Background: Smartphone dependence is increasingly prevalent among Chinese college students and has been linked to poorer sleep, reduced academic engagement, and elevated emotional symptoms. Physical exercise is frequently proposed as a protective behavioural factor, yet most studies report only direct associations and leave the underlying mechanism unexamined. Objective: This study tested whether self-control mediates the association between physical exercise and smartphone dependence among Chinese college students. Methods: A cross-sectional survey was conducted with 400 Chinese undergraduates recruited through stratified cluster sampling. Physical exercise was assessed with the International Physical Activity Questionnaire–Short Form, self-control with the Brief Self-Control Scale, and smartphone dependence with the Smartphone Addiction Scale–Short Version. Hypotheses were tested with hierarchical regression and a bootstrapped mediation model (PROCESS Model 4, 5,000 resamples). Results: Physical exercise negatively predicted smartphone dependence (β = −0.27, p < .001) and positively predicted self-control (β = 0.30, p < .001). The indirect effect of exercise on smartphone dependence through self-control was −1.12 (95% CI [−1.60, −0.70]), accounting for 40.9% of the total effect. The direct effect remained significant after self-control was entered (β = −0.16, p = .001). Conclusion: Physical exercise was associated with lower smartphone dependence partly through higher self-control. The cross-sectional design precludes causal inference; longitudinal and intervention designs are needed to confirm the direction of these associations.