While sports continue to drive outsized volume growth and mainstream narratives around prediction markets, exchange executives and other industry proponents maintain that the future of the category does not rest on sports. Robinhood CEO Vlad Tenev told CNBC’s Jim Cramer that crypto contracts are already claiming a “disproportionate share” of prediction-market activity, adding: “I think within a few years, sports will actually be in the minority, similar to active trading at large.”
That gap between what’s driving the numbers today and what the people inside the category say they’re building it for is a key storyline of prediction markets in late 2026.
Bernstein’s Gautam Chhugani has put a number on how far that shift could go. A research note published last week projects the industry growing from roughly $410bn in volume this year to $10T annually by 2035, driven mostly by financial contracts overtaking sports, not the category simply getting bigger.
JB Mackenzie, who runs futures and prediction markets at Robinhood, has been saying the same thing from his front-row seat. In a recent NEXTPredict podcast interview, Mackenzie predicted:
“I think in three years, people are going to be talking less about the sports component of this, and much more about the financial and economic products that have come to market and brought traders into the marketplace.”
Robin Hanson, the economist who spent decades arguing for prediction markets before they went mainstream, went further when he joined the show. Hanson acknowledges the demand for sports markets for entertainment, but said he’s not sure they clear the bar for why prediction markets deserve protection in the first place.
So why are sports carrying the conversation right now, and why does almost everyone building the category insist that won’t last?
Sports volume drives the headlines, but is it a mirage?
There’s no denying that sports are driving the majority of the current volume growth story, especially with the ramping up of combos, prediction markets’ version of parlay offerings. But part of that headline number is somewhat misleading, based on how volume gets counted and represented.
Notional volume counts $1 for every contract traded, regardless of whether the contract price is one cent or 99 cents. Combos or longshot parlays inflate that topline far more drastically than most single yes/no contracts do. If you place a $10 trade on a 20-1 combo, for example, it counts as 200 five-cent contracts. That means a $10 position registers as $200 in notional volume, even though most of that volume is from a market maker filling the other side of the order, not retail taker volume.
Relying on notional volume, by definition, overstates sports’ true share of retail spend across prediction markets. The rapid growth of parlay products is only intensifying that overstatement.
In August 2026, combos accounted for 44.6% of Kalshi’s $40.03bn in notional volume, pushing the total sports share to around 78% for the month. September has pushed that further. Through Sept. 27, Kalshi had already logged $54.8bn in notional volume per TickerTracker, well past August’s full-month $40.03bn, and combos made up 58.1% of it. Yet combos held just 32.5% of open interest, about the same as single-leg sports contracts (33.2%), despite generating more than twice the volume. The exaggerated combo volume is why sports look like an even bigger share of the pie than the actual capital at risk would suggest.
Third-party analysts are raising a similar flag from a different angle, not about how prediction-market volume gets counted internally, but about how it stacks up against the sports-betting industry it keeps getting compared to. Eilers & Krejcik Gaming recently projected that regulated sportsbooks will still take in roughly $31.7bn in NFL handle this season versus $8.4bn of what EKG calls “handle analog” (HAN) from sports prediction markets, a constructed proxy EKG uses to make prediction-market volume comparable to sportsbook handle.
By that estimate, sportsbooks would account for roughly 79% of combined NFL wagering activity to prediction markets’ 21%, or about 4x the size, despite the headlines and the valuations. While just an estimate based on proxy methodology for converting prediction market notional volume into something more akin to handle, it points in the same direction as the combo math above. If you strip out the way notional volume gets counted, prediction markets look like a much smaller slice of the sports-wagering pie than topline volume numbers suggest.
Rapid sports growth and an expanding pie
None of that means sports volume isn’t genuinely growing. It undoubtedly is, even with combos stripped out, as our NFL week 1 analysis made clear. But the real question is whether it’s growing as fast as everything else around it.
According to major exchanges, the answer to that is “no.” Kalshi CEO Tarek Mansour said in a conversation with a16z:
“Sports is bigger than it’s ever been right now … but it’s also a lower percentage than it’s ever been.
“You have the other categories actually growing faster than sports,” Mansour added. “We’re seeing it in politics and financials and crypto and culture.”
Indeed, at Kalshi, crypto market trading went from under 4% of volume share in January to nearly 19% by August, before the start of the NFL season pulled it back under 15% in September. And Kalshi has now filed for approval to offer leverage on event contracts, one of the last structural pieces institutions need to trade the category at real size.
For Mackenzie, sports tap into a shared human interest, and salience drives the trading demand. “We all have a passion for sports … And because those games are always on television, there’s a natural link, and people ask those questions.”
It’s also part of why Robinhood alone processed 8.8 billion event contracts in Q1 2026 and Kalshi pulled roughly 89% of its $263.5m in 2025 fee revenue from sports contracts alone.
Mansour made a similar point elsewhere. In an interview with journalist Max Raskin, he described sports as the familiar entry point precisely because “they understand it. It’s relatable. They get it.” He also explained how sports become a gateway to other trading categories.
“If you onboard people to come trade on politics or sports or whatever, then they start getting interested in things like understanding AI and its consequences,” Mansour said.
The common theme is that familiarity, especially with sports, gets people in the door, and curiosity takes them to other trading categories.
There’s also a demand-side data point that complicates the narrative that prediction markets are simply taking share from existing sportsbooks. Adhi Rajaprabhakaran, writing about Bernstein’s report, notes that roughly 80% of Kalshi’s users don’t use a sportsbook app at all, and that women make up about 26% of Kalshi’s user base versus 16% at FanDuel.
