There is controversy in the literature regarding the dose-response relationship of strength training in healthy older participants. The present study determined training frequency effects on maximum strength, muscle mass and functional capacity over 6 months following an initial 3-month preparatory strength training period. One-hundred and six 64–75 year old volunteers were randomly assigned to one of four groups; performing strength training one (EX1), two (EX2), or three (EX3) times per week and a non-training control (CON) group. Whole-body strength training was performed using 2–5 sets and 4–12 repetitions per exercise and 7–9 exercises per session. Before and after the intervention, maximum dynamic leg press (1-RM) and isometric knee extensor and plantarflexor strength, body composition and quadriceps cross-sectional area, as well as functional capacity (maximum 7.5 m forward and backward walking speed, timed-up-and-go test, loaded 10-stair climb test) were measured. All experimental groups increased leg press 1-RM more than CON (EX1: 3 ± 8%, EX2: 6 ± 6%, EX3: 10 ± 8%, CON: −3 ± 6%, P < 0.05) and EX3 improved more than EX1 (P = 0.007) at month 9. Compared to CON, EX3 improved in backward walk (P = 0.047) and EX1 in timed-up-and-go (P = 0.029) tests. No significant changes occurred in body composition. The present study found no evidence that higher training frequency would induce greater benefit to maximum walking speed (i.e. functional capacity) despite a clear dose-response in dynamic 1-RM strength, at least when predominantly using machine weight-training. It appears that beneficial functional capacity improvements can be achieved through low frequency training (i.e. 1–2 times per week) in previously untrained healthy older participants.