Smoking tobacco has been associated with incidence, response and outcomes after treatment of some cancers. We hypothesized that tobacco use could result in an observable effect on breast cancer stage and characteristics at diagnosis. There were 6,000 patients with Tis-4, N0-3 breast cancers who presented to a comprehensive cancer center at initial diagnosis between 1970 and 2006. Patients were included who had a known smoking history, and subdivided into any tobacco use 2683 (45%) or never tobacco use 3317 (55%). Analyses were performed to evaluate the association of smoking with clinical, pathologic and treatment-related factors at cancer presentation. Median age at diagnosis for all breast cancers was 55 years, for nonsmokers was 56 years, for any smoking history was 55 years, and the subgroup of current smokers was 52 years. The difference in median age for current smokers versus nonsmokers was statistically significant (p < 0.0001). The probability of age <55 years at breast cancer diagnosis for any smoking history compared to nonsmokers was 1.2 for white patients (p < 0.0003) but 0.81 for black patients (p = 0.25). There was no statistically significant association between smoking and T stage, N stage, ER/PR status, or Her-2/neu status, although smokers were less likely to utilize breast-conserving treatment. Smoking was associated with a younger age at diagnosis and lower utilization of breast conservation, and observed in the subgroup of white patients but not black patients. Further efforts to clarify potential reasons for any racial differences and lower utilization of breast conservation with smoking are warranted.