Reliability and generality of measures of acceleration , planned and reactive agility

Purpose: The purpose of the present study was to assess the reliability of a new protocol which examines different components of agility using commercially available timing gates. Methods: Seventeen physically active males completed four trials of a new protocol, which consisted of a number of 10 m sprints. Sprints were completed in a straight line or with a change of direction after 5 m. The change of direction was either planned or reactive, with participants reacting to a visual light stimulus. Results: There was no systematic bias in any of the measures although random variation was reduced in the straight acceleration and planned agility when considering only the final pair of trials, with mean coefficients of variation (CV) of 1.6% (95%CI 1.2-2.4%) and 1.1% (0.8-1.7%) respectively. Reliability of reactive agility remained consistent throughout with mean CVs of approximately 3%. Analyses revealed a high degree of common variance between acceleration times and both planned (r = 0.93) and reactive (r = 0.83) agility, as well as between the two agility protocols (r = 0.87). Conclusion: Both planned and reactive agility could be measured reliably. Protocol design and use of a light stimulus in the reactive test emphasize physical abilities comparable with other test measures. Therefore, inclusion of a reactive light stimulus does not appear to require any additional perceptual qualities.