Background: Bone marrow stimulation (BMS) is presently considered first-line surgical treatment for osteochondral lesions of the talus (OLTs); however, some patients still experience pain or dysfunction after surgery, and the reasons for success or failure remain somewhat unclear. This study aimed to investigate the effect of smoking on postoperative outcomes after arthroscopic BMS for OLTs. Methods: Consecutive patients with OLTs who underwent BMS between January 2017 and January 2020 were included. Smokers were defined as patients who actively consumed cigarettes before surgery and postoperatively, whereas nonsmokers were patients who never smoked. Visual analog scale (VAS), American Orthopaedic Foot & Ankle Society ankle hindfoot score (AOFAS), Karlsson-Peterson, and Tegner scores were assessed preoperatively and at follow-up. Additionally, a general linear model (GLM) was performed, followed by the interaction analysis to explore the potential influence of smoking. Results: The study enrolled 104 patients with a mean follow-up of 30.91 ± 7.03 months, including 28 smokers and 76 nonsmokers. There were no significant differences in patient age (35.2 ± 10.0 years vs 37.6 ± 9.7 years, P = .282) or OLT area (63.7 ± 38.7 mm2 vs 52.8 ± 37.0 mm2, P = .782). Both univariate analysis and GLM revealed that smoking was associated with worse postoperative pain levels, Karlsson-Peterson, and AOFAS scores ( P < .05). The interaction analysis showed a significant interaction between smoking and OLT area for postoperative Karlsson-Peterson scores (general ankle function) ( P = .031). Simple main effects analysis revealed that the negative effect of smoking on Tegner score significantly increased among patients >32 years old or with OLT area>50 mm2 ( P < .05). Conclusion: Smoking was associated with worse clinical outcomes following BMS of OLTs. As the size of OLTs increased, the difference in general ankle function between smokers and nonsmokers also increased. Furthermore, smokers who were older than 32 years or had larger OLTs were less likely to resume participation in high-level activities.