Eager to get America's businesses up and running again, presumably including his own, President Trump on Monday evening suggested that the country managed to maintain economic activity despite tens of thousands of deaths in car accidents each year. Why shut down over the coronavirus pandemic when we stay open despite car crashes?

“You look at automobile accidents, which are far greater than any numbers we're talking about,” Trump said. “That doesn't mean we're going to tell everybody no more driving of cars. So we have to do things to get our country open."

He was not the first to draw this analogy, nor was it the first time that he had done so. It is, however, a bad analogy.

When you hear someone compare the number of deaths that will result from the coronavirus pandemic to the number of car accidents in the United States, your first response should be something along the lines of, how do you know how many people will die from the pandemic?

We’ve sort of accepted that this comparison between the death toll from the virus and the 35,000 to 40,000 annual deaths from car accidents is a fair one. At the moment it isn’t; the number of people who have died from covid-19, the disease caused by the virus, is in the hundreds, not the tens of thousands. The comparison itself assumes that we will see at least a hundred times as many deaths from the virus.

Trump assures us that the number of deaths from car accidents is “far greater than any numbers we’re talking about” with the coronavirus, but he has no way of knowing that. That, from the outset, reveals a central problem with this and other comparisons. Maybe it will be the case that the death toll from the virus lands at around 40,000. Some estimates, though, suggest that the death toll will be more than an order of magnitude higher, perhaps over 1 million over the course of the pandemic. That total would occur over multiple years and assumes that we don’t take significant steps to limit its spread. Which, of course, is what the debate is about: how long measures need to be in place to limit the death toll.

So from the outset, this comparison to car accidents falls apart simply because we don’t know the scale of deaths from the virus. We do know something about the virus that makes obvious how bad the comparison is: that it’s a virus.

We know that, unlike car accidents, coronavirus cases — and, therefore, deaths — can expand exponentially. Someone with the virus can infect two to three others. Out of every hundred people who contract it, experts believe one will die. That rate of spread has meant that the number of coronavirus cases in the United States has jumped to nearly 50,000 from 8,000 a week ago.

To fairly compare car accidents to the virus, then, we would need to imagine that the bounds of the problem were similar. We’d need to imagine, for example, that from March 4 to March 14, the number of deaths from car accidents increased by a factor of 10. And that, from March 14 to March 24, it had done so again. Do you think that if the number of vehicle deaths increased 100-fold over two weeks there wouldn’t be a sudden push to limit driving?

Imagine that most cars used a shared operating system to control internal systems. Suddenly, one day, something goes haywire with the code and the system starts forcing cars to accelerate uncontrollably. We'd suddenly see a lot of deadly accidents and a lot of close calls.

Now imagine if we discovered that the problem could be broadcast over cars’ internal WiFi networks to other vehicles nearby, allowing the problem to jump from one car to the next. We would see places where the problem was worse, like on crowded highways. Very quickly, there would be restrictions about when and where you could drive your car. After all, something in the expected order had changed, and needed to be addressed.

Where the coronavirus-to-car analogy really breaks down, though, is that it ignores one of the immediate threats posed by the virus. When we talk about 40,000 deaths in car accidents, we're talking about that happening nationally over the course of a year. When we talk about the deaths from covid-19, we're talking so far about a spike in deaths in certain regions over a short period of time.

The scenario we’re in now is that there have been 1,000 serious car accidents on an interstate in a medium-sized city and now the drivers and their passengers are being rushed to several nearby hospitals. The hospitals are triaging patients and figuring out who needs what care, but they are very quickly out of beds in the emergency department and out of doctors to care for the patients. The sheer scale of the problem means that people will die waiting for a bed or waiting to be triaged.

People sitting in the waiting room with other ailments might die, too, unable to secure any of the limited resources. Even this analogy doesn’t go far enough, though, because it doesn’t account for the risk of doctors and health-care staff catching the coronavirus themselves, putting them at risk and reducing the available resources at the hospital.

It’s understandable that people would want to go back to the lives they enjoyed a month ago. I would like to do that, too; I’d like to go to a restaurant or take my son to a museum. I would like it to be the case that covid-19 offered no more substantial risk than driving in a car. But that’s not the case. Advocates of reverting to normal use the car-crash example specifically because it feels like a bit of normalcy for which we assume certain risks. But what’s happening now simply isn’t comparable, and comparisons between car accidents and the coronavirus are facile to the point of being ludicrous.

“I think that’s a false equivalency to compare traffic accidents with — I mean, that’s totally way out,” Anthony S. Fauci, a leading voice on the president’s coronavirus task force, said Friday. “That’s really a false equivalency.”

Ideally, the current measures we’ve taken (at Fauci’s recommendation) will limit the number of deaths from the virus to somewhere around the number of car accident deaths and, ideally, a vaccine or treatment is available by fall that limits the effects moving forward. But that ideal outcome would not change the fact that now, at this moment, comparing the virus to car accidents is wildly off the mark.