Background Admission stroke severity is an important clinical predictor of stroke outcomes. Pre-stroke physical activity contributes to stroke prevention and may also be associated with reduced stroke severity. Summarizing the evidence to date will inform strategies to reduce burden after stroke. Aims To summarize the published evidence for the relationship between pre-stroke physical activity and admission stroke severity and to provide recommendations for future research. Summary of review MEDLINE, Embase, Emcare, CENTRAL, and gray literature databases were searched on 14 February 2020 using search terms related to stroke and pre-stroke physical activity in adult stroke survivors. We screened 8,152 references and assessed 172 full-text references for eligibility. We included seven studies ( n = 41,800 stroke survivors). All studies were observational, assessed pre-stroke physical activity using self-reported questionnaires, and assessed admission stroke severity using the National Institute of Health Stroke Scale. Analyses were categorized as the presence of pre-stroke physical activity (four studies) or dose-response (five studies). In three studies, presence of pre-stroke physical activity was associated with milder stroke severity, and no association in one study. Greater pre-stroke physical activity duration and intensity (two studies) or amount (three studies) were associated with milder stroke severity. Studies ranged between moderate to critical risk of bias, primarily due to confounding factors. Pre-stroke physical activity may be associated with reduced risk factors for severe stroke, distal occlusion, smaller infarcts, and shorter time-to-treatment delivery. Conclusion Pre-stroke physical activity may be associated with reduced admission stroke severity. Lack of randomized controlled trials limited causality conclusions. Future research recommendations were provided.