On Meet the Press, no one noticed: What kinds of crazy "facts" can a person hear at the very top of our news chain?
Consider the indignities visited on those who watched yesterday's Meet the Press. Hugh Hewitt was part of the pundit panel. Things quickly went sideways from there:
TODD (6/28/20): Hugh, how would you advise the president to turn this around? I mean, it does look like at this point they've made the decision the federal government isn't going to own the response [to the pandemic]. I mean, Secretary Azar kept bringing it back to the states, back to the states. I understand that's a federalism response, but it's not working.Hewitt storylined further from there. But good God! Just look at the numbers he cited!
HEWITT: Well, yesterday 500 Americans died, Chuck. And in Germany, 680 Germans died. The United States' death toll has dropped dramatically from May, when it was 2,700...
Hewitt's current death count for the U.S. was reasonably accurate. According to the Washington Post's numbers, 546 deaths from coronavirus were recorded on Saturday, June 27.
That death count was reasonably accurate—but how about the crazy number he cited for Germany? "In Germany, 680 Germans died?"
Obviously, that number was crazily wrong. But nobody said a freaking thing about Hewitt's ridiculous statement—not Chuck Todd and not his other two pundit guests.
How crazy was Hewitt's number? Anyone who knows anything about this topic knows that Germany has stifled the coronavirus to a state of semi-extinction.
According to the WorldOMeter site, daily deaths in Germany have been averaging maybe 10 per day. The notion that Germany had suddenly pulled a 680 made no earthly sense.
(In population, Germany is roughly one-fourth our size.)
Still, the claim emerged from Hewitt's mouth. Later that day, in a tweet, he explained:
HEWITT (6/28/20): I was wrong on @MeetThePress this morning. Germany had 680+ new cases yesterday not deaths. My error entirely not @NBCNewsHewitt had deaths mixed up with cases! That's why he blurted that crazy number, a number suggesting that things aren't nearly as bad as the Trump-haters might have you think.
A journalist has to know nothing at all to think that Germany could have pulled a 680. We think of something NAME WITHHELD said to us, two decades ago, concerning the total lack of preparation displayed by many journalist guests on C-Span's Washington Journal.
(No, it wasn't Mortman.)
You have to know nothing at all to make a statement like Hewitt's. That said, no one corrected or challenged him on the air and, in the very next segment, Chuck Todd was offering this:
TODD: Welcome back. Data Download time. Let's take a look at how the United States is doing compared with other countries in controlling the coronavirus.The problems continued from there. Already, lack of clarity reigned.
At 107 cases per one million people, the United States has one of the highest infection rates in the world. Brazil's is higher, with a seven-day average of about 163 new infections per one million. Russia and India are also in double digits per one million people. And, of course, we can't be sure how accurate the reporting is in all cases...
Does the United States currently have "107 cases per one million people?" Actually, that would be the number of new (confirmed or reported) cases on a daily basis.
A viewer could have gleaned that information from the small print on the chart which appeared behind Todd. But Todd's entire presentation was very poorly written.
Just within the chunk we've posted, we were especially struck by this statement: "And of course, we can't be sure how accurate the reporting is in all cases."
We certainly can;t be sure of how accurate the reporting is when it comes to American cases! Just last week, the CDC reported that the actual number of cases, overall, may be as much as ten times the current reported number. "Cases" is a slippery metric, for several basic reasons and in several different ways.
Coronavirus statistics can be very hard! If you want to report accurate information, you have to be careful about you say.
In this case, Todd was conveying basic narrative, but the text he was reading was poorly composed. Earlier, Hewitt's unchallenged statement was a miracle of complete/total incomprehension.
While we're at it, we'll refer you to Kevin Drum's discussion of the recent divergence in U.S. statistics between "daily cases," which are rising, and "daily deaths," which have been dropping.
For pro-Trump reasons, Hewitt was trying to make more of the drop in deaths than anyone sensibly could. In recent weeks, we've noted several people overstating the daily number of deaths, perhaps for opposite narrative reasons.
Largely because deaths are "a lagging indicator," Drum says their number could "turn upward next week." Here's the bad news:
At least on a tiny level, daily deaths have already turned upward! The upward movement so far is slight. It may turn out to be a statistical blip, signifying nothing much beyond the lack of a steady drop.
But the seven-day rolling average has in fact inched upward in the last few days—and yes, we're adjusting for the infusion of deaths which suddenly occurred last Thursday, when New Jersey dumped the results of a change in its accounting system into the daily number.
To what extent have deaths inched up? Consider:
At the end of May, American deaths were averaging 915.6 per day (May 25-31). By June 25, the average was all the way down to 567.1 per day (June 19-June 25).
Since then, the seven-day average has nudged up to where it stands at present, 580.7 (June 22-June 28). That may turn out to be nothing much. But the rise in the past few days represents the first time the average has risen since we started keeping track last month.
At any rate, there was Hewitt, reporting live and direct from Pundit Know-Nothing Land. The people you see on your TV shows often know nothing at at all—except, of course, for the glorious shape of their prevailing narratives.
Those they memorize and recite. Quite often, that's pretty much all.
One last factual point: U.S. daily deaths never averaged 2700 in May, or at any earlier point. According to the Post numbers, the daily count went that high on four occasions, though that may have been an artifact of glitches in the day-to-day recording system.
The daily average of deaths exceeded 2000 per day for at most two weeks in April. The average was never as high as 2000 per day at any time in May.
Hewitt was cherry-picking and overstating in service to pro-Trump narrative. He overstated with respect to the American numbers. With respect to the German number, it was Upper-End Pundit Gone Wild.