It’s been two weeks since a blue wave swept the nation. Dozens of democratic politicians made history in the 2026 midterm election, and so did a newly popular tool to gauge public opinion –– prediction markets.
Nearly $2 billion was wagered in prediction markets, according to the BBC, and a recent New York Times analysis ranked top prediction markets as more accurate than traditional election forecasts for many large-scale races in the days leading up to the election.
Prediction markets allow users to bet on certain events happening, like Tina Kotek winning the Oregon gubernatorial race. If the event occurs, the bettor receives a payout. The price of a bet reflects how likely market participants believe an outcome is.
University of Oregon economics professor Keaton Miller said he’s not surprised, or too impressed, by the popularity of these markets during this election cycle.
“This is one of those things where I think there was a desire for something like a prediction market for a long time,” Miller said. “And we have a very sophisticated financial system that allows for a lot of money to move back and forth very quickly, and a lot of sophisticated traders who are looking for new asset classes and new ways to make money off of information.”
But as prediction markets enter the mainstream, political scientists and economists are questioning their scope, accuracy and the federal government’s minimal regulation.
Wisdom of the crowd?
The founders of the top prediction markets, Kalshi and Polymarket, advertise their businesses as a way to find a common truth through public opinion and participation.
UO political science professor, Chandler James explained how the “wisdom of the crowds” theory helps even out markets to come to a place of common opinion. Thousands of people, who may be engaging with a diverse set of information, are betting on a singular outcome.
“People, when they place these bets, they might be scouring these public polls, or they might be looking at campaign finance data or they might be reading the news and saying, ‘given what I know, my bet is that this outcome is going to occur,’” James said.
As the NYT analysis on prediction market accuracy in the 2026 midterms found that prediction markets made better forecasts as the election neared.
Miller said he thinks this is partly due to more knowledgeable traders getting involved later in the game, and the markets’ ability to adjust their prices quickly as Election Day nears.
“You have more sophisticated traders getting more involved in the market … which is going to drive the price to be sort of more accurate,” Miller said. “And that’s sort of the whole idea of prediction markets. You give people a financial incentive to make the best predictions they can, and then it sort of averages out everyone’s internal predictions.”
Adele Betik, a UO freshman who canvassed for Oregon Governor Tina Kotek, doesn’t see this wisdom of the masses idea as a positive, but instead thinks collectively betting will harm politics.
“(I’m) a little bit nervous of what it means for the future of voting,” Betik said. “I do think it does take away some of the power of voting and how important the individualistic aspect of it can be.”
Betik said she thinks betting on sports is a “very normal thing,” but placing a wager on politics is different.
“It can really demean people’s opinions in the way that people vote,” Betik said. “So I think that is something I worry about. If people start becoming dependent on these things, I think that is really harmful to the voting process.”
Prediction markets in 2026
During this unusual election season, while traditional election forecasters were more moderate, prediction markets’ risky nature may have aided in their accuracy.
“Partially what happened in this election is Democrats overperformed polls, and Democrats did very, very well. And prediction markets, it’s easier for them to sort of incorporate that overperformance, or the idea that might happen, than journalistic enterprise,” Miller said.
For example, neither prediction markets nor election forecasters thought Mississippi Senator-elect Scott Colom was going to win, but prediction markets were still more convinced that the Democrat would win than the forecasters.
“It’s sort of like how weather forecasters have a bit of an incentive to overstate the probability of rain,” Miller said. “If you predict rain and it doesn’t happen, nobody cares. If you predict rain and it does happen, everybody hates you.”
One tool among many
In this new era of analysing elections, James sees prediction markets as a “supplement” to other public opinion research.
“These tools are good, but political scientists already have tools to predict election outcomes that are pretty good. It is called the ‘fundamentals,’” James said. “You look at the economy, there’s some indicators like the unemployment rate, inflation, things like that, that do pretty well to inform people that are trying to make predictions about political outcomes.”
James said polling, the analysis of current affairs, the economy and prediction markets combined offers the most “compelling body of evidence” for forecasting.
One black hole when it comes to prediction markets is the ‘why’ behind each bet.
John Horvick, senior vice president of the public opinion research group DHM Research, sees this as a key difference between public opinion research and political betting.
“Most public opinion … even surveys where you’re trying to ask people what they plan to do for a future event, the vast majority of public opinion research is really to understand … people’s motivations,” Horvick said.
Analysing voter motivation can help inform political forecasts, and tell politicians how voters are thinking about an issue, considering trade-offs, and what persuades them to vote, donate or buy something.
“(With prediction markets) you can get a price on whether or not candidate ‘Smith’ is going to beat candidate ‘Jones’ in an election. But, it doesn’t tell you anything about what’s motivating people to vote a particular way (and) what are the issues that matter to them,” Horvick said.
Horvick also noted the difference in scale between prediction markets and public opinion research polls.
“Most political public opinion research is really at a much smaller scale than prediction markets are ever going to garner interest,” Horvick said.
Still, it doesn’t seem like prediction markets are going to fade into the sunset anytime soon, which Keaton said may mean more regulation is in store.
Betting on senators or betting on corn?
Currently, the regulation of prediction markets is a patchwork across the U.S. In Oregon, political betting is partially legal.
Miller said one reason regulating prediction markets is currently challenging is due to the federal government’s classification of them as commodities markets, originally established for trading farm goods. Commodities markets are regulated by the Commodity Futures Trading Commission.
“One of the whole points of prediction markets, and commodity markets more generally, is that insider trading is not only allowed, it’s kind of encouraged,” Miller said.
Miller said that, because of how commodity markets were set up to encourage farmers to have the upper hand when trading agricultural stocks, this type of market encourages insider trading.
“The whole point of commodity markets is to allow people who have inside information about their own particular needs for the commodities, or uses of the commodities, or production of the commodities, to hedge against that information and hedge against their risks in the broader market,” Miller said.
Another area for potential regulation surrounds market manipulation. For example, James said, large bettors can come in and bet against one candidate to encourage another candidate, even if they don’t think their bet will financially pay off.
“They could be manipulated to change the way people are thinking about a race because perception matters a great deal in politics,” James said. “So if you have polls that show some candidate is really up and you’re like ‘OK, that shows that the candidate is really powerful,’ that might lead some people who might challenge an incumbent to back away.”
Down the road, Miller said he thinks prediction markets will become more regulated, moving away from a model that treats these bets as physical goods.
“I think these markets will be around for a long time; I don’t think they are going anywhere,” Miller said. “I think that especially if there’s a Democratic presidential administration in a couple of years from now we’ll see some change in regulation, but between now and then it’s still going to be the wild, wild west.”
The post Nearly $2 billion was spent betting on the 2026 midterms appeared first on Daily Emerald.