Late registration has a long history as a practice and subject of research in American higher education. Approximately 10% of course registrations at colleges and universities are late. Prior research has produced mixed results on the impact of the practice on student success. This study sought to identify a predictive model that would guide students and administrators on the decision to add a course late. It examined the effect of late registration, term length, delivery method, and age on final course grade. The data source was a population of 2013--2017 archival course enrollment data from three private liberal arts colleges and universities in the southeastern and midwestern United States. The statistical method of multiple linear regression was applied to the data. The regression models failed to yield strong predictions of the association between registration variables and student success. While this study did not demonstrate a negative interaction between late registration and student success statistically, it sheds light on the approximately two million students who register late for courses annually and the number of them who are not successful. Researchers should focus their attention on institutional characteristics that, when coupled with late registration, negatively impact many students nationally each year. Professional ethics should compel institutions to develop thoroughly student-centered policies and procedures that lessen or counter the impact of late registration, particularly in time-compressed courses.