Ischemic and hemorrhagic strokes are established etiological factors for recurrent seizures. Still, only few prospective data are available to predict post-stroke epilepsy and to choose the best point in time and anticonvulsive agent for treatment. In a prospective study we evaluated 264 consecutive stroke patients and assessed their post-stroke epilepsy risk within a follow-up of 1 year. Data on ten risk items concerning the stroke localisation, persisting neurological deficit, stroke subtype, established diagnosis of vascular encephalopathy, early- and late-onset seizures were collected using a post-stroke epilepsy risk scale (PoSERS). All patients underwent brain imaging with either CT, MRI or both and 148 patients underwent electroencephalography. The overall frequency of early-onset seizures within 14 days was 4.5%, of at least one late seizure 6.4% and of epilepsy 3.8%. Chi-Square tests showed significantly higher relative frequencies of seven of the ten clinical characteristics in post-stroke epilepsy patients. The total scale showed moderate sensitivity (70%) and positive predictive value (87.5%) while specificity (99.6%) and negative predictive value (98.8%) were relatively high. The EEG showed little value in predicting post-stroke epilepsy. The PoSERS appears to be a valuable tool to predict the risk for post-stroke epilepsy within the first few days after a stroke.