Compare First-Visit and Every-Visit Monte Carlo prediction algorithms. Learn how to accurately evaluate policies from raw agent experience without a model.