Why the Current Narrative that
“Florida is Improving” is a Lie
If you browsed the New York Times Coronavirus tracker on August 9 (data through 8/8/2020), you’d likely conclude that Florida is “getting better”. After all, on the front page of the Latest Map and Case Count for the US, Florida shows up in a section called “Where new cases are decreasing”. Click deeper to the Florida details, and you see a nice graph which indeed appears to show that cases are dropping significantly. The obvious conclusion is the pandemic is coming under control.

https://www.nytimes.com/interactive/2020/us/florida-coronavirus-cases.html
This obvious conclusion would be wrong. It fails to take into account that “Confirmed Cases” is an easily manipulated statistic. Unfortunately, virtually the entire mainstream press bases its reporting on confirmed cases and is being fooled.
In Florida, as you can see in the Time’s graph, the 7-day average of Confirmed Cases peaked on July 17 @ 11,870 and dropped to 6,550 on August 8, an apparent decrease of 44.0% (See: Note 1).
Between these same dates, the 7-day average number of tests dropped 42%! Note: [Positive Tests] = [Confirmed Cases] by definition in this Florida data. So, if the total number of tests administered drops, and the Positive Test Rate stays the same, cases must drop.
In Florida’s case, Positive Test Rate was 17.5% on August 8 and 18.1% on July 17. As a result, 95% of the apparent reduction in Florida cases is due simply to the reduced number of tests. It’s why The Covid Tracking Project’s graphs for both New tests and New cases cases depict virtually identical trends during July and August (except for slightly different scaling).

Thus, by any reasonable standard, the Florida pandemic remains uncontrolled and extremely dangerous. The best you can say for it is that the Positive Test Rate is no longer climbing (albeit stabilized at a shockingly dangerous level).
Think Florida is alone in doing this? Think again. We looked at every state plus the District of Columbia over the most recent 5-week period (Weeks ending July 7 through August 4). All of the 10 states with the highest positive test-rate cut back testing during the most recent week. This means they all reported fewer new cases than if they’d held testing levels constant.

Contrast these states to the 10 with the lowest Positive Test Rate: 6 of these best-performing states tested the most during the latest week. Tests/100K averaged much higher as well: for example, NY, with a Positive Test Rate of 1.0%, tested at a rate of 2,401 tests/100K population. This means it tested at a rate more than 100% higher than 5 of the 10 worst-performing states, and 43% higher even than GA (the most aggressive-testing state in this group).
The fact is that every state in the worst-performing 10 cut back on testing, despite long-standing WHO protocols to increase testing when positive rates are this high. Nine out of these worst-performing states have Republican governors. It’s hard to credit that this behavior is an accident. Could they be heeding the President’s advice to “Slow the testing down”?
Note 1: This calculation (and those that follow) are based on the author’s original analysis using data released by the Covid Tracking Project. The author’s calculation of seven-day average case counts for Florida based on the Tracking Project data were identical to the NY Time’s number for August 8, and differed by a count of 5 cases (11,870 vs. 11,865) for July 17. https://covidtracking.com/data/state/florida