The visualization is based on graphics by Derrick Van Gennep (twitter.com/DerrickVanGenn2) for the international volunteer coalition Endcoronavirus, endcoronavirus.org/countries.
The data is smoothed out using the moving average to show the general trends. To see net official data, switch from average to raw data display.
The location of a country in one of three groups is automatically determined based on several criteria. I do not intervene manually in the order of individual countries. For instance, Ecuador showed a negative number of new cases on May 8, so it was at the very end of the list.
For country ratings in the lists I use a combined rating based on the amount of growth, the rate of growth change, and the accumulated number of confirmed cases. It is calculated automatically on the JHU data.
For example, a country may be able to stabilize growth, but the high number of patients accumulated still represents a risk. Or for countries with low absolute values, the high growth rate of these values also increases a country's position in the ranking. It's dynamic.
I have no influence on the source of data: if official statistics for a certain country give questionable figures, I do not change the display. The data is fully displayed in real time, updated daily.
The height of the charts is normalized: here I prioritize the form of the process rather than absolute values. However, the absolute number of daily new cases is displayed in color of individual line segments. The darker the line, the more cases were confirmed on that day.
The visualization shows only those countries where the total number of confirmed cases exceeds 1000.