Physical activity may decrease gastroesophageal cancer risk through a reduction of oxidative stress and decreased chronic inflammation, yet few epidemiologic studies have been able to report a clear inverse association between physical activity and gastroesophageal cancer. Because no meta-analysis has investigated the relation of physical activity to gastroesophageal cancer, we conducted a comprehensive systematic review and meta-analysis according to the PRISMA guidelines based on 24 studies with a total of 15,745 cases. When we compared high versus low physical activity levels and summarized associations according to anatomic site and tumor histology, risk reductions were evident for esophageal adenocarcinoma [relative risk (RR) = 0.79, 95% confidence interval (CI) = 0.66-0.94], gastric cardia adenocarcinoma (RR = 0.83, 95% CI = 0.69-0.99) and gastric non-cardia adenocarcinoma (RR = 0.72, 95% CI = 0.62-0.84). The risk reduction for esophageal squamous cell carcinoma (RR = 0.94, 95% CI = 0.41-2.16) became statistically significant (RR = 0.66, 95% CI = 0.46-0.96) after excluding an influential study. The test for heterogeneity by gastroesophageal cancer subtype was statistically non-significant (p-difference = 0.71). The RR of total gastroesophageal cancer for high versus low physical activity was 0.82 (95% CI = 0.74-0.90). A dose-response analysis of frequency of physical activity and total gastroesophageal cancer risk revealed that the greatest risk reduction was achieved among those engaging in moderate to vigorous physical activity five times per week (RR = 0.67, 95% CI = 0.58-0.79). Our results provide support for an inverse relation of physical activity, in particular exercise frequency, to gastroesophageal cancer risk.