If all your friends jumped off their couches, would you? According to new research on exercise “contagion,” chances are that you would. Past studies on the spread of certain lifestyle choices—from altruism to overeating—have shown that habits can travel like viruses, for good or bad. But this is the first study to rigorously analyze exercise contagion on a large scale. Using fitness tracker data posted to a social site by more than 1 million U.S. runners, researchers analyzed the distances, speeds, and total time they ran, covering 350 million kilometers over 5 years. But how to measure whether social network “friends” could influence one another? First, scientists had to tease out confounding effects like the propensity to run more on weekends. So they let the weather set up their experiment, on the understanding that nice weather is likelier to make people run. They found that as people in cities with nice weather went for especially long runs, their friends in unaffected cities would extend their own runs, increasing their running distance and time and burning more calories. That suggests that at least some of that extra exercise is thanks to social influence, the researchers write today in Nature Communications. Men appeared more susceptible than women: They were strongly affected by male friends and moderately affected by female friends, whereas women were moderately affected by other women and unaffected by men. But less active runners also had an outsized influence on others. As lazier runners put in more time or distance, their friends from afar responded in kind. The researchers say their findings suggest that to encourage healthy behavior, you should target an entire social network, not just one individual. After all, what’s a race with only one person?