By: Mark Glennon*

Can you walk across a river with an average depth of five feet?

Put in terms that simple, the fallacy of relying on an average should be obvious. Knowing the average doesn’t help. It depends where you cross and how tall you are.

But that same fallacy constantly appears in discussion about COVID-19 and policies to fight it. Averages that mean little have been overdone, nationally and here in Illinois.

Most recently, the COVID debate is moving in a new direction that demands better awareness of the fallacy of averages.

This is important not only to the coronavirus debate, but provides a broader lesson why primary schools ought to be requiring students to read books like Innumeracy: Mathematical Illiteracy and Its Consequences, How to Lie With Statistics and Thinking Fast, Thinking Slow. Too many people, including much of the media and government, seem unaware of the fallacy.

The fallacy most commonly appears in discussion about fatality rates. From the start of the pandemic, the average mortality rate from an infection has understandably been central to perceptions about the virus, and initial estimates almost always were provided as a single average. They were scary. Headlines were common in the spring with words like “staggering death tolls” of 1.3%. In March, Dr. Anthony Fauci estimated the mortality rate at about 2% and the World Health Organization pegged it at about 3.4%.

Estimates gradually dropped for a number of reasons, but a single, average number continued to be the focus. Until last month the Center for Disease Control published only one average number, which it put at 0.65% in July.

Hold on, many scientists are now saying. Focus on what’s inside those averages. Thousands of scientists and medical practitioners are now signing on to what’s called the Great Barrington Declaration, asking for what they call “focused protection.” Signers include what the Wall Street Journal calls “dozens of esteemed medical experts with blue-chip academic credentials.” Their statement says,

The most compassionate approach that balances the risks and benefits of reaching herd immunity, is to allow those who are at minimal risk of death to live their lives normally to build up immunity to the virus through natural infection, while better protecting those who are at highest risk. We call this Focused Protection. 

Underpinning their approach is the starkness of the age variance within the “average” fatality rate. Below is the breakdown newly provided by the CDC. This is its “current best estimate” of chances of dying if you get infected. Death rates are clearly far, far lower than originally said for most of the population. The averages were distorted up largely because most of the deaths are older people. In Illinois, 86% of deaths have been over age 60.

Stay laser-focused just on those at risk and leave younger people alone because they face no material risk – that’s what’s behind the new declaration. Since April, that’s the approach we’ve often said makes sense.

Instead of that focused protection, however, policy in Illinois and much of the nation would be better described as carpet bombing – flatten the entire economy by restricting behavior of the entire population. In fact, Illinois’ approach might be considered the opposite of focused protection since its policies toward retirement facilities were so poor, as reported by the Chicago Tribune. Over half of Illinois COVID-19 fatalities have been in those facilities.

The fallacy was first evident in a different way in Governor JB Pritzker’s initial shutdown order and reopening plan, which lumped the entire state into one unit. Regions with no COVID problem howled, and Pritzker gradually switched over to a more regionalized approach instead of state-wide averages.

But the state is still struggling under an inconsistent approach to the fallacy of averages.

St. Clair County’s positivity rates are better than the rest of its region, it insists, so it should stand alone in how restrictions are imposed. But the state is refusing to allow the county to be counted separately from Region 4. The county emergency management director says he sometimes feels like he is “in a war” with the Illinois health department, according to the Belleville News-Democrat.

And that’s inconsistent with how the state is dealing with Region 6. There, positivity rates are pulled way down by heavy testing at the University of Illinois – over 10,000 per day according to the News-Gazette. Heavy testing gets lots of negative results. So, the state decided to exclude the U of I county from Region 6 numbers. That has the rest of Region 6 squealing about the tougher restrictions it will face because of the exclusion.

That whole dust-up with excluding U of I, by the way, illustrates the silliness of focusing on positivity rates at all. They are heavily distorted by how much testing is being done and who is getting tested.

If all that is too much math for you, flunk with dignity and at least remember the main lesson: Don’t let anybody tell you whether it’s safe to walk across a river based on its average depth.

*Mark Glennon is founder of Wirepoints.

Oldest Most Voted
Inline Feedbacks
View all comments
3 months ago

Another little math lesson for you: never worry about average when it’s the total number what matters. Hospitals have the capacity they have. They either get overwhelmed, or not. They don’t treat “average patients”. They treat the people they get. The whole point of trying to contain coronavirus has absolutely nothing to do with any average death rate, and everything to do with the total number of people filling up hospitals. And the problem is, if hospitals are overwhelmed, they aren’t able to care for other patients that may have other diseases. If the ICUs are full, then nobody can… Read more »

Val W. Zimnicki
3 months ago

“….. flatten the entire economy by restricting behavior of the entire population.”

Just ‘average’ collateral damage which is ‘average’ thinking by our below ‘average’ politicians and bureaucrats…

3 months ago

We are a society of sheep. Just for a moment, accept that intelligence is normally distributed… That means that the average (which, arguably is not very smart) + maybe 50% of 1 STD DEV, doesn’t understand math… Or descriptive statistics… We’re doomed.

And, even worse, 45% or so, do not pay taxes yet get to vote. Really screwed.

3 months ago

Throwing a bomb to Keynesians here: Stimulus doesn’t work. Trump is taking a beating by the left wing media for killing stimulus talks. Unfortunately, he said they can talk about it after the election. Instead, he should have cited this data, and said to governors open up your economies and that will be the best stimulus.

3 months ago
Reply to  Jeff Carter

Getting people back to work is the best stimulus. The problem is NO politician is going to open themselves up to being accused of killing people because they opened their economy.

3 months ago

Lockdowns do NOT work, and are counterproductive to the wellbeing of the community. And yet, Joe Biden wants to impose lockdowns as ‘science’!!!

Edward Wezain
3 months ago

What’s that Mark Twain line, “Figures don’t lie, but liars do figure!”

accept reality
3 months ago

If Illinois, NY, CA (et al) policies were applied to zoos, all the animals would room free while the visitors viewed from the cages.

3 months ago

I hope that people are finally starting to understand that the total lockdown approach never made a lot of sense. Thanks for your good work on this. Second the recommendation on “Innumeracy”, it is an important book