Background: Stroke is a leading cause of perinatal brain injury and cerebral palsy. Term neonatal hemorrhagic stroke (NHS) is a common syndrome with poorly defined epidemiology. We aimed to determine incidence and mechanisms within a large population-based NHS sample. Methods: The Alberta Perinatal Stroke Project (APSP), a provincial registry ascertained NHS cases using exhaustive ICD-9/10 code searching (1992-2012, >2400 chart reviews). Prospective cases were captured through the Calgary Pediatric Stroke Program from 2007-2014 (n=387). All NHS cases underwent structured chart review using a data capture form and blinded review of neuroimaging. Provincial live births were obtained from statistics Canada. Outcomes were extrapolated to the Pediatric Stroke Outcome Measure (PSOM). Results: We identified 74 cases: 49 NHS (26 retrospective, 27 prospective), 4 presumed perinatal HS (PPHS), and 21 hemorrhagic transformation (HT) of ischemic injury. Incidence of NHS was 1:8800 live births (1:5820 for all forms). HT was common (28.4%) including global, arterial venous ischemic lesions. Presumed perinatal hemorrhagic stroke presented with epilepsy. No risk factor was identified in 68% of cases. Outcomes were abnormal (PSOM 1 or more) in 30% and better in the HT group. Conclusion: NHS occurs in 1:8800 live births. Imaging classification is essential to define mechanisms.