Poker Writer
The wisdom of the markets was borne out last night as Zohran Mamdani managed an upset in the New York City Democratic mayoral primary against Andrew Cuomo in a hotly contested and contentious election.
While the established candidate, Cuomo, was the assumed leader for most of the race, polls and several prediction markets swung toward Mamdani. Although the results aren’t yet official, preliminary numbers show Mamdani as the the very likely winner.
BREAKING: Mamdani surges as clear favorite in the NYC Mayoral primary.
61% chance he wins. pic.twitter.com/8KJeTHMyan
— Polymarket (@Polymarket) June 23, 2025
Cuomo was trading as high as close to 90% at the end of May and was the heavy favorite in Polymarket, as well as other prediction markets, all the way up until the day of voting, when the markets started to shift in favor of Mamdani. Polymarket, Kalshi, and PredictIt all had Mamdani as the favorite on the day of the primary.
Several factors contributed to a rapid shift in the markets toward Mamdani, including:
- A Cuomo campaign that was widely seen as poorly managed
- Endorsements by prominent Democrats, such as Bernie Sanders and Alexandria Ocasio-Cortez
- Increased name recognition for Mamdani nationwide
There was also the factor of the ranked-choice voting format. That was expected to benefit Mamdani in a competitive field for the progressive vote, while Cuomo’s support or lack thereof may have been more polarized. Still, Mamdani enjoyed a greater share of the first-choice vote than even the markets projected, and might have been the winner regardless.
Prediction Market Accuracy
The Mamdani win was a surprise to many who were following with interest, as he seemed to have an outside chance for most of the campaign. But the shift in the prediction markets changed what many people were thinking of as an outside shot to the real possibility of Mamdani’s election. Looking back now, it appears that the late money that came in on Mamdani was very sharp.
The swing in the predictions markets alarmed some social media users, who warned others not to trust them over polls.
However, around the same time that the prediction markets put Mamdani ahead, the polls were also moving rapidly in his favor. They showed him making great strides with white men, in particular.
NYC POLL: For the first time, Mamdani leads Cuomo in RCV Final Round
🔵 Zohran Mamdani: 52 (+6)
🔵 Andrew Cuomo: 48% (-6)
——
White: Mamdani 61-39
Black: Cuomo 62-38%
Hispanic: Cuomo 60-40%
Asian: Mamdani 79-21%
—
Men: Mamdani 56-44%
Women: Cuomo 52-48%
—
College: Mamdani… pic.twitter.com/dCzcIbM48l— InteractivePolls (@IAPolls2022) June 23, 2025
But even factoring in those late swings, Mamdani’s performance beat expectations by as much as 18 percentage points.
In this case, the prediction markets seemingly knew better than the polls did. As the closing of polls drew closer, the line continued to move further and further in the direction of Mamdani, suggesting that they were tapped into public sentiment in a real-time way that polls can’t achieve.
This result mirrors the 2024 Presidential election, where prediction markets also consistently gave Donald Trump a better chance of winning the presidency than most polls did.
Director of Crypto-economics Lab at UC Digital Futures, Michael Jones, explained to University of Cincinnati News why that could be:
The polls are just people’s opinions; the pundits had their opinions, but there really are no consequences if they got it wrong. Maybe they took a little heat in the media. But if you got it wrong in the prediction market side, then you lost significant amounts of money.
Image Credit: Karamccurdy, CC BY-SA 4.0, via Wikimedia Commons
