wtf is covid-19-data?

This repository contains a comprehensive record of Covid-19 cases and deaths across the United States, tracked since the start of the pandemic. Originally maintained by The New York Times, it provides daily cumulative counts at the national, state, and county levels, along with specialized datasets covering prisons, college campuses, mask use surveys, and excess deaths. The data comes in simple spreadsheet-like files (CSV format) that anyone can open or analyze. There are two main categories: historical files with finalized end-of-day counts for every day since January 2020, and live files updated multiple times daily with current-day numbers that may be partial. The live files also break down cases into confirmed (lab-tested) and probable (diagnosed by symptoms and exposure). County and state records include FIPS codes, which are standard geographic identifiers that make it easy to combine this data with maps or population datasets. Researchers, journalists, public health officials, and data analysts would use this data. A journalist could build an interactive map showing how cases spread through specific counties over time. A researcher could compare death rates between states with different policies. A founder building a health dashboard could pull these numbers to display trends. The pre-computed rolling averages are especially useful for anyone making charts or graphics, since the raw daily numbers can swing wildly due to reporting quirks. What stands out is the human effort behind the data. Dozens of journalists compiled it by monitoring press conferences, analyzing government releases, and contacting local health officials across more than 50 states and territories. The README documents the messy reality of American public health reporting, where states changed methodologies mid-pandemic, sometimes reported numbers differently, and occasionally moved cases between counties without explanation. The team made judgment calls, like counting patients where they were treated rather than where they lived, and revising historical entries when better information arrived.