Interactive Maps and ChartsThese graphics visualize data from Wikipedia's tracking of confirmed COVID-19 cases, reported by state governments. You can hover over them for more details. Automatic updates happen at 9pm Eastern Time.I've chosen to chart these numbers because I haven't come across these particular data being visualized on more comprehensive sites (many of which are listed at the bottom.) (For some reason, Google docs sometimes publishes nonsensical, extreme data that doesn't match the actual spreadsheet, especially in the evening. If something looks obviously incorrect, it probably is.) Daily New Cases per 100K residents. (7-day average)i.e. How bad is it? This is the average number of new cases per day adjusted for population size over the past week. At its peak, New York experienced a weekly average of 50 new cases per day per 100,000 residents. (New York's single day maximum was 11,571 new cases on April 15th.) Daily Growth Factor (7-day average)i.e. Is it getting better or worse? Growth factor is the the current day's new cases divided by the previous day's new cases. 1.01 and above means increasing case numbers, 1.0 means no growth, 0.99 and below means deacreasing case numbers. A growth factor of 1.11 means the number of new cases per day has doubled in a week. The value shown is a geometric mean of the previous 7-days to smooth out lack of reporting on weekends and random noise. (When states do not report, it currently causes a divide-by-zero error in my spreadsheet, so some states are greyed out.) Projected New Cases in One WeekThis map combines the data from the previous two maps to give an indication of which states to be most concerned about. High case numbers aren't so concerning if case numbers are decreasing (i.e. growth factor <1.0), nor is a high growth factor especially concerning if the number of new cases is very low. But high new case numbers that are also continuing to grow are a real problem. This map shows what new case numbers might be in 7 days IF cases grow at a rate half way between the state's growth rate over the past 7 days and the national growth rate. Averaging the state and national growth rates reduces the exagerration of highly anomolous data and builds in some likely regression toward the mean. Still, this isn't a highly reliable prediction. Selected States Over TimeComparison of the number of new cases in a number of states that have had the most serious outbreaks at various stages of the pandemic. The bold trend lines are the centered 7-day average, with the noisy daily numbers deemphasized. Deaths per dayNumber of fatalities reported by states each day with 7-day running average trendline. Daily New CasesTo compare to the chart of daily deaths, here are the number of new cases each day, showing the extent to which it parallels, but precedes deaths. (Note that the vertical scales are obviously different.) Daily New Cases in San FranciscoIf you are interested in wading into all the numbers and quite a few more charts, feel free to look at a web version of my complete spreadsheet. At this point, there are many people using maps and charts to visualize much more sophisticated data analysis than I can do. These are some that I find especially informative: |