C19 Update: The data, the whole data, and nothing but the data

I developed a new plot today showing the C19 data without predictions. Going forward what will decide if we have a campus open or closed will be state cases, specifically a rising trend of cases will likely trigger closure. It will be interesting to see how the raw data changes over the next few weeks when we get high school football practice restarted, then K-12 school restarted, then higher education restarted. 

My prediction now is that AR is going to see a significant increase in cases over the next 4-6 weeks and these plots will show it day by day. I also needed a form of plot that could expand indefinitely along the time axis, with each y axis linear scale but scientific notation so whether plotting deaths or cases, state or the world, the y tick labels will look reasonable. 

Plot rows from top down represent Arkansas, US and world. Plot columns left to right represent C19 death data and C19 confirmed cases.  Each main plot is the daily count of either deaths or cases as pale blue dots, the last daily value as an oversize black dot, and seven day running average shown as red line. The inset plots show the total of each quantity (un-averaged) with scientific notation on the vertical axis. 

I still think death data is far superior to case data, which has an unknown error bar because of testing protocols that vary with region and time.