The aim of this study was two-fold: (i) analyze the variation of kinematic measures between using preferred and non-preferred legs while braking during the 5–0–5 change of direction test; and (ii) test the relationships between kinematic measures, and change-of-direction deficit (CODD). A cross-sectional study using twenty adult male soccer players (21.6 ± 2.0 years; 73.2 ± 6.1 kg; 174.8 ± 4.5 cm) was employed. Players performed three repetitions of the 5–0–5 test using each leg during the braking phase. Players have used the Polar Team Pro to obtain the kinematic measures of peak speed, peak acceleration, and peak deceleration. Additionally, the CODD was also obtained using single-beamed photocells. Comparisons revealed a significantly greater peak acceleration (+ 0.22 m/s2; p < 0.001) and deceleration (+ 0.17 m/s2; p = 0.004) for the non-preferred leg. There were no significant correlations were found between CODD and peak accelerations (r = − 0.014, [− 0.193; 0.166]), peak decelerations (r = − 0.052, [− 0.229; 0.128]) or peak speed (r = 0.118, [− 0.063; 0.291]). This study revealed that preferred and non-preferred leg must be analyzed differently since they are different in the kinematic variables. However, CODD seems independent of leg preference and the kinematic measures of a 5–0–5 change of direction test.