Roller massage (RM) has been reported to increase range of motion (ROM) without subsequent performance decrements. However, the effects of different rolling forces have not been examined. The purpose of this study was to compare the effects of sham (RMsham), moderate (RMmod), and high (RMhigh) RM forces, calculated relative to the individuals' pain perception, on ROM, strength, and jump parameters. Sixteen healthy individuals (27 +/- 4 years) participated in this study. The intervention involved three 60-second quadriceps RM bouts with RMlow (3.9/10 +/- 0.64 rating of perceived pain [RPP]), RMmod (6.2/10 +/- 0.64 RPP), and RMhigh (8.2/10 +/- 0.44 RPP) pain conditions, respectively. A within-subject design was used to assess dependent variables (active and passive knee flexion ROM, single-leg drop jump [DJ] height, DJ contact time, DJ performance index, maximum voluntary isometric contraction [MVIC] force, and force produced in the first 200 milliseconds [F200] of the knee extensors and flexors). A 2-way repeated measures analysis of variance showed a main effect of testing time in active (p < 0.001, d = 2.54) and passive (p < 0.001, d = 3.22) ROM. Independent of the RM forces, active and passive ROM increased by 7.0% (p = 0.03, d = 2.25) and 15.4% (p < 0.001, d = 3.73) from premeasure to postmeasure, respectively. Drop jump and MVIC parameters were unaffected from pretest to posttest (p > 0.05, d = 0.33-0.84). Roller massage can be efficiently used to increase ROM without substantial pain and without subsequent performance impairments.