Objective To evaluate the accuracy and operability of this newly developed master-slave long bone fracture reduction robotic system, and to investigate the experimental data with statistical analysis. Methods The slave side of the robotic system is the previously developed long bone fracture reduction parallel manipulator robot on the orthopedic traction table. The ring shaped platforms are mounted on the fractured long bone model. The master robot takes the form of parallel manipulator. The operator maneuvers the master robot to control the slave robot via conditional mapping unit to perform the fracture reduction. One way ANOVA is used to analyze the reduction accuracies and the numbers of operations. Results There is no significant difference in reduction accuracies among the three operators (P > 0.05). The number of operations is significantly lower (P < 0.05) in viewing bone model directly than viewing it through camera screens. Conclusions This robotic system can satisfy the accuracy requirement for fracture reduction, reduce the repetitive usage of X-ray imaging and lower the working intensity of operating surgeons. And the system has short learning curve. Such system can provide an efficient method for long bone fracture reduction.