Follow the # of new deaths day by day. Right now there is still a lot of noise but if it smooths out and heads north we will know where we are.
the best numbers are the deaths as you say. All the rest are estimates. Because most people recover, the #1 problem is how many will die, so projecting this is the goal This days to a number appears to be one way. We’ll see.
I’m looking at plotting weekly deaths on a semi-log chart. For the US the numbers aren’t old enough yet to project and compare to the regular flu and the H1N1 pandemic.
In a few weeks, both should give us some answers.