The purpose was to determine the day-to-day variability of total testosterone (tT), total T4 (tT4), total 3,5,3'-T3 (tT3), and cortisol (C) levels in exercise-trained men [no.=63, 24+/-4 yr (X+/-SD)]. Resting blood samples were collected at the same time of day (morning) over 3 consecutive days, 24 h apart, under controlled conditions. Subjects were instructed to minimize mental stress and physical activity and to maintain a consistent diet, and sleep patterns during the period of blood collections. Radioimmunoassay procedures were used to measure all hormones. Repeated measures analysis of variances were used to test for differences within each hormone across days, and intra-class correlations (ICC) and typical error were used to assess reliability of hormone concentrations across days. Results revealed resting hormonal values were within accepted clinical ranges. No significant mean differences (p>0.05) from day-to-day were found in any of the hormones. Furthermore, each hormone displayed highly significant (p<0.01) ICC values; tT 0.891, tT4 0.861, tT3 0.705, C 0.886, and consistent daily typical errors. These findings suggest resting tT, tT4, tT3, and C concentrations are not highly variable and appear reliable from day-to-day in exercise-trained men. If researchers measure any of these hormones at rest, before and after (ie, 24-48 h later) an experimental exercise treatment, they can presume that any changes seen at the 24-48 h point (that exceed the typical error appropriately) are due to their treatment and not an artifact of day-to-day variability.