But the issue isn’t that clear-cut: Top-line vaccine efficacy numbers are largely misunderstood, and Johnson & Johnson’s vaccine is still plenty effective, according to public health experts.

“So Johnson & Johnson is a very good vaccine. Moderna and Pfizer are the best. And I am going to do everything I can to make sure the residents of the city of Detroit get the best,” Duggan said.

The main issue here is that the Pfizer and Moderna vaccines have a top-line efficacy rate of 95%, whereas Johnson & Johnson’s is closer to 66%. Not only do those numbers not fully account for how effective these vaccines are in preventing severe disease, hospitalizations, and deaths (all three are highly effective at preventing those most extreme cases), they can lead to a basic misunderstanding: the implication that 5% of people who get the Pfizer or Moderna vaccine could still get COVID, versus about 34% of those who get the Johnson & Johnson shot.

That, however, is not how these numbers work, owing to the statistical nature of clinical trials.

An effective clinical trial has two arms of participants: those who get the real thing, and those who get a placebo. The latter is what’s called the control arm.

When measuring vaccine efficacy, you’re pitting the relative risk of someone contracting a virus who received an actual vaccine dose (or two, as the Pfizer and Moderna versions require) versus those who got the placebo. What that top-line 95% efficacy number for those two shots means is that people vaccinated with them had a 95% lower risk of contracting symptomatic COVID—any kind of symptom, mild or otherwise. COVID can present without symptoms and still be transmitted, but the 95% number doesn’t measure those cases.

The more important number is a much smaller one in the case of Pfizer and Moderna: 0.04%. That’s the percentage of people who received a vaccine and then contracted symptomatic COVID. Fortune reached out to Johnson & Johnson for their data on that metric.

Johnson & Johnson’s 66% effectiveness does not mean that 34% of people who receive the vaccine will end up contracting COVID. It means that those who receive it have a 66% lower chance of developing symptomatic COVID—a stellar number given the relationship between the vaccine and placebo arms.

The vaccines were tested across a wide swath of demographics and multiple nations, and ultimately, all three of these shots are highly effective in preventing the worst possible outcomes. Not a single person receiving the Johnson & Johnson vaccine in clinical trials has been hospitalized or died of the disease 28 days after a single shot. And six to seven weeks following vaccine administration, no one taking any of these vaccines in trials has been hospitalized or died.

We’re still learning the nature of asymptomatic COVID and how it spreads. But it’s entirely possible, and perhaps even probable, that vaccines would, at the very least, lessen the chances of transmission.

“Vaccines have always decreased transmission. The clinical trials [for these vaccines] were not designed to test for asymptomatic infection, but there is every biological reason in the world to believe that they will reduce asymptomatic transmission,” Monica Gandhi, a professor of medicine and infectious disease specialist at the University of California San Francisco, told the Association of American Medical Colleges (AAMC). And in that instance, those who are at higher risk would be significantly more protected from a serious case of the disease.

This is precisely the reason that public health experts are urging anyone who qualifies to get whatever vaccine they can. After all, the ultimate goal of an immunization campaign is to achieve herd immunity, which requires at least 70% of a population to be immunized so they can protect others from getting infected. In short: any shot in a storm.