Our Robert VerBruggen lays out a bunch of good news in the fight against the virus. Health experts think we’re nearing the peak of the outbreak, the daily death count has stabilized, and the evidence suggests social distancing has had the desired impact — the virus is spreading much more slowly than a few weeks ago.

The most commonly-cited model, from the Institute for Health Metrics and Evaluation at the University of Washington, now calculates we hit peak resource use back on April 10, we passed the peak the day of peak deaths, and they think we will have around 68,000 total deaths by summer — well below the earlier projection of, at minimum, 100,000 deaths.

With the outlook looking much better than the models projected just a few weeks or even just a few days ago, you’re hearing a lot of complaints that “the models were wrong.”

You Might Like

The models are trying to account for a million little variables — how many people have it, how many people have it and don’t know it yet, how many more people the infected will come in contact with, the immune system strength of the people who have it and who will catch it, and so on.

The final death toll will be determined by millions of tiny decisions — does this person stand too close to another, does this other person leave their house, does this third person wear a mask, does this fourth person cough into their hand and then touch a doorknob. It’s impossible to know all of them, so the model statisticians take their best guess based upon the data they have so far. When you only have Monday’s data, it’s tough to get a good sense of what the situation will be on Friday. But once you have Tuesday’s data, Friday gets a little bit clearer, and the same thing happens Wednesday, and by Thursday, you have a much better sense of what Friday will be like.

But some of these small, unpredictable decisions can have far-reaching consequences. Imagine three people are on the beaches in Florida, and a woman unknowingly spreads SARS-CoV-2 to one of the two men she talks to, but not both. The first man lives in Immokalee, and the second lives in Kings Point.

If she infects the man who goes home to Immokalee, a small unincorporated community southeast of Fort Myers, the consequences are likely to be mild. Immokalee is the third-youngest community in the state, with a median age of 24.7 years. The population density of the community is just under 2,500 people per square mile. Most of the residents of that community are young and hopefully relatively healthy, and even if our Florida Man unknowingly spreads it, the odds are good that the vast majority of the people he encounters will not need hospitalization or an ICU unit, and the odds of people dying are pretty low.

Now imagine that instead, the virus spread to the man who lives in Kings Point, north of Boca Raton. Kings Point is tied for the oldest median age of any community in the state, at 77.7 years. There are about 6,700 people per square mile. A person unknowingly spreading coronavirus in this community would have serious consequences, and the many elderly residents of that community would be much more likely to require hospitalization and ICU beds — and would be much more likely to succumb to the virus.

No model can perfectly account for all of this, which means all of them represent a best guess — a projection based upon known data and current trends, not a prediction or a declaration from an oracle.