Boysforlife has a legit point. It isn't a matter of simple math when we just don't know how many people have it. There was a study which suggested that 86% of coronavirus patients are undiagnosed. That means it isn't 685 of 52,996. Instead, if one believes that study, it is 685 out of roughly 380,000. Even if that study is half accurate, it still improves the survival rate significantly. That 35 million number of people who got the flu doesn't come from 35 million people all going to the doctor and all getting diagnosed. It comes from years of statistical sampling based on the history of the influenza virus. We are not at the point where can actually do that for Coronavirus.