Menu
Login Get Started
All Sports Performance
AD TEST MODE
PERFORMANCE HEADER
Mock Placement 728x90
ML

MLB Baseball

baseball • Performance Deep Dive

Trained Never Predicted Never
All-Time Average (142 folds)
Flat ROI
+49.5%
Hit Rate
53.9%
+1.5pp vs b/e
Recent Trend (last 3 folds · Mar 25–Apr 28, 2026 ) ↑ Improving
Flat ROI
+62.6%
+13.1pp vs avg
Hit Rate
58.8%
Out-of-sample Accuracy
54.1%

Avg across 142 folds

AUC-ROC
0.549

0.5 = coin flip · 1.0 = perfect

Log Loss
0.6840

Lower is better

Brier Score
0.2455

Probability calibration

142 validation folds
Apr 26, 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 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: LightGBMModel
Calibration: platt
All-Time Average (140 folds)
Flat ROI
+38.9%
Hit Rate
58.1%
+5.7pp vs b/e
Recent Trend (last 2 folds · Mar 25–Apr 28, 2026 ) ↑ Improving
Flat ROI
+46.1%
+7.2pp vs avg
Hit Rate
60.0%
Out-of-sample Accuracy
53.4%

Avg across 140 folds

AUC-ROC
0.573

0.5 = coin flip · 1.0 = perfect

Log Loss
0.6840

Lower is better

Brier Score
0.2456

Probability calibration

140 validation folds
Apr 26, 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 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 (140 folds)
Flat ROI
+32.7%
Hit Rate
66.8%
+14.4pp vs b/e
Recent Trend (last 1 folds · Mar 25–Apr 28, 2026 ) ↑ Improving
Flat ROI
+38.3%
+5.7pp vs avg
Hit Rate
66.7%
Out-of-sample Accuracy
53.1%

Avg across 140 folds

AUC-ROC
0.560

0.5 = coin flip · 1.0 = perfect

Log Loss
0.6868

Lower is better

Brier Score
0.2468

Probability calibration

140 validation folds
Apr 26, 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 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 (142 folds)
Flat ROI
+38.9%
Hit Rate
65.3%
+12.9pp vs b/e
Recent Trend (last 3 folds · Mar 25–Apr 28, 2026 ) → Stable
Flat ROI
+38.5%
-0.4pp vs avg
Hit Rate
61.5%
Out-of-sample Accuracy
53.8%

Avg across 142 folds

AUC-ROC
0.567

0.5 = coin flip · 1.0 = perfect

Log Loss
0.6862

Lower is better

Brier Score
0.2466

Probability calibration

142 validation folds
Apr 26, 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 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
4.487

Root mean squared error

Avg MAE
3.552

Mean absolute error

R² Score
-0.0068

Explained variance

142 validation folds
Apr 26, 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
Avg RMSE
4.490

Root mean squared error

Avg MAE
3.568

Mean absolute error

R² Score
-0.0104

Explained variance

142 validation folds
Apr 26, 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