The aim of this study was to identify the relationship between dry-land and in-water strength with performance and kinematic variables in short-distance, middle-distance, and repeated sprint swimming. Fifteen competitive swimmers applied a bench press exercise to measure maximum strength (MS), maximum power (P), strength corresponding to P (F@P), maximum velocity (MV), and velocity corresponding to P (V@P) using F–V and P–V relationships. On a following day, swimmers performed a 10 s tethered swimming sprint (TF), and impulse was measured (IMP). On three separate days, swimmers performed (i) 50 and 100 m, (ii) 200 and 400 m, and (iii) 4 × 50 m front crawl sprint tests. Performance time (T), arm stroke rate (SR), arm stroke length (SL), and arm stroke index (SI) were calculated in all tests. Performance in short- and middle-distance tests and in 4 × 50 m training sets were related to dry-land MS, P, TF, and IMP (r = 0.51–0.83; p < 0.05). MS, P, and TF were related to SR in 50 m and SI in 50 and 100 m (r = 0.55–0.71; p < 0.05). A combination of dry-land P and in-water TF variables explains 80% of the 50 m performance time variation. Bench press power and tethered swimming force correlate with performance in short- and middle-distance tests and repeated sprint swimming.