The metric used is a little hand to understand.
“Mean daily case growth rate” expressed as a percentage. I think that means at the right end of the chart that new cases were still growing at 10% a day (averaged over all the states). But that is inconsistent with the news that new cases are declining, which would be a negative percentage figure. It may be the date of the study — it looks like it ends on April 30.
I look forward to reading the actual paper. Maybe it will be clearer.
Not hard at all. Mean daily growth rate for a given day is the number of new cases that day divided by the cumulative number of cases that were identified prior to that day. For instance, assuming no more cases reported for today (I’m writing this with just over an hour left on 4/9) there were 30,313 new US cases today. Yesterday’s cumulative total was 434,927. Today’s growth rate is 30313/434927 = 0.0697, or just under 7%.
Note that the daily growth rate can never be negative. It can be zero when there are no new cases, but it will be positive if any new cases are identified. Also a decline in daily growth rate does not imply a decline in the number of new cases. To illustrate, assume the numbers above remain accurate for 4/9. That would give a new cumulative total of 465,240. To maintain the same growth rate on 4/10, there would have to be 465240 x 0.0697 = 32,425.717 new cases (obviously the number of new cases can’t be a decimal). That means that 32,425 or fewer new cases tomorrow would be a decline in daily growth rate.