The purpose of this analysis was to assess the potential association between body mass index (BMI) and change in gambling symptom severity in a sample of non-treatment-seeking problem gamblers over a one year period. One-hundred sixty problem gamblers age 18–29 were recruited from the surrounding community and completed assessments of gambling behavior and related variables at baseline and one-year visits. Hierarchical multiple regression was used to examine the association between BMI and symptom change, while accounting for nicotine use, gender, age, and baseline gambling severity. The sample included 102 normal weight, 35 overweight, and 23 obese gamblers. Groups differed in sex, age, gambling risk level, and current nicotine use. There was a significant obesity x baseline gambling risk level interaction for the model predicting changes in GD-YBOCS scores over the 1-year follow-up. Obese participants with gambling disorder showed significantly smaller improvements in symptoms over one year than normal weight participants. Results suggest that obesity predicts smaller improvements in gambling symptoms over time in those meeting criteria for gambling disorder. Future research should clarify the possible influence of impulsivity on this association, and assess the clinical utility of weight management programs in treatments for gambling disorder.