The mortality rate, or death ratio, is all over the ballpark and yeah, we don’t really know what it is yet. And you touched on a key problem: Are we talking about deaths as a % of known cases or total cases? Those two figures will be quite different. Also, an average figure, if we had one, would be somewhat misleading, because the risk will be dramatically different for a child or teen vs. a young adult vs. an older person. I’ve written about it all here (and given the complexities, I’m avoiding mentioning the current best estimate here, because I don’t think it should be stated without all the context): https://medium.com/luminate/covid-19-infection-mortality-rates-can-be-misleading-636c7af0a213