Repeated-sprint ability (RSA) is now well accepted as an important fitness component in team-sport performance. It is broadly described as the ability to perform repeated short (~3-4 s, 20-30 m) sprints with only brief (~10-30 s) recovery between bouts. Over the past 25 y a plethora of RSA tests have been trialed and reported in the literature. These range from a single set of ~6-10 short sprints, departing every 20-30 s, to team-sport game simulations involving repeating cycles of walk-jog-stride-sprint movements over 45-90 min. Such a wide range of RSA tests has not assisted the synthesis of research findings in this area, and questions remain regarding the optimal methods of training to best improve RSA. In addition, how RSA test scores relate to player "work rate," match performance, or both requires further investigation to improve the application of RSA testing and training to elite team-sport athletes.