From elections and inflation to sports, tech, and more, Kalshi lets you trade on the real-world events you already follow. Buy “Yes” or “No” contracts based on what you think will happen, then earn returns if you’re right.

Pick a market, make your prediction, and put your knowledge to work.

Trade $25, Get Up to $500

_{Bonus credit varies from $15 to $500. Terms apply.}

ONE WILD WORLD — NEWSLETTER

ISSUE #045 · THIS CHANGED EVERYTHING · AUGUST 2026

Here is an uncomfortable fact about the last five years. The single largest change to everyday human behaviour in living memory — billions of people covering their faces on command — was ordered before anyone had properly tested whether it worked.

The randomised trials that could actually answer the question arrived afterwards. And when they did, the results were far murkier than either side has ever been willing to admit. This is not the story that masks are useless. It is not the story that masks are magic. It is the stranger, quieter truth that sits between the two, and that almost nobody told you: we chose first, argued second, and tested third — and the tests refused to hand anyone a clean victory.

Cast your mind back to the spring of 2020. The official advice was not “wear a mask.” For a while, it was the opposite.

At the start of the pandemic, the World Health Organization, the US Surgeon General and Canada’s chief public health officer all actively discouraged healthy members of the public from wearing masks. Part of that was a genuine gap in evidence. Part of it, less nobly, was supply: surgical masks were in desperately short order, and hospitals needed them more than you needed one for the supermarket.

Then, on the 5th of June 2020, the WHO reversed. It now advised governments to encourage the general public to wear fabric masks wherever the virus was spreading and distancing was difficult. Officially this was dressed up as an “update” — the Director-General was careful to say masks alone wouldn’t protect you and that nothing fundamental had changed. Everyone outside the building could see it for what it was: a U-turn. The public had been told “don’t,” then told “do,” with no dramatic new discovery in between to justify the switch.

That is the part worth sitting with. The guidance flipped before the evidence did.

To understand why the evidence stayed so slippery, you need one piece of plumbing: the randomised controlled trial, or RCT.

The idea is simple and rather beautiful. You take a large group of people and split them into two purely by chance — a coin toss, in effect. One group gets the thing you’re testing; the other doesn’t. Because the split is random, every hidden difference between people — age, wealth, caution, where they live — evens out across both groups. So if one group ends up healthier, you can credit the intervention rather than the sort of person who chose it.

This is why an RCT beats simply comparing “people who wore masks” with “people who didn’t.” Maskers and non-maskers differ in a hundred ways that have nothing to do with the mask, and any of those hundred ways could be doing the work. The RCT is the gold standard precisely because it strips them out. For most of 2020, on Covid masking, there wasn’t one. And then, at last, there were two.

The first was ours — Danish, and worth knowing about.

DANMASK-19 recruited just over 6,000 adults in April and May of 2020 and split them at random: half were told to wear a surgical mask outside the home, half were not. The question was narrow and specific — does a mask protect the person wearing it?

The answer was a shrug. Infection turned up in 1.8 per cent of the mask group and 2.1 per cent of the controls. Statistically, that difference was not significant.

Here the mask sceptics tend to shout “See — masks don’t work,” and here they overreach, because look at what the numbers actually permit. The result was compatible with anything from a 46 per cent reduction in infection to a 23 per cent increase. The trial simply wasn’t big enough to catch a modest benefit if one existed — it was built to detect a halving of risk, nothing subtler. Fewer than half the people told to mask actually did. And crucially, it only tested whether a mask protects you, not whether it protects the people around you — which is the main thing masks were ever supposed to do.

So DANMASK didn’t prove masks fail. It proved something more deflating: the tidy, wearer-protecting effect people wanted to believe in was never robustly there to be found.

The second trial was enormous, and it is the one that actually tells you something.

Across 600 villages in rural Bangladesh, researchers ran a mask experiment on more than 340,000 adults — the largest of its kind ever conducted. They didn’t force anyone to mask; they promoted it, handed out free masks, and got community leaders to model the behaviour. Even then, mask-wearing only climbed from 13 per cent to 42 per cent. Hold that in mind: in the “masked” group, most people still weren’t masked.

The headline result was modest — around a 9 per cent drop in symptomatic infections overall, sitting right on the edge of statistical significance. But dig in and one signal stands out and holds firm: in the villages given surgical masks, infections among people over 60 fell by 35 per cent. That is a real, robust, meaningful number, and it is the single best piece of evidence we have that a mask can do something.

