Maybe the media’s bored with the pandemic storyline at this point (millions of doses administered every day, cases basically stable nationally) but I’ve seen three prominent examples in the last three days of outlets underselling the vaccines by overplaying the risk of infection after you’ve gotten your shots. Which is weird, since the educated professional class to which journalists belong tend to be evangelists for vaccination. They see their role, I think, partly as an adjunct to the public-health bureaucracy in trying to convince wary Americans to get vaccinated.

Are we past that point now, though? The sheer tedium of late-stage pandemic news may be leading them to hype the chances of getting sick post-vax, just to squeeze a few last drops of sensationalism from the dangers of COVID. They don’t need much encouragement on that point either: Although liberals are more enthusiastic about getting vaccinated than conservatives, liberals are also more likely to overstate the risks from the disease and therefore may be more prone to blowing reports of illnesses in vaccinated people out of proportion. Combine that with the fact that many reporters just aren’t that bright, particularly when they’re called on to engage with a complex scientific matter, and it’s a recipe for scaremongering as the national vaccination effort continues to scale up.

Example one: The New York Post seems to be on a mission to convince its readers that the Johnson & Johnson vaccine is garbage. Two headlines from the past two days:

The first story is about a woman who developed a dry cough and fatigue three weeks after her vaccination. But the clinical data on J&J is based on effectiveness after 28 days, not 21, and there’s evidence that people gain a higher level of protection as more time passes after they get their shot. Even after 28 days, J&J is 85 percent effective at preventing severe disease, not 100 percent. It’s not “news” in any way that a person who got that vaccine might get COVID within a month of receiving their dose. It’s rare, but even the trial data suggests that it’ll happen.

The second story is about a man who was vaccinated and then started having symptoms 25 days later. He ended up in the hospital and is in critical condition with pneumonia. That is noteworthy since J&J was 100 percent effective at preventing hospitalizations and deaths in trials — again, after 28 days — but if you’re going to scare the hell out of New Yorkers by making them think that a dose of Johnson & Johnson won’t keep them from dying after all, I’d say you’re duty-bound to first investigate how much of an outlier this man’s case is or what underlying health conditions he might have. Call up the city ERs and ask them: How many patients have you seen who’ve been vaccinated? Are there a disproportionate number who got J&J? “Vaccine isn’t as effective as it’s cracked up to be” is a fully legitimate news story but one case can’t prove anything. So why showcase it and make it seem like it does?

Example two: Lotta heavy-breathing this weekend over a study that demonstrated the South African variant of the virus is capable of “breaking through” Pfizer’s vaccine.

This headline from Drudge is supremely irresponsible, and frankly, false pic.twitter.com/ejWaJ22ETY

— Ben Shapiro (@benshapiro) April 12, 2021

Drudge linked to this story, which confirmed something that we’ve known for awhile, namely, that the South African strain is the one that the vaccines seem to have the most trouble defeating. By no means, though, did the study show that Pfizer’s vaccine “can’t stop” the South African variant. And if you read the fine print, per this Twitter thread by one of the study’s authors, you’ll find that there’s a big silver lining to it:

1. We wanted to test what type of variant infects the very few vaccinated individuals who go on to become infected with SARS-CoV-2. To this end, we generated a case-control cohort: every vaccinee was matched with a non-vaccinated individual infected with the virus

— SternLab (@SternLab) April 10, 2021

3. All in all – most infections in both categories were from the “British” variant B.1.1.7. However, we noted 8 cases of infection with the “S.A” variant B.1.351 in fully immunized individuals, as compared to only 1 infection in the control (non-immunized).

— SternLab (@SternLab) April 10, 2021

5. To summarize: we see evidence for reduced vaccine effectiveness against the British variant, but after two doses – extremely high effectiveness kicks in. We see evidence for reduced vaccine effectiveness against the S.A. variant, but it does not spread in Israel.

— SternLab (@SternLab) April 10, 2021

“It does not spread in Israel.” That could have and probably should have been the headline, that the fearsome South African strain just doesn’t seem to be very contagious. We’ve seen the same thing here in the U.S., as there are documented cases of the South African variant in America but in far smaller numbers than the British variant. If you have the misfortune to encounter it, it *might* make you sick even if you’re vaccinated. (Although it seems strange to me that the researchers found more cases of the South African strain in vaccinated than unvaccinated people, which makes me wonder about the sample.) But you’re unlikely to encounter it in the first place, and that’s the point. The strain you do stand a good chance of running into, the British one, turns out to be ineffective against Pfizer. Which means reporters are once again accentuating the negative in vaccine news.

But here’s another detail about the study that most reporting has missed:

"One other key point that was left out of the analysis is that everyone who got the 1.351 variant who had been vaccinated, got the variant before they were fully vaccinated," @ScottGottliebMD on the new study from Israel saying the South African variant may evade Pfizer vaccine. pic.twitter.com/SUEYc9aC3Y

— Squawk Box (@SquawkCNBC) April 12, 2021

Is the South African variant actually breaking through against Pfizer recipients who’ve been fully vaccinated and are past the subsequent 14-day period? The jury still seems to be out on that. Kind of an important detail.

Example three: This is probably more a case of someone at CNN being an idiot than having too dim a view of whether the vaccines work or not, but it’s a doozy.

CNN article on how to safely fly claims that 90% vaccine efficacy means that🙄 for every million who fly, we could have 100,000 infections. NO NO NO. That’s not what that number means. Also, this didn’t even happen when millions flew unvaccinated. So how could it make sense now? pic.twitter.com/4MLW5oFw4f

— zeynep tufekci (@zeynep) April 11, 2021

Tufekci’s right. For all the fear about flying before the vaccines arrived, the evidence at the time suggested that infections on airplanes are a rare occurrence. That’s because all the air in the cabin is replaced every few minutes and all passengers are wearing masks. Even in close quarters, the virus just isn’t lingering for long before it’s flushed out. Phil Klein explains the actual math here:

Translated into actual reality, a 90 percent effective vaccine refers to the reduction in odds of a vaccinated person being infected relative to somebody who has not taken the vaccine. That is, in a given group of, say, 2,000 people, if 100 unvaccinated people would ordinarily get COVID-19, just 10 vaccinated people in the same sized group would get the virus. It doesn’t mean that 200 vaccinated people would get the virus!

For the CNN statistic to be true, it would mean that without a vaccine, everybody who flies on an airplane is guaranteed to get infected — that is, all million passengers in the scenario above would get COVID-19. This is nowhere near the truth based on months of data from people flying before vaccines were even released.

A vaccine’s effectiveness is determined by measuring how many people in a control group got infected versus how many in the vaccinated group did. Obviously, not everyone in the control group will get infected. Most won’t. They’ll dodge the virus through sheer luck or by taking precautions when exposed or just by having a naturally robust immune response. CNN’s dopey mistake would lead the average reader to believe that they’re still walking around post-vax with a 10 percent chance of getting infected, a probability that may be high enough for some skeptics to convince them that the benefit isn’t that great. In reality, you’re walking around with a 90 percent chance of fending off the virus in a situation where an unvaccinated person would get infected.

You’d think our media class would be doing better with all of this at a fateful moment, when we’re burning through the willing vaccine recipients and will soon encounter the hesitators who need more persuasion. But then, the media takes its cues from experts and the experts seem weirdly intent on underselling the vaccine as well. Do better, everyone.