avg	0.0026	↑ +0.066	⏳ yes
home__team.corners_volatility_l10	0.0024	↓ -0.005	no
x__match.ref_cards_avg__match.tempo	0.0024	↓ -0.001	no
home__t3__team.cards_avg_l5	0.0022	↑ +0.002	no
match.humidity_pct	0.0022	↓ -0.015	no
away__team.travel_km	0.0018	↓ -0.020	no
away__team.cards_avg_l5	0.0016	↑ +0.029	⏳ yes
ratio__team.rest_days	0.0016	↓ -0.011	no
home__team.h2_goals_l10	0.0014	↓ -0.059	⏳ yes
away__team.conversion_l5	0.0014	↓ -0.007	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_fouls >= 23",
  "sport": "football",
  "target": {
    "op": ">=",
    "value": 23,
    "metric": "match.total_fouls"
  },
  "features": "all"
}