All Sports Performance

AD TEST MODE

PERFORMANCE HEADER

Mock Placement 728x90

WNBA Basketball

basketball • Performance Deep Dive

Trained 1d ago Predicted 1h ago

All-Time Average (49 folds)

Flat ROI

+57.7%

Hit Rate

63.1%

+10.7pp vs b/e

Recent Trend (last 2 folds · Sep 11–Jun 17, 2026 ) ↑ Improving

Flat ROI

+81.3%

+23.5pp vs avg

Hit Rate

70.0%

Out-of-sample Accuracy

60.0%

Avg across 49 folds

AUC-ROC

0.609

0.5 = coin flip · 1.0 = perfect

Log Loss

0.6627

Lower is better

Brier Score

0.2350

Probability calibration

49 validation folds

Jul 13, 2017 – Jun 17, 2026

Final Holdout Test

Held back from all training and fold evaluation — the cleanest out-of-sample signal. Period: Jul 6, 2026 – Jul 13, 2026

Holdout AUC

0.773

0.5 = coin flip

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 (42 folds)

Flat ROI

+32.4%

Hit Rate

69.3%

+16.9pp vs b/e

Recent Trend (last 0 folds · Sep 18–Jun 10, 2026 )

Flat ROI

—

Hit Rate

—

Out-of-sample Accuracy

58.3%

Avg across 42 folds

AUC-ROC

0.623

0.5 = coin flip · 1.0 = perfect

Log Loss

0.6716

Lower is better

Brier Score

0.2400

Probability calibration

42 validation folds

Jun 22, 2018 – Jun 10, 2026

Final Holdout Test

Held back from all training and fold evaluation — the cleanest out-of-sample signal. Period: Jul 4, 2026 – Jul 11, 2026

Holdout AUC

0.577

0.5 = coin flip

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 (43 folds)

Flat ROI

+33.9%

Hit Rate

66.5%

+14.1pp vs b/e

Recent Trend (last 1 folds · Sep 18–Jun 10, 2026 ) ↓ Declining

Flat ROI

-4.4%

-38.3pp vs avg

Hit Rate

50.0%

Out-of-sample Accuracy

57.4%

Avg across 43 folds

AUC-ROC

0.622

0.5 = coin flip · 1.0 = perfect

Log Loss

0.6708

Lower is better

Brier Score

0.2395

Probability calibration

43 validation folds

Jun 22, 2018 – Jun 10, 2026

Final Holdout Test

Held back from all training and fold evaluation — the cleanest out-of-sample signal. Period: Jul 4, 2026 – Jul 11, 2026

Holdout AUC

0.583

0.5 = coin flip

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

13.632

Root mean squared error

Avg MAE

11.069

Mean absolute error

R² Score

-0.0530

Explained variance

49 validation folds

Jul 13, 2017 – Jun 17, 2026

Fold-by-Fold Metrics

Supporter+ required

Supporter access required

Unlock fold-level charts to identify training stability.

Unlock Supporter

Model: LightGBMRegressorModel

Calibration: platt

Avg RMSE

17.273

Root mean squared error

Avg MAE

13.668

Mean absolute error

R² Score

-0.0626

Explained variance

49 validation folds

Jul 13, 2017 – Jun 17, 2026

Fold-by-Fold Metrics

Supporter+ required

Supporter access required

Unlock fold-level charts to identify training stability.

Unlock Supporter

Model: LightGBMRegressorModel

Calibration: platt