Introduction. Running, of which backwards running (BR) is one type, is a basic skill that has to be maintained at a high level by athletes. Aim of Study. Use cluster analysis to evaluate some kinematic variables for BR. This analysis is applied to classify players and identify differences in their classification to determine relevant dynamic solutions to raise players’ performance levels. Material and Methods. Twelve volunteer university soccer players (age: 20.8 ± 0.83 years old; experience: 4.7 ± 0.78 years; height: 175.6 ± 6.01 cm; body mass: 68.63 ± 5.06 kg) participated in the present study. The participants tried two 10-m BR, in which the best attempt based on the shortest time was analysed. Results. The study showed that cluster analysis may be used to classify and divide participants into two groups via evaluations of selected biomechanical variables. The first group, which consisted of 7 participants, represents the indistinctive performance level, while the second group, which consisted of 5 participants, represents the distinctive performance level. Statistically significant differences were found between the classifications of the participants. The second group excelled regarding certain biomechanical variables, including average stride length, average speed, angular velocity of the arms, peripheral velocity of the arms, angular velocity of the legs, peripheral velocity of the legs, instantaneous force and time of achievement. Conclusions. This classification ensures the correct selection and full consideration of practical training to achieve the ideal biomechanical characteristics for BR in soccer.