I made additional changes to Florida’s data to provide more accuracy than my previous change, and much more accurate than what @COVID19Tracking adds to the US total, which excludes from the denominator everyone who has ever tested from counting again in the current test number.
I reached out to @COVID19Tracking to pass along the pitfalls of using the day-over-day difference in "total people ever tested once" as the current number of people tested today. The worst part about using this as a denominator is that it gets worse over time.
As of today, there are 4.8M+ people in Florida who, if they are tested again tomorrow, next week, or next month, will not count using the less accurate data. With tens of thousands of them testing daily, it's really skewing the test numbers (and the resulting % testing positive).
Even my newest method has flaws that likely makes the % testing positive higher, since it combines both PCR and antigen (like some other states). Since FL's antibody rate is almost certainly now higher than the daily PCR testing rate, it's likely nudging the PCR number up a bit.
But it's still much more accurate than using the daily difference in the FL dashboard number.
Hopefully @COVID19Tracking fixes this issue on their end. But regardless, I'll be incorporating better numbers on my end.
Here was my initial thread on the FL changes (which I've altered based on the above):
Share this Scrolly Tale with your friends.
A Scrolly Tale is a new way to read Twitter threads with a more visually immersive experience.
Discover more beautiful Scrolly Tales like this.