Record sports-modeling experiments with the hypothesis, data cut, validation charter, metrics, leakage status, decision, commands, and artifacts. Use for trials, model comparisons, and reproducible research history.
