This research focuses on a detailed exploration of late-onset offending. Using the National Youth Survey, this work seeks to answer three questions. First, is late-onset offending a real phenomenon? Second, if late onset does exist, is the evidence for it conditioned by how we define crime and delinquency? Finally, is late-onset offending an artifact of measurement methodology? Most literature evidencing late onset relies on official police contact and arrest data. Propensity or control theories in general posit that late onset should not exist. Propensity, namely self-control, should be instilled early in life and if absent, results in early initiation into crime and delinquency. Research in developmental psychology seems to support this notion. The findings from this study indicate that late-onset offending is almost nonexistent when self-reported measures are used leading one to conclude that contemporary evidence for late-onset is heavily conditioned by how we measure crime and delinquency. A comprehensive discussion includes future directions for research, and implications for theory development and methodology.