Prediction markets let you bet on global events
In early April 2026, some strange things were happening in Paris. On April 6, the temperature reading at Charles de Gaulle Airport spiked by four degrees Celsius, and then came right back down again just a few minutes later.
Then, it happened again on April 15—this time, a five-degree Celsius jump in the temperature, lasting just a couple of minutes. No other official readings in Paris saw a spike in temperatures on that day.
Paris’s airport, then, experienced a highly unusual meteorological event. Or at least it did if you believe the temperature readings were accurate. But it turns out there’s reason to doubt the accuracy of those temperature readings—and the reason is called prediction markets.
A prediction market is a platform where users buy and sell contracts tied to the outcome of real-world events. The price of the contract reflects the crowd’s collective estimate of probability. A contract trading at 65 cents implies a 65 percent chance that something happens.
If you’re right, the contract pays out $1. If you’re wrong, it pays out nothing.
There are two global prediction market leaders: Polymarket founded in 2020, and Kalshi, founded a year later.
These platforms let users bet on a lot of things. You can bet on election outcomes, political events, economic indicators, sports games, and celebrity and cultural news. Here’s a sample of actual Kalshi contracts available on May 14, the day I wrote and recorded this episode:
How many cars will Tesla deliver in the second quarter of the year? There’s a 48 percent chance it’s above 440,000 cars, per the market.
Who will win Eurovision 2026? Finland has a 46 percent chance; Greece a 13 percent chance.
How many major Atlantic hurricanes will hit this year? There’s a 44 percent chance that it’s more than three hurricanes.
Will Trump be impeached during his second term? Sixty-two percent chance that he will.
And so on. Kalshi is regulated by the U.S. government; Polymarket is based in Panama and faces almost no government regulation. Polymarket lets its users bet on things like war and military events. (Betting on those is illegal in the United States, but Americans can access Polymarket with a VPN.)
Each contract has something called an oracle. An oracle is the definitive agreed-upon source of truth, so that there is no arguing the outcome of a contract later. For example, a Kalshi contract lets you bet on which company will have a top AI model in 2026. This can be a matter of opinion and judgment. But Kalshi specifies the exact published ranking with a link and a cutoff date, so that there’s no ambiguity after the fact.
That brings us back to the odd temperature swings in Paris. There are contracts on prediction markets tied to the official temperature reading at Paris’s airport. On two days in April, a user on Polymarket placed low-probability bets on the high temperature that day. That means a user bought a lot of contracts for just pennies, and then won $34,000 when the low-probability high temperature was achieved.
But was it? The oracle—the official source of truth—gave high temperature readings. But the meteorological agency says there were signs of tampering. The leading theory is that someone accessed the weather station and held up a hairdryer or lighter long enough to trigger the temperature spike—and the payoff on Polymarket.
If this is indeed what happened, then it was a fraud. And this was not a harmless fraud. Other users who bought opposing contracts lost money, when they would have won under fair conditions. And there’s a real-world impact too: air traffic control uses temperature readings for takeoff and landing decisions. Falsifying temperature readings could have put planes in danger.
This was not the only fraud that could have put people in danger. An American army soldier involved in the operation to capture Nicolás Maduro opened a Polymarket account just one week before the military operation. He placed thirteen bets in the week before the raid, all of which related to the military capture of Maduro. His bets netted him $409,000, or twelve times his investment. Anyone watching these contracts closely would have seen someone placing a large bet on this low-probability event, and it could have ruined the surprise.
That one soldier was arrested, but he’s not the only one profiting from secret military information. In fact, a recent analysis showed that half of all long-shot bets on military action were successful. That’s well above the win rate on other types of long-shot contracts. This suggests a lot of people with classified information are using betting markets to profit from their knowledge of national security secrets.
Jeff’s take
Prediction markets can be a force for societal good. The price of a contract reveals information to the world that was otherwise hidden. And that information can be put to legitimate and good use.
Think of election polling. What is the purpose of opinion polling before elections? It’s to help people understand the chances of one candidate winning, so they can make better decisions ahead of time.
But election polling is famously inaccurate, expensive, and slow to take into account new developments. Prediction markets are far better at forecasting election winners and they incorporate new developments quickly. The same goes for things like economic statistics and the spread of disease: society can benefit when information comes to light sooner. So what, then, if insiders profit off their information, if it makes the world better overall?
In fact, optimists say that insider trading is a feature, not a bug: that you want people with insider information to place bets on the platforms because then their private information can be shared with the world.
I suppose. But I guess what worries me about this is that any person who has access to any type of confidential information will now face a very strong temptation to turn that information into a windfall. And companies and governments know this.
So what might they do in response? They might increase employee surveillance. They might involve fewer people in important decisions. All that has a cost, too.
Great stories make learning English fun
