Maximal fat oxidation (MFO) and the exercise intensity that elicits MFO (FatMax) are commonly determined by indirect calorimetry during graded exercise tests in both obese and normal-weight individuals. However, no protocol has been validated in individuals with obesity. Thus, the aims were to develop a graded exercise protocol for determination of FatMax in individuals with obesity, and to test validity and inter-method reliability. Fat oxidation was assessed over a range of exercise intensities in 16 individuals (age: 28 (26-29) years; body mass index: 36 (35-38) kg·m-2; 95% confidence interval) on a cycle ergometer. The graded exercise protocol was validated against a short continuous exercise (SCE) protocol, in which FatMax was determined from fat oxidation at rest and during 10 min of continuous exercise at 35%, 50%, and 65% of maximal oxygen uptake. Intraclass and Pearson correlation coefficients between the protocols were 0.75 and 0.72 and within-subject coefficient of variation (CV) was 5 (3-7)%. A Bland-Altman plot revealed a bias of -3% points of maximal oxygen uptake (limits of agreement: -12 to 7). A tendency towards a systematic difference (p = 0.06) was observed, where FatMax occurred at 42 (40-44)% and 45 (43-47)% of maximal oxygen uptake with the graded and the SCE protocol, respectively. In conclusion, there was a high-excellent correlation and a low CV between the 2 protocols, suggesting that the graded exercise protocol has a high inter-method reliability. However, considerable intra-individual variation and a trend towards systematic difference between the protocols reveal that further optimization of the graded exercise protocol is needed to improve validity.