“3,000 people didn’t die yesterday—they were just reported dead”

A dumb COVID minimizer trope

It is true that daily deaths reported on dashboards are not actually deaths that occurred the previous day. It generally takes a few days for deaths to be reported, and so we’re perpetually looking at a blend of deaths (and cases) that occurred in the recent past.

Nobody who has ever looked at the numbers is surprised by this. Deaths and cases do not stop every weekend—reporting simply slows. Well almost nobody is surprised:

From a bizarrely misguided conspiracy theory.

But for some reason, minimizers trot this out this bit of Covidsplaining every time deaths are rising. “No, X people didn’t die, they were just reported dead that day.”

For example, during the second wave, Justin Hart decided to pick a fight with media reporting a one-day record of 120 deaths reported in Florida July 9. When a reporter pushed back on him, pointing out that deaths really were rising to record levels, Justin stood his stupid ground:

Justin Hart @justin_hart

I love it when I'm "technically true" but somehow I'm wrong. The 120 death number was touted in EVERY HEADLINE yesterday implying 120 people died in one day. We called that technically FALSE. We stand by it.

Marc Caputo @MarcACaputo

There's a new bad-faith argument in FL to discount Covid-19's deadliness: the daily death totals reported by the state aren't actual daily deaths because the dates of various deaths differ a little from the reporting dates Technically true But deaths are up Here's the history https://t.co/mtH7LUsVKZ

Justin was right! Turns out 120 people did not die of COVID-19 in Florida that day.

128 did!

Justin’s complaint was both bad-faith and false:

Plot by the excellent @zorinaq

If you stop to think about it for a second, this isn’t surprising; we don’t get real-time data, but rising numbers do indicate growth. In fact, growth is understated by rising numbers, because each day’s report blends deaths/cases from days and sometimes weeks earlier, when the situation was not as dire.

This leads to really poor-timed pronouncements by COVID minimizers.

Alex Berenson @AlexBerenson

The headline: “5,000 Americans died of Covid last week!” The reality: “5,000 Americans who had tests showing they were exposed to #sarscov2 at some point in the last few months died in the last two months! Those deaths were reported last week!” Also, Orange Man Bad.

It’s always on the upswing when minimizers try to pooh-pooh the significance of record numbers. They either haven’t thought it through or believe their followers won’t be able to point out the obvious flaw in their reasoning.

I expect that cases will rise again as the more infectious strains take hold, and if they do, I hope fewer people get fooled again.

The reverse effect is seen when cases decline, as they are in most states and European courtiers. Cases and deaths will continue to look worse than they actually are for a few days because health departments are still catching up on past reporting. So when the line trends down, the real-time drop is likely even more substantial.

An example of reporting lag via Minnesota Public Radio

This is one reason jurisdictions have been quick to open up in the face of rapidly declining test results. The real-time data—if it were superhumanly possible to know—is almost certainly even better than the positive trend-lines suggest.

Minimizers less frequently complain about the lagged death effect. In fact, right now many of the COVID minimizers are twisting themselves into pretzels arguing that reopening in the face of collapsing case and hospitalization figures must be political. As I wrote last week, it’s not. Just weeks ago, the same people who now say the reopening are only “political” advocated reopening when the numbers were unambiguously worse.

Conditions have simply improved. I wish I could say forever, but probably not.