Physical activity (PA) and sedentary behavior (SB) levels in healthy adults are predominately based on self-reporting measures, which generally overestimate PA but underestimate SB. Patients with coronary artery disease (CAD) eligible for cardiac rehabilitation (CR) follow an individualized program; thus, objective assessment of physical performance and regular daily activity is required. This study aimed to compare self-reported and objectively measured PA and SB in patients with CAD prior to out-patient CR. We included 91 patients with CAD and assessed their PA with an accelerometer for 8 days prior to CR, along with the short form of the international physical activity questionnaire. We found that most patients were sedentary (61%, ~8 h/day), and on average performed 63 min/day of moderate-to-vigorous-intensity physical activity (MVPA). Males performed less daily light-intensity physical activity (−5%, p = 0.011) and performed more MVPA (+2%, p = 0.002) compared to females. Maximal aerobic capacity was significantly associated with MVPA (Spearman rho = 0.483, p < 0.001) and MVPA > 10 min bouts (Spearman rho = 0.391, p < 0.001). Self-reported measures overestimated MVPA (total MVPA, +108 min/day, p < 0.001; MVPA > 10 min bouts, +152 min, p < 0.001) and underestimated SB (−174 min/day, p < 0.001) compared to objective measures. There was no significant correlation between methods in MVPA (Spearman rho = 0.147, p = 0.165)), MVPA > 10 min bouts (Spearman rho = −0.059, p = 576), and SB (Spearman rho = 0.139, p = 0.187). Quantitative analysis demonstrated the huge proportional bias for MVPA, MVPA > 10 min bouts, and SB. Our findings demonstrate that self-reported physical activity provides inaccurate estimates of MVPA and SB in patients with CAD entering the ambulatory CR. This strongly supports the more objective assessments of daily PA, preferably using an accelerometer.