All Sports Performance

AD TEST MODE

PERFORMANCE HEADER

Mock Placement 728x90

NBA Basketball

basketball • Performance Deep Dive

Trained Never Predicted Never

All-Time Average (154 folds)

Flat ROI

+61.7%

Hit Rate

59.3%

+6.9pp vs b/e

Recent Trend (last 3 folds · Mar 18–Apr 28, 2026 ) ↑ Improving

Flat ROI

+105.0%

+43.3pp vs avg

Hit Rate

75.0%

Out-of-sample Accuracy

59.0%

Avg across 154 folds

AUC-ROC

0.576

0.5 = coin flip · 1.0 = perfect

Log Loss

0.6666

Lower is better

Brier Score

0.2374

Probability calibration

154 validation folds

Nov 21, 2017 – Apr 28, 2026

Final Holdout Test

Held back from all training and fold evaluation — the cleanest out-of-sample signal.

No holdout data available. Re-run prediction-system run --sport nba to generate holdout metrics.

Fold-by-Fold Metrics

Supporter+ required

Supporter access required

Unlock fold-level charts to identify training stability.

Unlock Supporter

Betting Simulation

Supporter+

Recommendation ROI

�

Recommendation Win Rate

�

Universe ROI

�

Universe Win Rate

�

Supporter access required

Unlock Supporter

Model: LightGBMModel

Calibration: platt

All-Time Average (154 folds)

Flat ROI

+30.4%

Hit Rate

68.2%

+15.8pp vs b/e

Recent Trend (last 1 folds · Mar 18–Apr 28, 2026 ) ↓ Declining

Flat ROI

+22.1%

-8.3pp vs avg

Hit Rate

63.6%

Out-of-sample Accuracy

55.8%

Avg across 154 folds

AUC-ROC

0.582

0.5 = coin flip · 1.0 = perfect

Log Loss

0.6759

Lower is better

Brier Score

0.2416

Probability calibration

154 validation folds

Nov 21, 2017 – Apr 28, 2026

Final Holdout Test

Held back from all training and fold evaluation — the cleanest out-of-sample signal.

No holdout data available. Re-run prediction-system run --sport nba to generate holdout metrics.

Fold-by-Fold Metrics

Supporter+ required

Supporter access required

Unlock fold-level charts to identify training stability.

Unlock Supporter

Betting Simulation

Supporter+

Recommendation ROI

�

Recommendation Win Rate

�

Universe ROI

�

Universe Win Rate

�

Supporter access required

Unlock Supporter

Model: XGBoostModel

Calibration: platt

All-Time Average (154 folds)

Flat ROI

+26.2%

Hit Rate

66.1%

+13.7pp vs b/e

Recent Trend (last 0 folds · Mar 18–Apr 28, 2026 )

Flat ROI

—

Hit Rate

—

Out-of-sample Accuracy

57.1%

Avg across 154 folds

AUC-ROC

0.592

0.5 = coin flip · 1.0 = perfect

Log Loss

0.6766

Lower is better

Brier Score

0.2422

Probability calibration

154 validation folds

Nov 21, 2017 – Apr 28, 2026

Final Holdout Test

Held back from all training and fold evaluation — the cleanest out-of-sample signal.

No holdout data available. Re-run prediction-system run --sport nba to generate holdout metrics.

Fold-by-Fold Metrics

Supporter+ required

Supporter access required

Unlock fold-level charts to identify training stability.

Unlock Supporter

Betting Simulation

Supporter+

Recommendation ROI

�

Recommendation Win Rate

�

Universe ROI

�

Universe Win Rate

�

Supporter access required

Unlock Supporter

Model: XGBoostModel

Calibration: platt

Avg RMSE

20.245

Root mean squared error

Avg MAE

16.003

Mean absolute error

R² Score

-0.0428

Explained variance

154 validation folds

Nov 21, 2017 – Apr 28, 2026

Fold-by-Fold Metrics

Supporter+ required

Supporter access required

Unlock fold-level charts to identify training stability.

Unlock Supporter

Model: LightGBMRegressorModel

Calibration: platt