How to bet on the presidential election—and why prediction markets might be better than polls

Rey Mashayekhi
·7 min read

Our mission to help you navigate the new normal is fueled by subscribers. To enjoy unlimited access to our journalism, subscribe today.

With less than two weeks until Election Day, Joe Biden’s heavy lead in the polls is not enough to convince many that President Trump is doomed to defeat. After all, the polls notoriously got it wrong four years ago when Trump shocked the world by beating Hillary Clinton, despite trailing Clinton in most polls for virtually the entire campaign.

Because of 2016, there has been heightened public interest in an alternative means of predicting which way the 2020 election may go. And many are choosing to follow the money: looking to political betting markets, as well as free-to-play prediction markets, for a better read on who will be the next President of the United States.

While betting on political elections remains illegal in all 50 states—and is subject to strict oversight by the federal Commodities Futures Trading Commission (CFTC)—there is more interest than ever in betting on U.S. elections. Gambling companies based in the U.K., where political betting is legal, expect wagering on the 2020 election to break records set four years ago, while bookmakers in Las Vegas have speculated that the presidential election could prove a bigger market than the Super Bowl.

The current occupant of the White House—polarizing lightning rod of attention and opinion that he is—has undoubtedly played a major role in this heightened popularity. “I think Trump has definitely poured some gasoline on the political betting landscape,” says Brian Pempus, a betting industry expert at NJOnlineGambling.com, who cited how European bookmakers reported an exponential increase in election betting activity from 2012 to 2016.

“With politics, everyone already has an opinion and some skin in the game, as they call it,” Pempus notes. “It feels like a natural fit for gambling, and I know a lot of the legal sports books [in the U.S.] would be willing to offer it.”

The state of West Virginia even went so far as to legalize betting on elections for an extremely brief period in April—a move that saw sports books like FanDuel immediately seek to jump into the market. But the state quickly reversed its decision, citing existing election laws that ban the practice and scuppering any designs bookmakers may have had on accepting election bets.

“We certainly like the idea of being able to bring more people into our product—just as we’d like to offer Oscars betting, because there are lots of people who love to follow it,” says Cory Fox, FanDuel’s vice president of government affairs. The sports book was able to move quickly on offering presidential betting lines in West Virginia since its parent company, Irish gaming operator Flutter Entertainment (formerly Paddy Power Betfair), “has offered political betting for a number of years” in foreign markets, according to Fox.

There is at least one operator legally permitted to field wagers on political outcomes in the United States. PredictIt, a New Zealand–based online prediction market, was granted a rare no-action letter by the CFTC in 2014, thanks to an academic mandate that sees the website partnered with hundreds of universities and academic partners around the world. In August, the website—a self-described “stock market for politics”—reported record trading activity over speculation about who Joe Biden would select as his vice presidential nominee.

“This is the busiest time we’ve ever had in terms of eyes on the site and people wanting to participate,” says Will Jennings, PredictIt’s head of public engagement. The website now has between 110,000 to 120,000 “funded accounts,” including up to 1,000 highly active users “who treat this as a part-time to full-time job” in terms of their trading volume, Jennings says.

PredictIt remains highly regulated and constrained by the terms of its agreement with the CFTC: Users can’t trade more than $850 on an individual bet, and there can be no more than 5,000 traders with positions in a given “market” (i.e. the winner of the presidential election, the party that will control the U.S. Senate, etc.). But Jennings considers the growth of PredictIt’s platform, which allows traders to buy and sell “shares” in a given outcome, as part of a trend toward the “decentralization of punditry” when it comes to forecasting future events.

“Our media environment is too chaotic to keep up with. The world is getting more complex, and people are going to keep looking for better, more responsive ways to track risk,” he notes. With mainstream media sources surrounded by more public skepticism than ever, prediction markets are part of an emerging tendency to “trust the crowd” rather than relying on polling and other traditional means of gauging public opinion.

Jason Trost, founder and CEO of London-based betting exchange Smarkets, compared the effectiveness of prediction markets to Wikipedia and noted how crowdsourcing can prove an efficient, accurate means of gaining a perspective on events. “It’s fascinating how a diverse group of unrelated people can together very quickly synthesize information, and prediction markets are the same thing,” he says. “Different traders with different perspectives are coming together in real time.” (The most valuable “lifeline” on the popular quiz show Who Wants to Be a Millionaire?, according to Trost, is the Ask the Audience poll.)

In the U.K., Smarkets has taken more than $11 million in wagers on the U.S. presidential election to date, and the company is able to parlay its betting data to offer probabilities on all sorts of U.S. political results. To that end, the company’s ambitions extend well beyond recreational gambling. “We are trying to turn event trading, including politics and sports, into an asset class,” Trost says, describing his vision for Smarkets as “a real-time, short-term insurance market” where investors can “hedge out [their] exposure” to various outcomes.

Smarkets has plenty of company vying to shape the future of political betting markets. New York–based BallStreet Trading, a free-to-play online prediction market primarily focused on live sports, has also offered odds on a wide variety of political outcomes—from events during the presidential debates to whether another round of economic stimulus measures will arrive before the election. BallStreet plans to offer up to a dozen unique political markets on Election Day, according to founder and CEO Scott San Emeterio, who also believes there’s an appetite for predictive election data beyond “conventional polling.”

“I think prediction markets have a way of cutting through the noise,” San Emeterio says, adding that it’s important to factor a given platform’s audience in evaluating the data being fed into it. (BallStreet’s traders, for instance, survey as leaning “more right,” according to San Emeterio, meaning “there’s obviously a bias” at play.) In another four years and beyond, he believes that “election wagering will be a much bigger part of the conversation”—particularly if the polls once again prove way wide of the market in predicting the winner of the presidential election.

“When you look at polling four years ago, it couldn’t have been more wrong,” he says. “If Trump wins again, then we’re asking the same questions.”

While polls may erroneously indicate that one candidate or issue tends to have more support than others, prediction markets work completely differently in reflecting the likelihood of a result occurring—providing an arguably more nuanced, open-ended perspective in an increasingly unpredictable political environment.

“Right now, according to our market, Biden has a 70% chance of winning,” Smarkets’ Trost noted last week. “But if Trump wins, it’s not that our market was wrong—it’s that Trump had a 30% chance to win, and it happened. It’s like a 30% chance of rain.”

More politics coverage from Fortune:

This story was originally featured on Fortune.com