Young, WB and Murray, MP. Reliability of a field test of defending and attacking agility in Australian football and relationships to reactive strength. J Strength Cond Res 31(2): 509–516, 2017—Defending and attacking agility tests for Australian football do not exist, and it is unknown whether any physical qualities correlate with these types of agility. The purposes of this study were to develop new field tests of defending and attacking agility for Australian Rules football, to determine whether they were reliable, and to describe the relationship between the agility tests to determine their specificity. Because the reactive strength (RS) of the lower limb muscles has been previously correlated with change-of-direction speed, we also investigated the relationship between this quality and the agility tests. Nineteen male competitive recreational-level Australian Rules football players were assessed on the agility tests and a drop jump test to assess RS. Interday and interrater reliability was also assessed. The agility tests involved performing 10 trials of one-on-one agility tasks against 2 testers (opponents), in which the objective was to be in a position to tackle (defending) or to evade (attacking) the opponent. Both agility tests had good reliability (intraclass correlation > 0.8, %CV < 3, and no significant differences between test occasions [p > 0.05], and interrater reliability was very high [r = 0.997, p < 0.001]). The common variance between the agility tests was 45%, indicating that they represented relatively independent skills. There was a large correlation between RS and defending agility (r = 0.625, p = 0.004), and a very large correlation with attacking agility (r = 0.731, p < 0.001). Defending and attacking agility have different characteristics, possibly related to the footwork, physical, and cognitive demands of each. Nonetheless, RS seems to be important for agility, especially for attacking agility.