Visualizing the Growth of COVID-19 in the U.S., Organized by State Peak Date

The exponential nature of viral spread means that pandemics are fast-moving and dynamic.
Combine this with the high interconnectedness of modern life—even when social distancing and lockdowns are applied—and pandemics can evolve quickly. In just a few weeks, previous hotspots can cool down, while new high risk areas can crop up seemingly out of nowhere.
In the United States, like many other places in the world, the virus is hitting regions differently, and this landscape is constantly changing over time.
COVID-19 Growth, by State
Today’s first visualization above comes to us from Reddit user bgregory98, and it uses data from the New York Times to plot confirmed active COVID-19 cases by state.
States are organized by the date that weekly average cases peaked, from top to bottom. Data is normalized and is current until June 16th, and states are colored based on regional definitions (i.e. Northeast, Midwest, West, South) as defined by the U.S. Census Bureau.
As you can see, when looking purely at active cases, the situation has evolved considerably from a geographical perspective.
Early on, COVID-19 cases were more concentrated in coastal population centers, especially in the Northeast. New York, New Jersey, and Massachusetts, the three hardest hit states per capita so far, saw cases peak in April.
However, a look at the bottom half of the visualization shows that generally, states in the South and West are starting to heat up with cases. Recent daily numbers confirm this, with California, Texas, Florida, Arizona, and North Carolina all gaining more than 1,000 new cases on June 17th.
Growth by State, Part Deux
The following visualization by Reddit user jawsem17 is designed using a similar concept, and is current as of June 17th.
This version uses the same data set from the New York Times. However, it also includes deaths as a metric, showing a comparison of peak deaths to peak cases for each state.

Although one would expect peak deaths to follow peak cases, this is not always the case.




