Why prediction markets swung toward Harris before Election Day
Add Axios as your preferred source to
see more of our stories on Google.
Former President Trump's chances of recapturing the White House fell sharply over the weekend across prediction markets like Polymarket and Kalshi. Then they climbed back up.
Why it matters: Volatility like this underlines why using betting markets to gauge the status of the presidential race is tricky. And yet, prediction and betting markets have emerged this year as a powerful force in shaping public opinion — and campaign messaging — about the horse race.
- Follow Axios' live coverage featuring reporting and analysis from across the country.
The big picture: Prediction markets are misnamed.
- They are betting lines intended to provide a price to wager on at a given moment in time — not necessarily to forecast whether the outcome is probable.
By the numbers: Trump's chances of recapturing the White House fell from 67% on Polymarket on Oct. 30 to 53% on Sunday and then back to 62% on Tuesday morning.
- Vice President Harris went from 33% to 47% to 38%.
The intrigue: The public may judge the success of prediction markets based on the outcome of the election, with a Trump victory meaning the markets were right all along, while a Harris win means they messed up.
- That's not necessarily fair. It's like "claiming the stock market fails to value companies accurately simply because share prices change over time," says Lloyd Danzig, managing partner at Sharp Alpha, an investment firm focused on sports, gaming and entertainment.
- "In both cases, it is investor sentiment surrounding then-available data that drives pricing in a given moment," he added in an email.
Zoom in: On prediction markets, bettors — or shall we call them investors? — can sell their contracts at a loss or a gain, depending on how the markets have moved since they first bought in.
- Someone who bought Trump at 52%, or 52 cents on the dollar, could've sold at 64%, or 64 cents, for a sizable gain without waiting for the election to actually take place.
- Anyone who holds on to their contracts through the ballot count will either see them go to $0 or go to $1.
Follow the money: It's possible that Trump's odds fell over the weekend because more investors were cashing out their bets.
Flashback: Hillary Clinton's odds of winning in 2016 on PredictIt were 82% the day before Election Day.
- Bookies put 90% odds on U.K. voters rejecting Brexit. That's not what happened.
Zoom out: Like prediction markets, political polls are also a measure of current sentiment and not necessarily predictive.
- However, unlike polls, single individuals can sway prediction markets with big bets, as Axios' Brady Dale explained last month.
The bottom line: Voting day is upon us and we'll soon get the real numbers on who's ahead.
