I have two sets of small multiples below for 2 month time frames ...
Top: State trends in cases / deaths from mid-Mar to mid-May
Bottom: State trends in cases / deaths from early May - today
A few obs ...
1) Increases in deaths are delayed by a few weeks, which is obvious with the first spike in Louisiana
2) Cases are not reported by test date ... this is frustrating to use
3) A deaths spike should follow the ones in the bottom visual within the week IF they cases-deaths spike trend from earlier follows BUT I would have expected to see some increase by now
4) Increases in hospitalizations give me some confidence that a subsequent deaths spike will happen
5) I have read we are treating it better than before and more cases are young persons. It is possible, dare I say likely, that the spike in deaths will not be nearly as high as back in Mar-Apr
I would be interested to know how seriously those 60 and older are taking staying at home and not getting out. The folks I know at highest risk are still not going into public. It may be that, given the relative youth of new cases, that older generations are simply being smarter about their risks. If this death trend stays like this I don't think anyone can see that as bad news.
The cases and deaths trend for the US as a whole from 3/1 forward ...