I look at the worldometer web site pretty frequently to see what’s going on with the pandemic. I’ve noticed something odd with the numbers being reported for Florida, which you can see here:
The number of daily new cases started going up at the end of June, and now they’re at their highest point ever for the state. The number of daily deaths started to go up in early July, which is what you’d expect. But the reported death rate decreased sharply starting around August 5th, and is now about the same as it was before the latest spike.
This makes no sense to me. Why would the number of deaths plummet when the case rate is high and increasing? Other states with high case rates, like Louisiana, show no such pattern. It’s almost enough to make me suspect deliberate data suppression, although in general I don’t believe in conspiracy theories (it’s too hard to keep a secret when so many people would have to be involved).
Worldometers appears to be backfilling older dates so that, for example, if someone died on August 1 but it was reported by the state on August 10, it gets added to the total for August 1. Other data trackers that don’t do this, like this one from the New York Times, show no such dropoff.
Florida changes their reporting methodology recently. Deaths used to show up when they were reported - now they are added to the day of the actual death, regardless of when it is reported. This may result in them not being noticed as much.
Florida changed its COVID reporting method Tuesday and now reports when cases or deaths actually occurred rather than when they were communicated to the state — allowing health officials to assign cases or deaths to days in the past rather than the present.