The prediction market reached a new record for monthly trading volume of $28.4 billion in May, according to Artemis data. This figure surpasses the previous record high of $27.1 billion set in January of this year and marks the fourth consecutive month of increase in trading volume.
The streak itself says more than the headline numbers. The January surge fed on leftovers from the election cycle and the positioning of the new year. May, on the other hand, didn’t really have an equivalent catalyst, but still won. This seems to suggest that activity on prediction market platforms is moving away from sudden event-based spikes towards a stable trough. The idea that these markets are places of occasional curiosity is gradually transitioning into permanent spaces.
Kalsi recorded a trading volume of $17.3 billion last month. This is itself a monthly record for the platform, with a 29% increase in volume month-over-month. Even more noteworthy is the fact that approximately 61% of the total trading volume came from Kalsi, which processed almost double the trading volume of Polymarket, which recorded $8.4 billion. But together, a duopoly still exists, as both of these platforms accounted for nearly 90% of May’s trading volume.
If we went back a year, this split between the two largest platforms would have been reversed. Polymarket built this category and maintained market dominance from 2024 all the way to September 2025 when Kalsi tipped the scales. We tracked a similar reversal within crypto-themed contracts earlier this month, with Kalsi shares rising since February.
The reason for this reversal is actually quite understandable. Kalsi’s regulated status in the United States helps drive its sports business, which accounts for the majority of its sales. Conversely, Polymarket ran into a number of regulatory hurdles throughout most of the last quarter. This does not mean that Polymarket’s trading volume is falling off a cliff from month to month; it simply has stopped rising.
Why break one record four months in a row?
A single record could be a fluke. Sports calendars, viral markets, and one big election can all spike in a month and then wane. After 4 months in a row, it’s hard to give up. This means that the floor beneath the sector continues to rise even when the catalyst stops rotating.
That baseline is the maturity signal. Prediction markets are no longer waiting for elections to become important. Newsrooms quote odds, hedge funds monitor odds, and market makers like Wintermute are now quoting two-sided prices across the largest venues. This plumbing is starting to look more like a genuine derivatives market than a new gamble.
Regulation and construction companies take the next step
Mr. Kalsi’s lead is in line with regulatory winds. The CFTC has indicated that clearer predictive market guidance favoring licensed exchanges is coming. Polymarket is playing the same game from the other side, buying a CFTC-licensed venue to re-enter the US market from which it had been largely shut out.
The builder tier is a wildcard. Hyperliquid launched the HIP-4 Outcomes Agreement on Mainnet on May 2, allowing developers to develop their own markets on top of an exchange that already settles hundreds of billions of dollars in monthly trading volume. The initial share is modest at about $87.7 million, well below 1% of the sector. But it does signal the next step. It is a marketplace that anyone can launch and settle on-chain, building on a base of 1.4 million existing traders.
For now, keep the shape simple. There are two large venues, one of which is regulated and exiting, and there is nothing to force it up, sitting in a sector that just recorded its strongest month.
