PURPOSE:Prior studies exploring the reliability of peak fat oxidation (PFO) and the intensity that elicits PFO (FATMAX) are often limited by small samples. This study characterised the reliability of PFO and FATMAX in a large cohort of healthy men and women. METHODS:Ninety-nine adults [49 women; age: 35 (11) years; [Formula: see text]O2peak: 42.2 (10.3) mL·kg BM-1·min-1; mean (SD)] completed two identical exercise tests (7-28 days apart) to determine PFO (g·min-1) and FATMAX (%[Formula: see text]O2peak) by indirect calorimetry. Systematic bias and the absolute and relative reliability of PFO and FATMAX were explored in the whole sample and sub-categories of: cardiorespiratory fitness, biological sex, objectively measured physical activity levels, fat mass index (derived by dual-energy X-ray absorptiometry) and menstrual cycle status. RESULTS:No systematic bias in PFO or FATMAX was found between exercise tests in the entire sample (- 0.01 g·min-1 and 0%[Formula: see text]O2peak, respectively; p > 0.05). Absolute reliability was poor [within-subject coefficient of variation: 21% and 26%; typical errors: ± 0.06 g·min-1 and × / ÷ 1.26%[Formula: see text]O2peak; 95% limits of agreement: ± 0.17 g·min-1 and × / ÷ 1.90%[Formula: see text]O2peak, respectively), despite high (r = 0.75) and moderate (r = 0.45) relative reliability for PFO and FATMAX, respectively. These findings were consistent across all sub-groups. CONCLUSION:Repeated assessments are required to more accurately determine PFO and FATMAX.