Decision Threshold
A decision threshold is a cutoff probability used to convert probabilities into yes/no decisions, such as acting only when p is above 0.60.
Definition
A decision threshold is the probability cutoff you use to take an action. For example, you might treat p >= 0.60 as “Yes” and p < 0.60 as “No”.
Why it matters
Forecast quality is not the same as decision quality. Scoring rules like Brier score evaluate probability accuracy, while thresholds define operational decisions. A threshold should reflect costs, payoffs, and risk tolerance.
Common pitfall
Judging forecasters by accuracy only: If you pick a threshold and evaluate with plain accuracy, you throw away probability information. Prefer proper scoring rules for evaluation.
Related
Thresholds are tied to expected value and to how you define “actionable” edges versus a benchmark.