GARBAGE IN, GARBAGE OUT?

Michael Fumento has long been a skeptic of faddish medicine, and he's at it again.  "After Repeated Failures, It’s Time To Permanently Dump Epidemic Models."  Among the authorities he cites is the president's contagion czar.

Then [Dr. Anthony] Fauci finally said it. “I’ve spent a lot of time on the models. They don’t tell you anything.” A few days later CDC Director Robert Redfield also turned on the computer crystal balls. “Models are only as good as their assumptions, obviously there are a lot of unknowns about the virus” he said. “A model should never be used to assume that we have a number.”

Which, of course, is exactly how both a number of public health officials and the media have used the them.

Only one significant model appears to have been correct. But wasn’t. The University of Washington’s Institute for Health Metrics and Evaluation has actually been dramatically reduced and reduced.

That model isn't even statistical inference as we understand it. But statistical inference is difficult in the early days of an outbreak.  At this point, somebody with a home computer and a simple logit model could start estimating contagion curves, and perhaps in another month of repeated runs, day by day, something resembling a handle on the curve with a large prediction error might be available.

All that statistical legerdemain is unlikely to improve on the simplest explanation.

The only “model” with any success is actually quite accomplished and appeared in 1840, when a “computer” was an abacus. It’s called Farr’s Law, and is actually more of an observation that epidemics grow fastest at first and then slow to a peak, then decline in a more-or-less symmetrical pattern. As you might guess from the date, it precedes public health services and doesn’t require lockdowns or really any interventions at all. Rather, the disease grabs the low-hanging fruit (with COVID-19 that’s the elderly with co-morbid conditions) and finds it progressively harder to get more fruit.

That’s not proof that public health interventions are worthless; merely that since the Plague of Athens four centuries B.C. and before, epidemics have risen and fallen quite on their own. Nobody needed Big Brother looking over their shoulder and cracking a whip; nobody needed to implode their economies and leave their citizens with tops reading: “I survived the ‘worst epidemic in history’ and all I have left is this crummy t-shirt.”

That's too strong: it's still helpful to know that the latest plague is in fact new, it is still prudent to remove yourself, to the extent you are able, from being part of the chain of contagion, and it is still helpful to know how contagious the latest plague is.  In the Wuhan coronavirus, those clusters in rest homes are likely to be with the unwitting participation of asymptomatic carriers, and there are apparently a lot of them, and the people who develop symptoms are more contagious before the fevers and coughs show up.