Thirty-six sedentary, mildly obese (30-40% fat), premenopausal women (29-49 yr) were randomly assigned to one of two dietary regimens for an 8-wk double-blind, parallel study of the effect of moderate caloric restriction and macronutrient variation on maximal physical performance. Group 1 (N = 19) consumed 4186 kJ.d-1 (1000 kcal.d-1) with a diet composition of 40% fat, 40% carbohydrate (CHO), and 20% protein. Group 2 (N = 17) also consumed 4186 kJ.d-1 but varied the percent kilojoules to 20% fat, 60% CHO, and 20% protein. Maximal aerobic power was measured using a modified Balke protocol, strength was assessed using isotonic bench and leg press machines and a Cybex 340 isokinetic device during knee flexion and extension, and muscular endurance was taken as the number of sit-ups performed in 1 min. Pre- to post-testing revealed significant (P less than or equal to 0.05) reductions in body weight (group 1, -7.4 kg; group 2, -6.5 kg) within both groups. No significant changes were seen in maximal aerobic power within both groups. Strength measures for both groups showed a trend toward reduction but statistical significance was only evident in right knee extension for group 1 and left knee extension for group 2 (P less than or equal to 0.05). Sit-up number increased nonsignificantly for both groups. Between group differences were not found in any body weight, body composition, or physical performance index. We therefore conclude that in mildly obese women, maximal exercise performance is relatively unimpaired after 8 wk of caloric restriction with either of two diets differing moderately in their macronutrient content.