Bloomberg published an analysis of every active Polymarket wallet since January 2025.
Over 100,000 accounts lost at least $1,000 each — nearly double the number that made equivalent gains. Around half of all 2 million active accounts showed results between -$10 and +$10. A separate academic study from the University of Toronto covering 2.4 million users and $67 billion in volume found that 68.8% of accounts are in the red, with 76.5% of all profit going to the top 1%.
The concentration is stark. Around 5% of bot-like accounts generated 75% of all trading volume. Just 823 accounts each made over $100,000. Together, they captured $131 million — the rest of the user base lost the equivalent amount.
Here's the detail that stood out most. Retail traders actually predicted the correct outcome more often than the bots. They lost anyway, because they entered positions later at worse prices. The edge isn't in knowing what happens. It's in getting there first.
The Paris hairdryer story
I also find this story particularly funny, because it shows the creativity of humans in earning money.
Polymarket runs daily markets on city temperatures. On April 6, 2026, at around 9:30 pm local time, a weather sensor at Paris Charles de Gaulle airport suddenly recorded 22°C — four degrees higher than the 18°C it had been all afternoon. No other sensor in the area recorded anything similar. A Polymarket user won $14,000 on that spike. Nine days later, on April 15, the exact same thing happened at the same sensor. Another user won $20,000.
Météo-France filed a criminal complaint for tampering with an automated data processing system. French climate enthusiasts on forums speculated that someone had used a battery-powered hairdryer on the sensor. The winning trader's bet was 20 times larger than their typical wager. Shortly after scrutiny began, they deleted their account.
It gets worse. Polymarket did not cancel the contracts or refund the bets. The resolved payouts stood, even though the data they settled on was almost certainly manipulated.
What do you think it would take to regulate this properly? What's the most unusual thing you've ever bet on — or would consider betting on if you could?
Sources:
Bloomberg analysis — bloomberg.com/news/newsletters/2026-04-29/most-traders-lose-on-polymarket-and-winners-look-like-bots Paris weather sensor story — npr.org/2026/04/23/nx-s1-5797876/polymarket-paris-weather-bet
CNN investigation — cnn.com/2026/04/23/europe/france-weather-sensor-polymarket-bet-intl-latam
Over 100,000 accounts lost at least $1,000 each — nearly double the number that made equivalent gains. Around half of all 2 million active accounts showed results between -$10 and +$10. A separate academic study from the University of Toronto covering 2.4 million users and $67 billion in volume found that 68.8% of accounts are in the red, with 76.5% of all profit going to the top 1%.
The concentration is stark. Around 5% of bot-like accounts generated 75% of all trading volume. Just 823 accounts each made over $100,000. Together, they captured $131 million — the rest of the user base lost the equivalent amount.
Here's the detail that stood out most. Retail traders actually predicted the correct outcome more often than the bots. They lost anyway, because they entered positions later at worse prices. The edge isn't in knowing what happens. It's in getting there first.
The Paris hairdryer story
I also find this story particularly funny, because it shows the creativity of humans in earning money.
Polymarket runs daily markets on city temperatures. On April 6, 2026, at around 9:30 pm local time, a weather sensor at Paris Charles de Gaulle airport suddenly recorded 22°C — four degrees higher than the 18°C it had been all afternoon. No other sensor in the area recorded anything similar. A Polymarket user won $14,000 on that spike. Nine days later, on April 15, the exact same thing happened at the same sensor. Another user won $20,000.
Météo-France filed a criminal complaint for tampering with an automated data processing system. French climate enthusiasts on forums speculated that someone had used a battery-powered hairdryer on the sensor. The winning trader's bet was 20 times larger than their typical wager. Shortly after scrutiny began, they deleted their account.
It gets worse. Polymarket did not cancel the contracts or refund the bets. The resolved payouts stood, even though the data they settled on was almost certainly manipulated.
What do you think it would take to regulate this properly? What's the most unusual thing you've ever bet on — or would consider betting on if you could?
Sources:
Bloomberg analysis — bloomberg.com/news/newsletters/2026-04-29/most-traders-lose-on-polymarket-and-winners-look-like-bots Paris weather sensor story — npr.org/2026/04/23/nx-s1-5797876/polymarket-paris-weather-bet
CNN investigation — cnn.com/2026/04/23/europe/france-weather-sensor-polymarket-bet-intl-latam
