Betting markets have become a big search term when people are trying to determine the outcome of the close race between Vice President Kamala Harris and Donald Trump - but former government officials are warning people not to bet on betting markets’ accuracy.
In recent weeks, betting odds on the candidates from markets like Polymarket, Betfair, Kalshi, PredictIt and more have become a widely cited statistic. Especially, since Trump’s odds have surged.
But betting markets are not based on real polling, they’re easy to manipulate, are not historically reliable and most operate outside of the United States since betting on elections only recently got the ambiguous OK, Jeffrey Sonnenfeld, Anthony Scaramucci and Steven Tian wrote for Time.
All three are former government officials, Sonnefeld has been an informal advisor to President Joe Biden, Trump, Bill Clinton and George H.W. Bush. Scaramucci was the White House Communications Director under Trump for 10 days. Tian, served in the U.S. State Department working on Iranian nuclear nonproliferation.
And all three are worried about the public’s potential reliance on betting markets.
According to their analysis, fewer people actually use the prediction markets than news stories may indicate. They claim it’s nearly impossible to place bets on certain battleground state races because the spread (the difference between the bid price versus the ask price) is so large.
This means the odds can be manipulated with wagers of a few thousand dollars.
“The lack of liquidity and lack of trading volume makes it all too easy to manipulate these prediction markets for anyone who wants to, foreign or domestic,” Sonnerfeld, Scaramucci and Tian wrote.
There are key differences between polling and betting markets to note. The polls show who people intend to cast their ballot for - therefore reflecting the popular vote rather than who will ultimately take the White House via the Electoral College.
Betting markets instead look at who people think will become president. But the gulf between the two methods of predicting the eventual presidential election outcome does raise questions about how effective each one is.
There are other concerns about betting markets to consider too.
Betting on U.S. elections is a very new opportunity for citizens thanks to a recent lawsuit involving Kalshi. There are few regulations and rules about each platform which Sonnerfeld, Scaramucci and Tian raise as a major red flag.
For example, Polymarket, a crypto-based market that U.S. citizens cannot use, puts Trump’s odds at more than 60 percent while Harris’s is below 40 percent.
But Polymarket’s CEO, Shayne Coplan, is friendly with Trump allies and it’s backed by Peter Thiel, an investor who has also backed MAGA political candidates.
In national polls, Harris is leading Trump narrowly 48 percent to 46 percent and have remained largely neck and neck throughout the race.
Sonnerfeld, Scaramucci and Tian point out that while some market betting sites are legal to use, many operate outside of the U.S. meaning those placing bets are not indicative of U.S. voters.
Additionally, the leading markets – Polymarket, Kalshi and Interactive Brokers – have no historical accuracy to secure their reputation. All three had not been established or were inactive before the 2024 election cycle.
Kalshi currently has Trump’s odds at 58 percent while Harris’s is at 42 percent. On Interactive Brokers, Harris’s odds of winning the election are 39 percent.
National polls, while not always the most reliable, currently show Trump and Harris are neck-and-neck with no real indication of who will win.
