When we launched Charting Covid in March, there weren't resources for finding out how COVID was trending in your county. Today, there are many such resources, resources that are maintained by folks working on them full time.
For an excellent presentation of county-level data, see this resource by the NY Times.
We all need to do our part to avoid catching and spreading COVID-19. Here, we attempt to visualize how the places we all live are doing at preventing the virus from spreading.
Disclaimer: New York City data is not broken out by county.
This chart is useful for seeing where on the curve a particular locale is. Since the x-axis shows the number of days since 50 cases were reported in a locale, it allows for comparing the curve across time.
| Row | County | State | Confirmed Cases | Cases Per Capita | Growth Factor | Doubling Time (Days) |
|---|
We're using the New York Times county-level data for reporting this data. Thanks to the Times for making this information publicly available.
The code we used to process and present this data is publicly available: https://github.com/mattparrilla/chartingcovid.
We tried to keep the data as close to the source as we could, however, many counties had entries erroneously reporting the same or a decreasing case count for a given date. This is a known issue in the dataset.
To account for these holes in the data, we used the growth rate (discussed below) to estimate the case count for days with zero new cases. We only did this for historical data and we only did it when we had good case data for adjacent dates.
The original data included cases and deaths. We focused on cases and pulled some additional metrics out of that figure: new cases, growth factor, and doubling time.
New cases is the difference in total cases between adjacent days.
The growth factor is defined as the ratio of new cases in one day to the ratio of new cases the previous day. It is a useful metric for knowing where we are on "the curve". A growth factor greater than 1 implies we are still on the exponential portion of the curve, while a growth factor less than 1 implies new cases are slowing down.
To provide a smoother, less noisy figure we averaged the growth factor for the previous 5 days.
3Blue1Brown has an excellent video explaining the importance of growth factor
The doubling time of cases was calculated for every locale in the dataset for which we had sufficient data. It measures the days that it will take for the number of cases to double at the current growth rate.
The growth rate used to calculate doubling time is the exponential growth rate (as opposed to the growth factor mentioned above).
This website was made by Matt Parrilla and Brian Holdefehr. We're a couple of software guys who wanted to get a better idea for what was going on around us.