Reproducibility and validity of the PowerCal device for estimating power output during sprints in well-trained cyclists

BACKGROUND: The PowerCal device purports to effectively estimate power output (PO) using an algorithm computed from the HR response recorded during exercise. OBJECTIVE: To assess the reproducibility and validity of PO estimation using the PowerCal during sprints. METHODS: Ten well-trained male cyclists (maximal oxygen uptake-62.7 ± 5.9 ml ⋅ kg -1 ⋅ min -1 ) performed four separate experimental trials over a two-week period. Cyclists completed an incremental exercise test until volitional exhaustion followed by three maximal sprint sessions on separate days. Each sprint session consisted of 3 repeated maximal-effort sprints (15, 30, and 45 s) formatted with an effort/recovery duration ratio of 1:5. RESULTS: Peak power output (PPO) and mean power output (MPO) of the PowerCal showed high within-subject variation (6.7 to 21.5%) and low to high intra-class coefficient of correlation (ICC = 0.39 to 0.85). Furthermore, the PPO from PowerCal was underestimated by 6.6 to 13.9% and the MPO by 14.9 to 27.6% when compared to the Velotron. Furthermore, the data exhibited high bias (32 W to 129 W) and large confidence limits (-236 to 315 W) between the PO derived from the Powercal and the Velotron. CONCLUSION: We do not recommend competitive cyclists or sport scientists to use the PowerCal if reliable and valid PO information is required during maximal sprints.