These figures suggest that sports isn’t cannibalizing sports betting so much as building an entirely different customer base around it. It’s this new audience that Robinhood and Hanson are describing as driving the retail demand for whatever comes next, even if sports markets themselves fall outside of the core purpose for the vertical’s existence in the first place.
Why sports isn’t the most interesting part
For many exchange executives, builders and prediction market purists, sports is far from the end game. In fact, most admit that sports contracts fall somewhere in the range of pure entertainment markets to a distraction from the serious use cases that prediction markets are intended for.
For Kalshi and Robinhood, sports offerings are meeting customer demand and driving overall industry growth.
“What we find is that we’re really catering to the customer who wants to trade the football game on television because they want to do it casually — but within their overall platform, together with everything else,” Mackenzie said.
Hanson has spent his career arguing prediction markets deserve protection because they aggregate dispersed information into an honest price that provides inherent informational value. In his opinion, most sports markets don’t do that work.
“I don’t think most sports markets would pass that threshold of questions that matter,” he said on the NEXTPredict podcast, adding: “I do regret it if there’s a backlash because of this, because I might want to protect the questions that matter from the sports questions that matter less.”
Ethereum co-founder Vitalik Buterin has made a similar case, warning in February that the category was “over-converging to an unhealthy product market fit: embracing short-term cryptocurrency price bets, sports betting, and other similar things that have dopamine value but not any kind of long-term fulfillment or societal information value.”
Are sports becoming a minority category?
Beneath the sports-dominant headlines is a sprawling discussion of real-world use cases, emerging categories and important questions. Mackenzie points to 15-minute crypto volatility contracts, a fast build-out in crude oil, gold, and silver, and a “tremendous increase” in people trading inflation prints and Fed rate-cut decisions, with single-stock contracts and direct central-bank markets next.
“Prediction markets give you the ability to be very precise in the trading you want to do,” Mackenzie explained.
Tenev said the shift is already happening. In his recent CNBC appearance, he pointed to Bitcoin price, Fed decisions, and the CLARITY Act as the kinds of contracts already overtaking sports in Robinhood’s own books. Meanwhile, event-contract revenue hit $156m in Q2 2026, ahead of crypto transaction revenue.
Bernstein’s own modeling also backs the direction. Chhugani expects the industry to compound at roughly 70% a year through 2035, with sports share falling from about 61% to 38%, financial assets rising from 12% to 49%, and institutional volume alone reaching roughly $3T, concentrated almost entirely outside sports. In July, Macquarie projected a similar rotation, with sports’ share of taker volume falling from 88% in 2025 to 47% by 2030 as total taker volume grows from $22b to $1.5T.
Beyond entertainment
Hanson predicts more widespread use of prediction markets as decision markets, priced for decision-makers rather than bettors. He gave the example of a market estimating a candidate’s odds of winning the presidency alongside their odds of being nominated, which gives a political party direct, priced advice on who to run.
Insurance is another use case gaining increasing attention, and for good reason. On the NEXTPredict podcast, host Pierre Lindh put a different long-run framing to Mackenzie: not “prediction markets versus sportsbooks” but rather “prediction markets versus the insurance industry.” Mackenzie buys into the idea.
“We’ve heard a lot of stories about weather, the impact of hurricanes or natural disasters — especially the slow-moving ones, where you might know a hurricane is coming to the US a couple of days ahead of time,” said Mackenzie. “That gives people the ability to be very precise in placing a trade to insure against some property or something else they have.”
As Hanson puts it, financial markets have always served two kinds of customers: people trading “for fun and excitement” and people trading “for risk hedging and transfer.” In his estimation, prediction markets have barely begun serving the second kind. Real risk transfer at scale is one of the clearest examples of a category sports simply can’t be. It needs institutional depth and precise, fast-settling contracts, not a bigger audience of casual bettors.
Robinhood, for its part, says its own safeguards already extend past the sports-market basics, the kind of groundwork institutional players will want in place before they show up at size. Asked what integrity looks like beyond baseline regulatory requirements, a company spokesperson told NEXTPredict.io that Robinhood monitors “for the misuse of material nonpublic information and suspicious trading activity,” and bars “individuals with the ability to influence an outcome” from trading the markets they could influence.
Where demand will take prediction markets
Put together, the shape of the next few years looks less like a rotation out of sports and more like an expansion around it. Sports drives adoption and liquidity. But many of the businesses worth watching are the ones building toward a buyer who needs an answer, not just a bet: decision markets, macro precision products, private-equity exposure and real risk transfer at institutional depth.
Sports are taking prediction markets to new heights in demand, customer acquisition and mainstream awareness. But where the markets go next gets us closer to the “questions that matter.”
For Hanson, there’s an even bigger purpose prediction markets can serve, beyond risk hedging.
“I’d say the world collectively has enormous demand for better information — far more, actually, than the demand for risk hedging or entertainment through markets,” he said. “So I see this eventual demand as much larger than all current financial markets, much less current prediction markets.”
As Mackenzie reminds us, it’s still early in the game:
“It feels like a sprint, because prediction markets have grown so much in two years, but it’s really a marathon, and I think we’re at the beginning stages — maybe not at the starting line like two years ago, but in that first quarter of the marathon.”
Hanson agrees the category is just getting started: “I see current prediction markets as still tapping only a small fraction of what they could eventually reach — but they are bigger than they were before, and that’s good. It’s lowering infrastructure costs, creating legal precedent, getting customers familiar. All of that will help the information markets I’d like to produce later, but we’re not there yet.”
