SCIENCE IS A PROCESS.

Robert Verbruggen gets it.  "[S]cience is a very messy project and it’s best not to get too invested in any single number."  He's referring to the pitfalls of anticipating the effects of a novel coronavirus, in a world where the defenders of conserving capacity in hospitals, which stretches out the duration of the contagion without reducing the contagion, are somehow treating the continued shutdowns as somehow reducing the contagion.  (That's why any good technical paper mentions directions for additional research.)

Rich Lowry extends the argument.

[A]n extraordinary feature of the [novel] coronavirus is how poorly understood it is. We don't know how many people have it, what the death rate is, what the long-term health consequences of having a severe case are, or how best to treat it, among other things.

The models of how the virus would spread and how many would be hospitalized and die were invested with a certainty that they didn't deserve. They were all over the map, and some have failed to accurately predict the course of the disease even a couple of weeks in the future.

Anomalies are like that, which is why I inserted that [novel].  I'd note further that tackling anomalies isn't what science, as we understand it, does very well.  It's relatively easier, and probably more rewarding, to engage in puzzle-solving within a Kuhnian paradigm.  Then comes the refractory phenomenon with immediate consequences and the kind of contemplation over five or ten years that would work many of the anomalies isn't available.