Can we walk and chew gum at one time?: When we read the reporter's identity line, we knew we were in good hands.
Her report appeared on July 3 at the New York Times. The report appeared online only. It never appeared in print editions of the Times.
Her report concerned a "counterintuitive" fact. Even as coronavirus "cases" are rising nationwide, coronavirus deaths continue a steep decline.
Below, you see the headlines which appear above her report. You also see the reporter's identity line:
U.S. Coronavirus Cases Are Rising Sharply, but Deaths Are Still DownWith credentials like those, we could hardly go wrong! As it turned out, WITHHELD graduated from Stanford in 2014. She received her doctorate from Harvard four years later.
This seemingly counterintuitive trend might not last, experts said. But the nation can still learn from the decline.
[NAME WITHHELD] is a reporter for The Times, where she covers science and health. She holds a Ph.D. in microbiology and immunobiology from Harvard University.
Today, she reports for the Times. Right in her first two paragraphs, she—or perhaps some meddling editor—decided to tell us this:
WITHHELD (7/3/20): After a minor late-spring lull, the number of confirmed coronavirus cases in the United States is once again on the rise. States like Arizona, Florida and Texas are seeing some of their highest numbers to date, and as the nation hurtles further into summer, the surge shows few signs of stopping.Frankly, we were puzzled.
And yet the virus appears to be killing fewer of the people it infects. In April and May, Covid-19, the disease caused by the coronavirus, led to as many as 3,000 deaths per day, and claimed the lives of roughly 7 to 8 percent of Americans known to have been infected. The number of daily deaths is now closer to 600, and the death rate is less than 5 percent.
For the prior three days, the 7-day average of daily deaths—except for some specialized purpose, there's no other sensible way to report this statistic—had actually been closer to 500 than to 600.
More significantly, we were puzzled by the claim about the "3,000 deaths per day" which had sometimes been recorded in April and May.
We checked the New York Times database to which the reporter linked. When we did, we saw that those data included exactly zero days when the reported death count in the U.S. ever went as high as 3,000.
Then we clicked the other link in those two paragraphs. When we did, we were taken to this capsule report by David Leonhardt—a July 2 capsule report which actually said this:
"Coronavirus deaths in the U.S. have been falling for most of the last 10 weeks—to about 600 a day recently, down from more than 2,000 in late April."Even Leonhardt may have been pushing it a bit with that "about 600 a day" statistic. But somehow, WITHHELD (or her editor) saw Leonhardt say "more than 2,000 in late April" and decided the story would be better if that accurate statement was changed to this more exciting claim:
"In April and May, Covid-19 led to as many as 3,000 deaths per day."That wasn't what the Times database said. That wasn't what Leonhardt had said.
But now, it was what the New York Times said! Indeed, as Leonhardt's accurate statement was drowned in the harbor, the online report was executing the press corps' great motto and watchword:
No embellishment left behind!No embellishment left behind! It's a time-honored way of life.
Do mainstream reporters always embellish? Actually, no, they don't.
That said, our own reporting at this site is "all anthropology now." And the world-class experts with whom we consult assure us that the impulse to exaggerate is very much bred in the bone of our war-inclined, tribal species.
They refer us to Professor Harari's endlessly best-selling book. Our brains are wired to induce us to tell thrilling tribal tales, these despondent experts all say.
Concerns about "accuracy" are a much later cultural addition routinely honored in the breach. Or so these scholars insist.
Whatever! We offer this as an introduction to the latest actual data.
In his own July 3 post, Kevin Drum had once again noted the somewhat puzzling battle between cases and deaths, making this accurate statement:
"As we all know, the number of COVID-19 cases is skyrocketing but the COVID-19 death rate is continuing to decline."So Drum correctly said. (Admittedly, he was referring to something which very few people know.)
We've been following these data too, in part due to our fascination with the way this topic is being reported:
On cable, deaths have virtually disappeared. Only "cases" are now discussed.
In print, "cases" has become the default statistic, replacing the previous "deaths." And everywhere, statistics tend to be overstated. Sometimes, if it weren't for all the overstatements, there would be no statements at all!
Are we humans able to walk and chew gum at the same time? As a species, could we handle a public discussion in which we were told about the rise in cases and about the decline in deaths, with possible explanations offered?
Would we be able to handle that? Anthropologists say we probably could—but they despondently add that this hypothesis will likely never be tested.
At any rate, here we go:
The rise in (recorded/confirmed) cases is indeed a major, significant fact. Will this rise in (recorded) cases eventually lead to a renewed rise in deaths?
Life is easier, and more exciting, when such questions aren't asked. But just for the record, here are some 7-day rolling averages, including the 7-day rolling average as it currently stands:
Daily deaths from covid-19, nationwide;(We're using the Washington Post's numbers. We've gone back and made adjustments based on changes the Post has made in the numbers it originally posted.
7-day rolling average
May 18-May 24: 1136.9
May 25-May 31: 916.1
June 1-June 7: 806.4
June 8-June 14: 713.6
June 15-June 21: 586.4
June 24-June 30: 538.7
June 29-July 5: 477.1
(We've adjusted for the 1,854 retroactive "probable" deaths New Jersey dumped into the system on June 25. The current average may turn out to be artificially low due to reduced reporting over a 3-day holiday weekend.)
The 7-day average still exceeded one thousand deaths per day as late as May 25 (1093.7 deaths per day, May 19-May 25). Since then, it has dropped all the way down to the current 477.1 deaths per day. That number may be artificially low, though probably not by any giant amount.
Don't get us wrong! Even that reduced number is a sign of our rolling national failure.
Elsewhere, there are very few daily deaths at this point. That's true in Germany, and even in initially hard-hit Italy and France.
Elsewhere, daily deaths are few. Our own vastly reduced number of deaths remains a sign of our national failure, a failure which leads directly to the disordered person who currently sits in the White House.
All roads lead to Donald J. Trump, whose apparent cognitive and/or psychiatric impairment our "press corps" will won't discuss. Still, we'd like to see a serious attempt to analyze the "counterintuitive" coexistence of the rise in cases and the decline in deaths.
Why haven't nationwide deaths begun to rise again? We'd like to see that basic question discussed—but at such moments, despondent anthropologists rush to remind us of our species' inheritance:
"We simply weren't wired for such discussions," these despairing scholars exclaim, ominously referring to our specie in the past tense. At that point, they turn and shamble back into their caves. Soon, loud weeping is heard.
Is that death rate likely to rise again? We'd like to see the question discussed.
But that would require a serious discussion of an obvious question—and that had ceased to be a part of our culture back when Arianna sewed that fourth button on the disfavored candidate's deeply troubling suit. It represented a 33% increase in deeply troubling buttons!
Remember, our work at this site is all anthropology now! If you want to hear preapproved tribal tales, you'll have to go somewhere else.
Why haven't deaths been shooting up...: Why haven't deaths been shooting up, even in states where "cases" have been skyrocketing?
Provisional answers have floated around. Silly us! We wonder which ones may be correct!
At any rate, the doctorate came from Harvard itself. But sure enough! Right there in paragraph 2, the statistical claim was embellished!
Even after twenty-two years, we still find the pattern surprising. But its incidence never declines, and we're told that it never will!