The “obesity paradox” was reported in patients with stroke. We aimed to evaluate the pattern and magnitude of association between body mass index (BMI) and prognosis of stroke in patients with minor ischemic stroke or transient ischemic attack (TIA). A total of 5163 patients with available BMI data from the Clopidogrel in High-risk patients with Acute Non-disabling Cerebrovascular Events (CHANCE) trial were included. Patients were classified into three groups according to their BMI values: normal weight (< 23.9 kg/m2), overweight (24–27.9 kg/m2), and obese (≥ 28.0 kg/m2). The efficacy outcomes included a new stroke (ischemic or hemorrhagic), poor functional outcome defined as modified Rankin scale ≥ 2 points and death from any cause within 90 days. The interaction effects were determined using multivariable Cox or logistic regression models. After 90 days of follow up, there were 513 new strokes. Overweight (BMI 24–27.9 kg/m2) patients had a higher risk of recurrent strokes than those with normal weight (10.8% vs 8.8%; HR = 1.24, 95% CI 1.02–1.50) after adjusting for the baseline covariates, but no significant association was observed for those who were obese (P = 0.37). No significant association was found between being overweight or obese and poor functional outcome or death. For patients with a minor ischemic stroke or TIA, being overweight was associated with an increased risk of recurrent stroke compared to being of normal weight in our study. Trial registration: URL:http://www.clinicaltrials.gov. Unique identifier: NCT00979589. © 2019, Springer Science+Business Media, LLC, part of Springer Nature.