Background: It is commonly believed that diet and exercise are important for weight loss, but the relative importance of each is unclear, and clinical trials have yielded conflicting results. Aims: To evaluate the degree to which aerobic exercise (AE) and resistance training (RT), with or without dietary changes, lead to weight loss in obese adults. Methods: MEDLINE, Embase, and Web of Science were searched from inception through November 2017. Studies were limited to randomized trials reporting weight loss outcomes with either aerobic exercise (AE) or resistance training (RT), or each with or without diet (D) interventions (AE + D, RT + D, respectively). Included studies evaluated the effect of these interventions (duration ≥ 12 weeks) in overweight or obese (BMI ≥ 25) adults without potentially limiting medical conditions such as diabetes or coronary artery disease (CAD); studies of postpartum or lactating women were excluded. Subgroup analyses and meta-regression were used to calculate differences between modalities. Random effects or mixed models were used for primary outcomes. Results: Of 368 identified articles, 40 randomized trials, including 3,885 subjects (66% female) met inclusion criteria and were analyzed. Across four groups of weight loss interventions (AE, RT, AE+D, RT+D), there was a significant difference between mean weight loss across all groups, -1.6 kg ± 0.3 (AE), -1.5 kg ± 0.7 (RT), -8.6 kg ± 0.7 (AE + D), and -6.2 kg ± 1.3 (RT + D), p<0.001, although weight loss with AE or RT, in the absence of diet changes, was comparable and small in magnitude (-1.6 kg ± 0.3 (AE) and 1.5 kg ± 0.7 (RT), p=0.88). Among studies incorporating dietary changes, a trend of superiority of AE compared to RT was seen; -8.6 kg ± 0.7 (AE + D) vs. -6.2 kg ± 1.3 (RT + D), p<0.095. Subgroup analyses showed greater weight loss in any study involving both diet changes and exercise (AE + D or RT + D) compared to studies involving exercise only (AE or RT alone), -1.3 ± 2 kg vs. -0.35 ± 0.07, respectively. Meta regression adjusting for exercise duration (mins/week) was performed and confirmed the small contribution of exercise to weight loss in either exercise only (1-AE, 2-RT) or exercise + diet (3-AE+D, 4-RT+D) groups (Figure). There was no significant difference in weight loss between subjects allocated to any aerobic exercise (AE with or without diet changes) compared to those allocated to any RT (RT with or without dietary changes), -4.2 ± 0.7 kg (AE or AE + D) vs. -3.7 ± 1.1 kg (RT or RT + D), p 0.69. Conclusion: These results suggest that dietary changes are the primary driver of weight loss, with exercise playing a less important role. Aerobic and resistance training appear to be equivalent, and do not induce clinically-significant weight loss in the absence of dietary changes.