Background: Accurate breast cancer recurrence risk perceptions might motivate health-promoting behaviors and alleviate undue anxiety. Although a few studies have examined early-stage breast cancer survivors' perceived risk of recurrence, none have assessed the accuracy of survivors' perceived risk of recurrence. Methods: First primary ductal carcinoma in situ and early-invasive breast cancer survivors reported their perceived risk of recurrence during 6- and 12-month postsurgery interviews. We estimated the patients' 10-year risk of recurrence from published clinical trials, and for early-invasive breast cancer patients, risk of distant recurrence was based on their breast cancer-specific mortality calculated using Adjuvant! Online. Patients' perceived risk was compared with their calculated risk and categorized as "Accurate," "Underestimated," "Overestimated," and "Uncertain. " Multinomial logit marginal effect models were fitted using Accurate as the reference. Results: Only 17% of 531 patients accurately perceived their risk at 6 months, most of whom inaccurately perceived their risk at 12 months (P = 0.0143). Patients who were nonwhite [odds ratio (OR), 1.70; 95% confidence interval (95% CI), 1.12-2.56] and received radiation therapy (OR, 2.01; 95% CI, 1.07-3.77) were more likely to underestimate their risk. Patients with ductal carcinoma in situ (OR, 1.69; 95% CI, 1.08-2.70), lower social support (OR, 0.71; 95% CI, 0.53-0.95), and anxiety (OR, 1.58; 95% CI, 1.01-2.47) were more likely to overestimate their risk. Conclusion: Few breast cancer survivors accurately perceived their risk of recurrence. Impact: The accuracy of perceived risk may be increased by better physician-patient communications about their prognosis, provision of social support, and treatment for coexisting anxiety.