Now the part almost nobody quotes. What about the cloth masks — the ones most ordinary people actually wore? Here the study goes quiet. The authors were careful and honest: they explicitly declined to claim surgical masks beat cloth ones. What they said was that the evidence for surgical masks was more robust — the cloth-mask effect appeared or vanished depending on how you handled the missing data. It was never nailed down.

Read that back slowly, because it is the whole story in a sentence: the mask that billions of people spent two years wearing is the mask the biggest trial in history could never confirm did anything at all.

Which brings us to the document that detonated the argument.

In January 2023, Cochrane — the outfit widely regarded as the gold standard for weighing medical evidence — published a giant review of the physical measures used against respiratory viruses. Its verdict on the trials was blunt: pooled together, they did not show a clear reduction in infection from wearing surgical masks, and, among healthcare workers, no clear difference between surgical masks and N95 respirators.

Cue the victory laps. Newspaper columnists and politicians seized on it as proof that “masks don’t work” and that every mandate had been theatre.

And then came the twist that makes this a One Wild World story rather than a talking point. Cochrane’s own editor-in-chief stepped in to say the review had been badly misread. It had not shown masks don’t work, she explained; it had shown that studies of schemes to get people to wear masks were inconclusive — largely because, as in Bangladesh, barely anyone in the “mask” groups reliably masked. Absence of evidence, she reminded everyone, is not evidence of absence.

That should have settled it. It didn’t. Because in June 2024, after a bruising public fight, Cochrane quietly decided not to change the review’s wording after all — a soft reversal of the reversal. The review’s own authors accused the editor of throwing them under a bus to appease a newspaper columnist. By this point the story had folded in on itself so many times that the only honest summary was this: everyone was misrepresenting the same study, in opposite directions, at the top of their voices.

So strip away the shouting. Here is what the best available evidence supports, stated as plainly as it can be.

A surgical mask, worn by an elderly person, in a place where the virus is spreading, probably offers a real if modest benefit. That much is solid. Almost everything past that point — the cloth masks, the community-wide mandates, the promise that your mask was protecting you — was never robustly established, in either direction. Not proven useless. Never proven to work. The entire argument should have lived in the gap between those two phrases, and almost nobody was willing to stand in it.

You might assume this is settled history by now. It isn’t. In 2026, researchers are still publishing papers accusing each other of misrepresenting these exact trials — the same handful of studies, refought line by line, six years on.

And that is the real lesson, the one that has nothing to do with your feelings about masks. We took the largest behavioural decision of the century, applied it to billions of people, and only got round to testing it properly afterwards. Then, when the tests came back ambiguous, we treated the ambiguity as an insult rather than an answer. Choose first. Argue second. Test third. Refuse to accept the result whenever it fails to flatter you.

You were told to mask. Then you were told you’d been conned.

Both sides were quoting the same studies — studies that never said what either of them claimed.

Bundgaard H et al., “Effectiveness of Adding a Mask Recommendation to Other Public Health Measures to Prevent SARS-CoV-2 Infection in Danish Mask Wearers” (Annals of Internal Medicine, 2021) — the DANMASK-19 trial. Infection 1.8% vs 2.1%; odds ratio 0.82, 95% CI 0.54–1.23; the trial was powered only to detect a 50% reduction.

Abaluck J et al., “Impact of community masking on COVID-19: A cluster-randomized trial in Bangladesh” (Science, 2022) — 600 villages, 342,183 adults. Surgical-mask villages: ~11% overall reduction and a 35% reduction among over-60s; the cloth-mask effect was not robust to how missing data were handled. Mask-wearing rose from 13.3% to 42.3%.

Jefferson T et al., “Physical interventions to interrupt or reduce the spread of respiratory viruses” (Cochrane Database of Systematic Reviews, 2023); the Cochrane Editor-in-Chief’s statement of 10 March 2023 that the review had been misinterpreted; and the June 2024 update in which Cochrane declined to amend the review’s wording.

World Health Organization, “Advice on the use of masks in the context of COVID-19,” interim guidance of 5 June 2020, and the Director-General’s remarks of the same date; contemporaneous reporting on the earlier advice against public masking by the WHO, the US Surgeon General and Canada’s Chief Public Health Officer.

Ongoing 2026 academic exchange contesting the interpretation of the Cochrane and Bangladesh trials — evidence that the dispute remains live six years on.

Most IRA Millionaires missed the first Roth conversion window. Craig Wear, CFP® — who's helped 3,000+ families avoid a projected $10B+ in lifetime taxes — wrote Roth Conversion Reset to make sure you don't miss the second. The full book is free, delivered to your inbox in minutes.

Send Me the Book