Background and objective: new population-level studies are needed to better assess the relationship between physical inactivity and mortality from COVID-19. The aim of the study was to evaluate the correlation between population prevalence of physical activity and standardized mortality rates by COVID-19 in Brazilian capital cities and the Federal District. Methods: this is an ecological study, whose analysis is secondary. The prevalence of physical inactivity, insufficient physical activity, and physical activity during free time was obtained from the Surveillance of Risk Factors and Protection for Chronic Diseases by Telephone Survey 2019 (VIGITEL), according to minutes spent on leisure, commuting, and household activities. The COVID-19 mortality data was obtained from the Influenza Epidemiological Surveillance Information System (SIVEP-Gripe), adding the accumulated deaths until December 31, 2020. The resident population was estimated from the Instituto Brasileiro de Geografia e Estatística (IBGE) for the year 2020. Pearson Correlation evaluated the correlation between the prevalence of different physical activity practices and the standardized mortality rate from COVID-19, in total, and according to age groups. Results: there was a significant positive correlation (r = 0.420; p = 0.029) between the overall prevalence of insufficient physical activity and the standardized COVID-19 mortality rate. No correlation was observed between the other prevalence of physical activity and the standardized mortality rate from COVID-19. Conclusion: there was a correlation between insufficient levels of physical activity and the standardized mortality rate from COVID-19 in people living in Brazilian capital cities.