Cardiopulmonary exercise testing (CPET), particularly maximal oxygen consumption (VO₂max), has been used to assess the outcome for patients with palliated congenital heart disease (CHD). Small studies correlating VO₂max with noninvasive imaging measures of ventricular function have led to the hypothesis that VO₂max reflects cardiac performance. In other settings, physical training is associated with increased VO₂max. The authors hypothesized that habitual exercise correlates with VO₂max to a greater degree than ventricular function in a relatively healthy cohort of patients with palliated CHD. The habitual exercise behavior of 208 subjects with conotruncal abnormalities (tetralogy of Fallot, truncus arteriosus, and interrupted aortic arch) at the time of previously performed, study-based CPET and cardiac magnetic resonance imaging (CMR) were retrospectively assessed via questionnaire. The association of VO₂max with habitual exercise duration and CMR measures of ventricular function was tested. Of 208 subjects, 89 (43 %) completed questionnaires, and 78 % of the 89 patients had a concurrent CMR. The mean VO₂max was 76 ± 21 % of that predicted. The CMR-assessed left ventricular ejection fraction was 67.6 ± 7.4 %. The hours of habitual exercise per week correlated with VO₂max (p < 0.001; r ² = 0.14). No association was found between right and left ventricular ejection fraction, cardiac index, or right ventricular end-diastolic volume and VO₂max. In this study, VO₂max correlated with habitual exercise to a greater degree than CMR measurements of ventricular function. These findings highlight the importance of considering the contribution of noncardiac factors when exercise data in both clinical and research settings are interpreted. The contribution of these factors to clinical outcomes deserves further study.