Proving the Unprovable: The White House Ties Itself Into A Statistical Knot Over Covid-19 Harm

The problem with never admitting a mistake as President is that it requires others to defend it no matter how indefensible.  That is the problem with declaring that “99 percent” of U.S. coronavirus cases are “totally harmless”  is that statistics are tricky things that often demand actual proof.  Mark Twain once said “facts are stubborn things, but statistics are pliable,” just not this pliable.  Rather than just admit that the President overstated this point, the White House proceeded to try to prove the unprovable with predictably ridiculous results. Even the President’s top health advisers refused to support the statement. It is another example of the expenditure of unnecessary energy and focus to avoid admitting a mistake.  One can still maintain that most people exposed to this virus show mild or no symptoms without dying on this statistical hill (with graphs that actually show that the statistical claim is wrong).

I have been critical of the unrelentingly negative coverage of many news outlets over the last three years.  There is a loss of objectivity in much of the reporting as we have discussed in prior postings. However, many of these negative pieces are legitimate stories like the President telling the public that 99 percent of virus cases are harmless.

During his July 4 “Salute to America” speech on the South Lawn of the White House, Trump noted that the administration conducted more than 40 million coronavirus tests. That is an impressive figure. However, he added that “by so doing, we show cases, 99 percent of which are totally harmless.”

On its face, the statistic is clearly wrong.  It is certainly true that a very high percentage of those who test positive for the virus do not show serious symptoms.  However, President Trump elected to put a statistical twist on that fact.  As a result, it was immediately ridiculed.  The President has been repeatedly criticized (often fairly) for downplaying the threat of the virus or making dangerously ill-informed comments.  This is another example of sending a message directly at odds with what his public health officials are trying to convey to the public.

Putting aside the public health dangers for such statements dismissing the threat to individuals, it makes even less sense politically. With the President struggling with dropping polls, it would seem obvious that fighting on this issue would play directly into the hands of his critics.  Yet, rather than walk it back, White House press secretary Kayleigh McEnany defended the statement as factual with two charts illustrating the country’s COVID-19 death rate.

There were two obvious problems. First, just showing a chart of deaths does not show the percentage of harm.  Many of those who recover from the virus are experiencing long lasting symptoms. Moreover, even those with “mild cases” report some serious symptoms, including neurological symptoms now being documents even in mild case recoveries.

Second, McEnany’s slides showed a case fatality rate of 4.6 percent, not 1 percent. So even if you are just talking death rather than harm that is almost 500 percent off.

What is intriguing about this latest pile up is that it was again entirely avoidable with a simple correction. This is why Benjamin Disraeli said “There are three types of lies — lies, damn lies, and statistics.”