Betting odds more accurate than polls in forecasting Trump’s win

Several prominent polling experts, including Nate Silver, The New York Times and FiveThirtyEight, had the race at essentially a 50-50 toss-up on Election Day, while The Economist gave Harris a 56 percent chance of winning.

Yet Trump remained a solid -162 betting favorite over Harris, the +142 underdog, on Tuesday afternoon at offshore sportsbook BetOnline. The -162 price (which meant bettors had to wager $162 to win $100 on Trump to be elected) implied he had a 61.8 percent chance of winning.

Trump briefly dipped to -145 after the first results came in at 4 p.m. Tuesday. But his odds quickly soared to -225 by 5 p.m. and to -320 by 6 p.m., topping Trump’s high mark during the campaign.

Before Election Day, Trump was as high as a -300 favorite to win the White House again after surviving an assassination attempt in July. Harris flipped to a slight favorite after the Sept. 10 presidential debate. But Trump flipped back to the betting favorite in early October and never relinquished the role.

The president-elect skyrocketed to a prohibitive -2,500 favorite by 8 p.m. Tuesday at BetMGM sportsbook in Ontario, Canada — the only BetMGM location that accepted bets on the election, as betting on politics isn’t permitted in the U.S.

“Undoubtedly, betting markets called this better than anyone,” BetOnline.ag political oddsmaker Paul Krishnamurty told the Review-Journal. “The huge gamble two to three weeks ago on Trump (flipping to the favorite) raised many an eyebrow, and never fit the polls.”

Trump also shot up to -1,200 at offshore sportsbook Pinnacle by 9:30 p.m. Tuesday to win the popular vote. Harris was the -400 favorite at BetOnline on Monday night to win the popular vote over Trump (+320).

As of Wednesday afternoon, the president-elect led Harris by almost 5 million votes and was on track to become the first Republican to win the popular vote since George W. Bush in 2004.

“I was frankly shocked to see Trump win the popular vote, let alone by that margin,” Krishnamurty said. “Even after pollsters adjusted samples to compensate previous underestimates, Trump outperformed by at least five percent.”

Trump’s odds to win the election dropped to as low as -125 on Saturday partly because of a surprising Des Moines Register/Mediacom Iowa poll that showed Harris ahead of Trump in the state by a margin of 47 percent to 44 percent among likely voters. Trump dropped from a heavy -2,000 favorite at BetOnline to win the electoral college vote in Iowa, but was still a solid -500 favorite, with an implied probability of 83.3 percent.

The betting market turned out to be a much better barometer than the well-respected poll, as Trump won Iowa as a -600 favorite, taking 55.9 percent of the votes to 42.7 percent for Harris.

History books

Oddsmakers have correctly called 23 of the past 27 elections.

In the past 104 years, only four betting underdogs on Election Day have won the presidency, according to SportsOddsHistory.com: Trump in 2016 (+375), Jimmy Carter in 1976 (+100), John F. Kennedy in 1960 (+110) and Harry Truman in 1948 (15-1).

The odds before 2012 are taken from newspaper accounts at the time that included conversations with oddsmakers and other betting analysts.

$750M exchanged

New York-based Kalshi exchange and London-based Betfair Exchange also had Trump as the favorite entering the election. A federal appeals court on Oct. 2 cleared Kalshi — a prediction market, not a sportsbook — to resume taking action on election trading/betting.

Kalshi had more than $430 million traded on the election. Betfair had 249 million pounds, which equates to more than $321 million, traded on the election.

BetOnline, which isn’t regulated in the U.S., won on the election.

“It was a good election for BetOnline. We took plenty of five- and six-figure bombs on Trump, but the majority of our bets taken were for Harris,” Krishnamurty said. “Especially during the final days.”

Contact reporter Todd Dewey at tdewey@reviewjournal.com. Follow @tdewey33 on X.