The purpose of the study was to analyze the relationships between sprint swimming performance, dry-land power, and kinematics in master swimmers. Twenty-two male master swimmers were separated in two groups based on their chronological age: (i) 30–39 years and; (ii) 40–49 years. Maximum dry-land power was determined through counter movement jump and 3 kg medicine ball throwing (Hmax and Tmax, respectively). Kinematic determinants of performance were measured during a maximal bout of 15, 25 and 50 m front crawl (T15, T25, T50). Stroke frequency (SF), stroke length (SL) and stroke index (SI) were calculated as kinematical aspects of the stroke. In the 30-39 group, SI25 was correlated to T25 (r = –0.76, p < 0.01, η2 = 0.96), the same was observed between SI50 and T50 (r = –0.83, p < 0.01, η2 = 0.96). Only SI50 was significantly correlated to T50 (r = –0.86, p < 0.01, η2 = 0.97) in the 40–49 years age cohort. In dryland power variables, Hmax and Tmax were only correlated in the younger master swimmers group (r = –0.87, p < 0.01, η2 = 0.97). There were no significant differences (p < 0.05) between younger (30–39 years) and older (40–49 years) swimmers groups in dry-land tests (Hmax 28.5 ± 5.9 vs. 26.5 ± 3.9 cm and Tmax 4.2 ± 1.0 vs. 4.2 ± 1.1 m). Our results suggest that swimming performance in younger master swimmers (30–39 years) seem more dependent on kinematic swimming variables than on strength parameters, which were most related to swimming performance in the older master swimmers (40–49 years).