The Reliability and Validity of Fatigue Measures During Multiple-Sprint Work: An Issue Revisited

The ability to repeatedly produce a high-power output or sprint speed is a key fitness component of most field and court sports. The aim of this study was to evaluate the validity and reliability of eight different approaches to quantify this parameter in tests of multiple-sprint performance. Ten physically active men completed two trials of each of two multiple-sprint running protocols with contrasting recovery periods. Protocol 1 consisted of 12 x 30-m sprints repeated every 35 seconds; protocol 2 consisted of 12 x 30-m sprints repeated every 65 seconds. All testing was performed in an indoor sports facility, and sprint times were recorded using twin-beam photocells. All but one of the formulae showed good construct validity, as evidenced by similar within-protocol fatigue scores. However, the assumptions on which many of the formulae were based, combined with poor or inconsistent test-retest reliability (coefficient of variation range: 0.8-145.7%; intraclass correlation coefficient range: 0.09-0.75), suggested many problems regarding logical validity. In line with previous research, the results support the percentage decrement calculation as the most valid and reliable method of quantifying fatigue in tests of multiple-sprint performance.