Aim The aim of this study was to analyse the feasibility and reliability of the tests used to determine muscle strength, sprint power, and aerobic capacity in athletes and non-athletes with cerebral palsy (CP). Methods Twenty individuals with spastic CP (four females, 16 males; age range 18-49y; Gross Motor Function Classification System level I, n=15; II, n=5; unilateral CP, n=10; bilateral CP, n=10; athletes, n=12; non-athletes, n=8) participated in the study. Isometric and isokinetic knee flexor and extensor strength, sprint power, and aerobic capacity were determined, using, respectively an isokinetic dynamometer, a Wingate cycling test, and a graded maximal bicycle exercise test, on three occasions. Intraclass correlation coefficients (ICC), standard error of measurements, and smallest detectable differences (SDD) were calculated. Results The feasibility of the isometric strength test, Wingate test, and graded exercise test was good; the isokinetic strength test was difficult to perform for five participants. The strength parameters showed moderate to good ICCs (isometric, 0.74-0.94; isokinetic, 0.88-0.93) but high SDDs (isometric, 25-45%; isokinetic, 30-45%). Sprint power (ICC 0.98; SDD 24%) and aerobic capacity (ICC 0.98-0.99; SDD 16-21%) showed good ICCs and moderate SDDs. Interpretation All tests, except for the isokinetic strength test, seemed to be feasible for almost all participants. All tests are suitable for evaluating changes in a group; however, only large improvements (16-45%) can be detected when monitoring individual changes. © The Authors. Developmental Medicine & Child Neurology © 2012 Mac Keith Press.