A theory-based protection and risk model was applied to explain variation in college students' cigarette smoking. Key aims were to examine whether psychosocial and behavioral protective and risk factors can account for cross-sectional and developmental variation in smoking, and to examine whether protection moderates the impact of risk on smoking involvement. Data for this three-wave longitudinal study were collected in fall 2002, spring 2003, and spring 2004 from 549 male and 427 female first-semester college students at the University of Colorado, Boulder. A 32-page questionnaire was used, with content theoretically derived from the constructs in problem-behavior theory. Cigarette smoking (number of cigarettes smoked on an average day in the past month), three types of psychosocial protection (models protection, controls protection, support protection), three types of psychosocial risk (models risk, opportunity risk, and vulnerability risk), two types of behavioral protection (church involvement, academic achievement), and two types of behavioral risk (problem drinking, marijuana use) were assessed. Psychosocial and behavioral protective and risk factors accounted for significant variation in smoking involvement, and protection moderated the impact of risk. Findings were consistent, for the most part, for both genders and across three separate waves of data. Key predictors of smoking involvement included controls protection, models risk, vulnerability risk, behavioral protection, and behavioral risk. Antecedent protective and risk factors were associated with the initiation of smoking in the college setting. A model of protective and risk factors can be useful in understanding college smoking behavior and suggesting targets for intervention.