The underuse of cardiovascular recovery as an adjunct to reactivity may stem from a lack of research on how to assess the process reliably. We explore the test–retest reliability of three simple, intuitive approaches to measuring recovery, and of a more sophisticated curve-fitting technique. Eighteen young normotensive subjects experienced three stressors twice each, with 10-min baseline, 3-min task, and 20-min recovery periods and continuous monitoring of heart rate and blood pressure. Reactivity showed moderate reliability, but the three simple approaches to measuring recovery revealed essentially none. However, the curve-fitting approach, using a three-parameter (amount, speed, and level of recovery) logistic function was reliable. This approach, capturing the inherently dynamic process of cardiovascular recovery, may allow researchers to usefully add the assessment of recovery to paradigms exploring reactivity as a risk factor for cardiovascular disease.