Agreement of Total Distance and Speed Metrics Across Multiple Football Tracking Devices

We conducted a comparison of athletic metrics collected by five football tracking devices: A GNSS vest, GNSS/IMU smart shin guards, and three IMU lower-limb devices, worn simultaneously during 24 training sessions. These devices are designed to monitor the external load of the players, including key athletic metrics such as total distance covered and maximum speed. Limits of Agreement and Intraclass Correlation Coefficient (ICC) were computed for these metrics. Results reported a high reliability for distance covered (ICC = 0.833) but comparatively lower reliability for max speed (ICC = 0.310). We revealed variance in biases among the devices, with GNSS-based systems showing congruent measurements and notable biases detected when comparing GNSS with IMU-based devices, as well as among IMU devices themselves. These findings indicate that while these devices are generally reliable for assessing athletes' total distance covered, measuring maximum speed proves more complex due to the transient nature of velocity, varying device processing capabilities, and different sensor technologies. Future enhancements will extend the scope to a wider range of athletic and technical metrics and incorporate standardized tests to evaluate the devices' measurements against known distances.