ObjectiveTo examine the effects of strength training on patient-important outcomes of stroke recovery and to quantify the influence of the exercise prescription on treatment effects.DesignSystematic review and meta-analysis.Data sourcesEight electronic databases (MEDLINE, EMBASE, EMCARE, AMED, PsycINFO, CINAHL, SPORTDiscus, and Web of Science) and two clinical trial registries (ClinicalTrials.gov and WHO International Clinical Trials Registry Platform) were searched from inception to 19 June 2024.Eligibility criteriaRandomised controlled trials were eligible if they examined the effects of strength training compared with no exercise or usual care and reported at least one exercise prescription parameter. An advisory group of community members with lived experience of stroke helped inform outcomes most relevant to stroke recovery.ResultsForty-two randomised trials (N=2204) were included. Overall risk of bias was high across most outcomes. Strength training improved outcomes rated as ‘critical for decision-making’ by the advisory group, including walking capacity (standardised mean difference (SMD)=0.95 (95% CI 0.34–1.56)), balance (SMD=1.13 (0.51–1.75)), functional ability and mobility (SMD=0.61 (0.09–1.14)), and habitual (mean difference (MD)=0.05 m/s (0.02–0.09)) and fast-paced walking speed (MD=0.09 m/s (0.01–0.17)), with very low to moderate certainty of evidence, mainly due to risk of bias and inconsistency. More frequent strength training, traditional strength training programmes and power-focused intensities (ie, emphasis on movement velocity) were positively associated with walking capacity, health-related quality of life and fast-paced walking speed.ConclusionStrength training alone or combined with usual care improves stroke recovery outcomes that are important for decision-making. More frequent strength training, power-focused intensities and traditional programme designs may best support stroke recovery.PROSPERO registration numberCRD42023414077.