All Sports Performance

AD TEST MODE

PERFORMANCE HEADER

Mock Placement 728x90

MLB Baseball

baseball • Performance Deep Dive

Trained 12h ago Predicted 12h ago

All-Time Average (70 folds)

Flat ROI

+19.5%

Hit Rate

54.5%

+2.1pp vs b/e

Recent Trend (last 2 folds · Apr 1–Jun 23, 2026 ) ↓ Declining

Flat ROI

+12.2%

-7.4pp vs avg

Hit Rate

53.8%

Out-of-sample Accuracy

55.4%

Avg across 70 folds

AUC-ROC

0.567

0.5 = coin flip · 1.0 = perfect

Log Loss

0.6793

Lower is better

Brier Score

0.2433

Probability calibration

70 validation folds

May 10, 2017 – Jun 23, 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 mlb 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 (70 folds)

Flat ROI

+54.0%

Hit Rate

64.4%

+12pp vs b/e

Recent Trend (last 3 folds · Apr 1–Jun 23, 2026 ) ↓ Declining

Flat ROI

+26.6%

-27.5pp vs avg

Hit Rate

52.3%

Out-of-sample Accuracy

57.0%

Avg across 70 folds

AUC-ROC

0.582

0.5 = coin flip · 1.0 = perfect

Log Loss

0.6749

Lower is better

Brier Score

0.2411

Probability calibration

70 validation folds

May 10, 2017 – Jun 23, 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 mlb 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 (69 folds)

Flat ROI

+16.8%

Hit Rate

60.1%

+7.7pp vs b/e

Recent Trend (last 2 folds · Apr 2–Jun 24, 2026 ) ↓ Declining

Flat ROI

+13.3%

-3.5pp vs avg

Hit Rate

59.2%

Out-of-sample Accuracy

54.2%

Avg across 69 folds

AUC-ROC

0.549

0.5 = coin flip · 1.0 = perfect

Log Loss

0.6859

Lower is better

Brier Score

0.2464

Probability calibration

69 validation folds

May 11, 2017 – Jun 24, 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 mlb 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 (70 folds)

Flat ROI

+18.6%

Hit Rate

62.7%

+10.3pp vs b/e

Recent Trend (last 2 folds · Apr 1–Jun 23, 2026 ) ↑ Improving

Flat ROI

+22.2%

+3.6pp vs avg

Hit Rate

63.3%

Out-of-sample Accuracy

55.0%

Avg across 70 folds

AUC-ROC

0.564

0.5 = coin flip · 1.0 = perfect

Log Loss

0.6820

Lower is better

Brier Score

0.2446

Probability calibration

70 validation folds

May 10, 2017 – Jun 23, 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 mlb 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

4.497

Root mean squared error

Avg MAE

3.554

Mean absolute error

R² Score

-0.0090

Explained variance

70 validation folds

May 10, 2017 – Jun 23, 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

4.493

Root mean squared error

Avg MAE

3.568

Mean absolute error

R² Score

-0.0026

Explained variance

70 validation folds

May 10, 2017 – Jun 23, 2026

Fold-by-Fold Metrics

Supporter+ required

Supporter access required

Unlock fold-level charts to identify training stability.

Unlock Supporter

Model: LightGBMRegressorModel

Calibration: platt

All-Time Average (143 folds)

Flat ROI

+23.6%

Hit Rate

54.1%

+1.7pp vs b/e

Recent Trend (last 2 folds · Apr 1–May 12, 2026 ) ↓ Declining

Flat ROI

-5.1%

-28.7pp vs avg

Hit Rate

31.3%

Out-of-sample Accuracy

52.9%

Avg across 143 folds

AUC-ROC

0.521

0.5 = coin flip · 1.0 = perfect

Log Loss

1.0137

Lower is better

Brier Score

---

Probability calibration

143 validation folds

Apr 26, 2017 – May 12, 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 mlb 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