On the evaluation of optimal and nonoptimal control strategies