OBJECTIVES To assess independent predictors of stent thrombosis (ST) in an all-comer trial. METHODS This is an observational case-control study based on a retrospective analysis of the Basel Stent Kosten Effektivitäts Trial (BASKET) (n = 826). Patients with ST were compared to controls with regard to baseline parameters. Multivariate models were performed to identify independent predictors of ST. RESULTS At 36 months, there were 53 (6.4%) patients with ST, 17 (32%) of whom had early ST and 36 (68%) of whom had late/very late ST. Patients with ST were at a higher cardiovascular risk but received lower doses of statins than the controls (n = 212). Stents in ST patients were longer, had more overlap and were not as well expanded, with significantly more remaining stenoses than the stents in the controls. Multivariable analysis revealed interventions in saphenous vein grafts, malapposed stents, an overlap >3 mm, complex coronary anatomy and treatment with low-dose/no statins as risk factors for ST, while interventions in saphenous vein grafts, underexpanded or malapposed stents, a history of myocardial infarction and treatment with low-dose/no statins were risk factors for late ST. CONCLUSIONS The use of statins might have a protective effect against ST. This observation is new, hypothesis-generating and should be evaluated in an adequately powered randomized trial.