While opinion polls in the 2024 election are just now starting to swing toward Donald Trump, the online political gambling markets have been betting big on the former president for weeks.
A week before Election Day, Polymarket, a leading platform, has Mr. Trump a 2-1 favorite to win the election — the best odds he’s seen since just after he survived an assassin’s bullet in July.
And at Kalshi, a U.S.-based election prediction market, Mr. Trump went from tied with Vice President Kamala Harris earlier this month to better than 60% odds of winning.
So what gives?
“The message I get right now is the markets are moving in Trump’s favor, they’re favorable to Trump, but they’re not a slam dunk,” said Koleman Strumpf, an economics professor at Wake Forest University who has studied the markets for years and says they’re usually onto something.
The markets are quickly coming of age after a court gave the go-ahead to Kalshi earlier this year to become the first largely unhindered U.S.-based operation.
Mr. Strumpf said that while polls ask a set of largely random would-be voters who they would cast their ballot for at that moment, the markets ask a self-selected group of people with at least some financial means who they think will win when all is said and done.
“To me, the markets are always asking the question I’m interested in,” Mr. Strumpf said. “At the end of the day, I don’t want to trade in these markets, but I do want to consume the information.”
Kalshi currently has bets on everything from the winner of the national presidential popular vote, to the winner of Senate races, to control of the House.
But the Electoral College winner is by far the dominant question, with Kalshi reporting nearly $82 million in bets placed as of Sunday.
Ms. Harris’s dropping odds have sparked waves of hand-wringing on social media, with accusations of market manipulation and attempts to skew the election being shared in left-leaning circles.
Kalshi, perhaps hoping for more betting, gleefully eggs them on, telling them to put their money where their keyboard is.
“So if you think the Harris price is inaccurate, stop running your mouth and make a trade,” Kalshi’s Terry Oldreal chided the doubters.
Skeptics say the markets are too small right now to be a good predictive tool — and they’re too easy to manipulate.
A group of researchers led by Yale University’s Jeffrey A. Sonnenfeld said it tried to bet on some key races only to find “zero sellers in those markets.”
“All this means that the so-called price cited by media accounts is merely a phantom figure and not representative of reality,” they wrote.
Even in the overall presidential election, the researchers said the daily betting tops out in the lower hundreds of thousands of dollars, which means a bet of a few thousand dollars can quickly swing the odds several percentage points.
Perhaps the most contentious area of debate is the researchers’ claims that the markets have a poor track record when it comes to predicting outcomes.
Mr. Sonnenfeld and his fellow researchers said the leading prediction market in 2012 had GOP nominee Mitt Romney with a tiny edge over President Obama. In 2016 Democratic candidate Hillary Clinton was the large favorite. And in 2020 one platform gave Mr. Trump the edge with a late surge in betting. In all those cases, they were wrong.
Mr. Strumpf, though, said polls are often even worse, such as in 2016, when Mrs. Clinton had a lead of five percentage points or more in the final Real Clear Politics average of polls for Michigan, Wisconsin and Pennsylvania. She lost all three states en route to losing the White House.
Real Clear said gamblers were just as bad. A week out from the 2016 election, Mrs. Clinton was a better than 4-1 favorite.
In 2020, though, Real Clear said then-candidate Joe Biden was nearly a 2-1 favorite over Mr. Trump a week out from the election — and went on to win handily.
Mr. Strumpf has studied political markets dating back more than a century, when investors used to gather outside the official financial markets in New York to bet on election winners. He said even before the advent of modern polling in the 1930s, the markets were “super accurate” in picking the outcomes.
He also rejected the notion that allowing betting skews elections by changing how some people vote.
“In a century of data I looked at, I can find absolutely no evidence of this,” he said.
Michael McKenna, a pollster who served as a senior legislative aide in the Trump White House, said the markets try to answer one big question — who wins — but lack the nuance of polls, which can say a lot about why someone is leading.
“I have great respect for the wisdom of the crowds. I think that’s what markets are at the moment,” he said. “I think they’re a little faster than pollsters, but not as precise.”
There are three basic types of investors in markets right now. There are deeply partisan Democrats, deeply partisan Republicans and people who are in it for the cash.
Mr. McKenna, who also pens a column for The Washington Times, said he worries about a future when people with serious bank accounts get involved in a big way, opening the door to activities that can corrupt an election.
“It’s not going to stay that way,” he said. “Once the finance guys figure out there’s money to be made in being right, that the presidential election is essentially a commodity, they’re going to figure out a way to make money off of it.”
That’s one of the worries of the Commodity Futures Trading Commission, which battled to try to block Kalshi from opening its election market this year. The CFTC took its corruption to the federal courts but lost in district court and saw its initial appeal rejected by the U.S. Circuit Court of Appeals for the District of Columbia.
The judges said the CFTC could try to regulate election betting, but it would need to go through a full regulatory process.
• Stephen Dinan can be reached at sdinan@washingtontimes.com.