This article focuses on the impact of increased duration of exercise therapy on functional recovery after stroke. A comprehensive literature search using multiple databases was used to identify all relevant randomized controlled trials. Their quality was reviewed by two independent assessors, and a narrative systematic review and meta-analysis was completed. Methodological quality of all the 20 randomized controlled trials (RCTs) identified had a median of 6 points (range 5–8) on the 10-point PEDro scale. A meta-analysis was completed for studies that had a common outcome measure. For each outcome measure, the estimated effect size (ES) and the summary effect size (SES) were expressed as standardized mean differences (SMD). The results of the meta-analysis demonstrated that increased duration of exercise therapy time has a small but positive effect on activities of daily living as measured by the Barthel Index (SES 0.13; CI 0.01–0.25; Z = 2.15; p = .03) and that these improvements are maintained over a 6-month period (SES 0.15; CI 0.05–0.26; Z = 2.8; p = .00). Pooling reported differences in the various upper and lower extremity outcome measures demonstrated no significant SESs. However, the meta-analysis is supportive of the hypothesis that additional, focused exercise on the lower extremity has a favourable effect on lower extremity impairment and walking speed. The narrative review raises a number of issues that need to be considered in the development of future RCTs.