Objectives: The pieces of evidence regarding whether metabolic syndrome (MetS) is a better predictor than its individual components, especially diabetes, for recurrent stroke are limited. This study aimed to examine these associations. Methods: A total of 1087 ischemic stroke patients were recruited consecutively from 2003 to 2004. They were followed up until the end of 2008. Baseline clinical and laboratory characteristics and new stroke event during follow-up were recorded. MetS was defined by the definition issued by the Chinese Medical Association/Chinese Diabetes Society. Results: One hundred forty-three new stroke cases were recorded. After adjusting for baseline age, gender, education, marriage status, subtype stroke, length of index stroke to baseline assessment, history of cardiac diseases, smoking status, drinking status, clinics, aspirin treatment, and fibrinogen by Cox regression models, the risk of recurrent stroke was 43% higher in MetS patients than in non-MetS patients (hazard ratio [HR] = 1.43, 95% confidence interval [CI]: 1.01-2.01). The strength of this association is weaker than MetS individual components such as elevated glycemia (adjusted HR = 1.78, 95% CI: 1.26-2.52), elevated blood pressure (adjusted HR = 1.91, 95% CI: 1.11-3.30), or low high-density lipoprotein cholesterol (adjusted HR = 1.57, 95% CI: 1.08-2.51). Compared with the group with neither MetS nor diabetes, the adjusted risk of recurrent stroke was highest in the group with diabetes (HR = 2.77, 95% CI: 1.66-4.63), followed by those with both MetS and diabetes (HR = 1.91, 95% CI: 1.25-2.94). The risk of recurrent stroke in patients with MetS in the absence of diabetes was similar to those with neither. Conclusion: MetS is not superior to its individual components in predicting future recurrent stroke in patients who experience mild-to-moderate ischemic stroke.