Primary dysmenorrhea is cramping abdominal pain associated with menses. It is prevalent, affects quality of life, and can cause absenteeism. Although evidence-based medical treatment options exist, women may not tolerate these or may prefer to use nonmedical treatments. Physical activity has been recommended by clinicians for primary dysmenorrhea since the 1930s, but there is still no high-quality evidence on which to recommend its use. We sought to determine the effectiveness of physical activity for the treatment of primary dysmenorrhea. Systematic literature searches of MEDLINE, Embase, Cochrane, Web of Science, CINAHL, PsycINFO, SPORTDiscus, PEDro, Allied and Complimentary Medicine Database, World Health Organization International Clinical Trials Registry Platform, Clinicaltrials.gov, and OpenGrey were performed, from database inception to May 24, 2017. Google searches and citation searching of previous reviews were also conducted. Studies were selected using the following PICOS criteria: participants were nonathlete females experiencing primary dysmenorrhea; intervention was physical activity delivered for at least 2 menstrual cycles; comparator was any comparator; outcomes were pain intensity or pain duration; and study type was randomized controlled trials. Study quality was assessed using the Cochrane risk of bias tool. Random effects meta-analyses for pain intensity and pain duration were conducted, with prespecified subgroup analysis by type of physical activity intervention. Strength of the evidence was assessed using GRADE. Searches identified 15 eligible randomized controlled trials totaling 1681 participants. Data from 11 studies were included in the meta-analyses. Pooled results demonstrated effect estimates for physical activity vs comparators for pain intensity (-1.89 cm on visual analog scale; 95% CI, -2.96 to -1.09) and pain duration (-3.92 hours; 95% CI, -4.86 to -2.97). Heterogeneity for both of these results was high and only partly mitigated by subgroup analysis. Primary studies were of low or moderate methodological quality but results for pain intensity remained stable during sensitivity analysis by study quality. GRADE assessment found moderate-quality evidence for pain intensity and low-quality evidence for pain duration. Clinicians can inform women that physical activity may be an effective treatment for primary dysmenorrhea but there is a need for high-quality trials before this can be confirmed.