draws v2 (mismatch theory)

← lab · AUC 0.526 (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.matches_since_win	0.0034	↓ -0.001	no
away__team.goals_for_avg_l5	0.0031	↓ -0.000	no
home__team.dominance_l5	0.0030	↓ -0.049	⏳ yes
away__team.shots_for_avg_l5	0.0027	↓ -0.012	no
away__team.dominance_l5	0.0025	↓ -0.001	no
home__team.venue_ppg_l5	0.0024	↓ -0.025	⏳ yes
match.style_clash_corners	0.0023	↓ -0.028	⏳ yes
home__team.corner_diff_avg_l5	0.0016	↓ -0.041	⏳ yes
match.temp_c	0.0016	↑ +0.002	no
match.rainfall_mm	0.0014	↑ +0.013	no
home__team.form_points_l5	0.0014	↓ -0.031	⏳ yes
away__team.congestion_21d	0.0006	↑ +0.003	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": "draws v2 (mismatch theory)",
  "sport": "football",
  "target": {
    "equals": "D",
    "metric": "match.result"
  },
  "features": "all"
}