This study examined the effect of cycle sprints as a potentiating stimulus for power and strength adaptation in semi-elite athletes. Eighteen rugby players were assigned into training groups that completed either a 40-s cycle sprint (T(SPRINT)) or rested (T(CONTROL)) before each workout (n = 6-8) of a 4-week programme. Squat jump (SJ) peak power (PP) and mean power (MP), and box squat (BS) one repetition maximum (1RM) strength were assessed every workout. Saliva was collected across each workout and assayed for testosterone (Sal-T) and cortisol (Sal-C). The T(SPRINT) and T(CONTROL) groups both showed significant improvements in SJ PP (8.2 +/- 2.9 vs. 11.9 +/- 3.6%), SJ MP (11.8 +/- 2.6 vs. 18.6 +/- 4.8%) and BS 1RM (20.5 +/- 2.6 vs. 23.2 +/- 1.3%), respectively. However, there were no group differences in training adaptation, workout performance or the workout hormonal responses. As a combined group (all players), significant relationships were demonstrated between resting Sal-T and/or Sal-C concentrations and absolute SJ power (r = 0.20-0.30) and BS strength (r = 0.36-0.44) across all workouts. For individual players, the respective relationships with SJ power (r = 0.22-0.42) and BS strength (r = 0.41-0.49) were, on average, found to be stronger. In conclusion, leg workouts performed with or without prior cycle sprints can produce similar power and strength improvements in semi-elite rugby players. Resting salivary hormone concentrations appear important for workout performance, especially for individuals, thereby potentially moderating training adaptation.