BACKGROUND Muscle stretching is widely used in sport training and in rehabilitation. Considering the important contribution of joint position sense (JPS) to knee joint stability and function, it is legitimate to question if stretching might alter the knee JPS. OBJECTIVE To evaluate if a stretch regimen consisting of three 30 s stretches alters the knee JPS. METHODS A blinded, randomised design with a washout time of 24 h was used. METHODS 39 healthy students (21 women, 18 men) volunteered to participate in this study. METHODS JPS was estimated by the ability to reproduce the two target positions (20 degrees and 45 degrees of flexion) in the dominant knee. The absolute angular error (AAE) was defined as the absolute difference between the target angle and the subject perceived angle of knee flexion. AAE values were measured before and immediately after the static stretch. Measurements were repeated three times. The static stretch comprised a 30 s stretch followed by a 30 s pause, three times for each muscle. RESULTS The AAE decreased significantly after the stretching protocols for quadriceps (3.5 (1.3) vs 0.7 (2.4); p<0.001), hamstring (3.6 (2.2) vs 1.6 (3.1); p = 0.016) and adductors (3.7 (2.8) vs 1.7 (2.4); p = 0.016) in 45 degrees of flexion, but no differences were found for values of the gastrocnemius and popliteus muscles in this angle and for the values of all muscles in 20 degrees of flexion (p>0.05). CONCLUSIONS The accuracy of the knee JPS in 45 degrees of flexion is improved subsequent to a static stretch regimen of quadriceps, hamstring and adductors in healthy subjects.