MODEL v8.0 OPERATIONAL · AUC 0.919 —Z
Clawd Front

Performance Lab

Model skill · Calibration · Feature importance — the same numbers we use internally.

Model Skill

AUC-ROC across held-out storm years. Strict temporal split — held-out years never touch training data.

AUC-ROC by Version
Higher is better. 0.5 = coin flip. 1.0 = perfect discrimination.
0.96 0.92 0.88 0.84 0.80 v5.0 v5.1 v5.2 v8.0 0.919 0.814 0.882
EVALUATION: STRICT TEMPORAL SPLIT · HELD-OUT YEARS NEVER TOUCH TRAINING
Calibration · Reliability Diagram
When the model says 15%, tornadoes should occur ~15% of the time.
FORECAST PROB → OBSERVED FREQ → PERFECT
ISOTONIC-CALIBRATED · v8.0 · BRIER-SCORED VS NWS LSRs

Feature Importance

What the model actually weighs when computing tornado probability — v8.0 XGBoost gain scores.

Top Features · v8.0
Normalized feature gain. Higher = more influential in final probability output.
neighbor_sig_cnt0.27
synoptic_forcing0.12
STP0.09
CAPE (sfc-based)0.07
SRH 0–3km0.06
LCL height0.05
0–6km bulk shear0.04
SPATIAL CLUSTERING OF SIGNIFICANT TORNADO REPORTS (150KM RADIUS) DOMINATES — TORNADOES TRAVEL IN PACKS

Verification Standards

How we measure performance — and what triggers automatic retraining.

0.919
AUC-ROC · Held-out storm days

The model correctly ranks a tornado day above a non-tornado day 91.9% of the time across years it has never seen.

EF1+ / 25MI
Verification target

A hit is any EF1+ tornado within 25 miles of the forecast point. No soft definitions, no EF0 inflation.

90-DAY
Retraining trigger

If rolling AUC drops below 0.87 over any 90-day window, the model automatically enters retraining. No manual intervention required.