PUBLIC SCORING

Track Record

Every prediction we publish is scored against reality. This page is our permanent, public accountability record. No retroactive adjustments. No cherry-picking.

Every prediction is scored against the actual outcome and surfaced on the chart below. We share both hits and misses — the regulation reset taught the model new patterns and our rookie predictions are still calibrating, so each scored race shifts the picture.

0.035

Brier score

Lower is better, 0 is perfect

+26%

Skill score

vs grid baseline

4

Races scored

2026 season

BRIER DECOMPOSITION

Brier = Reliability - Resolution + Uncertainty

Reliability

Calibration error — how well probabilities match reality

0.0270

Resolution

How much predictions vary from the base rate

0.0335

Uncertainty

Inherent unpredictability of F1 outcomes

0.0423
BASELINES
ModelBrierSkill Score
The Data Driver0.035reference
Grid baseline0.047+25.7%
Championship baseline0.060+41.8%
Random uniform0.054+35.3%

Skill Score = improvement over baseline. BSS = 1 - (model / reference).

CALIBRATION
0%10%20%30%40%50%0%10%20%30%40%50%Predicted probabilityActual frequency

When we predict a 30% chance, it happens ~30% of the time. Points on the diagonal indicate perfect calibration.

LOG LOSS

Log loss penalises confident wrong predictions more heavily than Brier score

R01Australian Grand Prix
0.153

log loss

0.040

brier

R02Chinese Grand Prix
0.106

log loss

0.035

brier

R03Japanese Grand Prix
0.106

log loss

0.036

brier

R06Miami Grand Prix
0.088

log loss

0.029

brier

SHARPNESS

Mean favourite probability

Higher sharpness = model is more decisive

32.3%

11%
Australian
36%
Chinese
41%
Japanese
42%
Miami

Trend: increasing — model confidence grows as season data accumulates

BRIER SCORE BY RACE
Grid baseline (0.047)
R01Australian Grand Prix
0.040Beats grid
Predicted: Max VerstappenActual: George Russell
R02Chinese Grand Prix
0.035Beats grid
Predicted: George RussellActual: Kimi Antonelli
R03Japanese Grand Prix
0.036Beats grid
Predicted: George RussellActual: Kimi Antonelli
R06Miami Grand Prix
0.029Beats grid
Predicted: George RussellActual: Kimi Antonelli
HISTORICAL VALIDATION

Walk-forward backtest

Validated on 114 races (2021-2025)

Model retrained before each race using only past data — no future leakage

0.041

Brier score

0.558

Spearman rank

PREDICTION HISTORY

Australian GP

Russell wins the Australian Grand Prix

Correct
Prediction: 26%Brier: 0.548

Australian GP

Safety car at Melbourne

Correct
Prediction: 55%Brier: 0.203

Australian GP

Antonelli on the podium in Australia

Correct
Prediction: 40%Brier: 0.360

Australian GP

Mercedes 1-2 in Australia

Correct
Prediction: 15%Brier: 0.723

Australian GP

Hamilton top 5 in Australia

Correct
Prediction: 62%Brier: 0.144

Australian GP

Verstappen on the podium in Melbourne

Wrong
Prediction: 42%Brier: 0.176

Australian GP

Leclerc on the podium in Melbourne

Correct
Prediction: 35%Brier: 0.423

Australian GP

More than 3 retirements in Australia

Correct
Prediction: 40%Brier: 0.360

Chinese GP

Russell wins the Chinese Grand Prix

Wrong
Prediction: 31%Brier: 0.096

Chinese GP

Safety car deployed in China

Correct
Prediction: 62%Brier: 0.144

Chinese GP

Antonelli on the podium in China

Correct
Prediction: 45%Brier: 0.303

Chinese GP

Hamilton podium at Shanghai

Correct
Prediction: 30%Brier: 0.490

Chinese GP

Same winner in Sprint and Race in China

Wrong
Prediction: 38%Brier: 0.144

Chinese GP

More than 3 retirements at the Chinese GP

Correct
Prediction: 42%Brier: 0.336

Chinese GP

Mercedes 1-2 in China

Correct
Prediction: 18%Brier: 0.672

Chinese GP

Verstappen on the podium in China

Wrong
Prediction: 28%Brier: 0.078

Transparency

Predictions published before each race. Track record computed automatically after each race using Brier score. We never modify predictions retroactively. Every probability is timestamped and immutable.

Scoring methodology: Brier score = mean squared error of probabilistic predictions. Skill score = 1 - (model Brier / baseline Brier). Grid baseline uses qualifying order.