Bernstein Predicts Annual Trading Volume of $10 Trillion for Prediction Markets
Bernstein has released a research report predicting that the annual trading volume of global prediction markets will reach $10 trillion by 2035, with financial contracts linked to crypto assets, stocks, and commodities replacing sports as the largest trading category.
A team led by Bernstein's chief analyst Gautam Chhugani estimates that the global prediction market trading volume will be around $50 billion in 2025. In the first eight months of 2026, the trading volume has already reached approximately $300 billion, with an annual forecast of $410 billion, a 71% increase from the agency's prediction in April. In April, Bernstein had only predicted that prediction markets would reach $1 trillion by 2030, but this time extended the forecast to 2035 and increased the figure tenfold. According to this prediction, the compound annual growth rate from 2025 to 2035 will be about 70%; even starting from the 2026 baseline, it will still need to grow by about 43% each year thereafter.
In terms of category structure, sports contracts will account for 61% of the prediction market trading volume in 2025, while financial contracts such as crypto, stocks, and commodities will only account for 12%. Bernstein predicts that by 2035, the share of sports will drop to 38%, while financial contracts will rise to 49%, corresponding to an annual trading volume of about $4.7 trillion. The core assumption of this model is that crypto, stocks, and commodities constitute a potential trading pool of about $700 trillion, which will expand to $900 trillion by 2035. The prediction market only needs to capture 0.5% of that trading volume to reach the aforementioned scale.
Kalshi is the biggest beneficiary of this growth. By August 2026, Kalshi accounted for about 60% of the industry's trading volume, up from about 35% a year earlier. The share of crypto contracts on its platform rose from less than 5% in January to about 20% in August, primarily driven by short-term Bitcoin contracts. The trading volume of commodity contracts increased from less than $2 million for the entire year of 2025 to $590 million since 2026, with $410 million contributed in August alone.
Polymarket's structure is shifting towards sports. This year, sports accounted for about 52% of its global trading volume, down from 39% in 2025. Political contracts dropped from 32% to 22%, while crypto contracts accounted for about 21%. Brokerage channels are also expanding: Robinhood's event contract business reached $156 million in revenue in the second quarter of 2026, a tenfold year-on-year increase.
On the product and regulatory front, platforms are expanding categories to include binary options on public company KPIs, perpetual contracts, and contracts on individual stocks. Cboe is advancing the "Cboe Clear US" central counterparty clearing solution, preparing to enter the event contract space. Bernstein believes that the regulatory discrepancies between federal and state levels for U.S. sports prediction markets will be difficult to fully clarify before 2027 to 2028. This is also one of the reasons why the model predicts that the share of financial contracts will surpass that of sports.
In terms of market mechanisms, the buyers are mainly retail investors: crypto and stock traders trading short-term binary contracts through platforms like Kalshi, Polymarket, and Robinhood, as well as users who originally placed bets on sports betting platforms. The sellers and liquidity providers are primarily professional market makers like Susquehanna, who profit from bid-ask spreads and hedging. Funds are flowing from traditional sports betting, retail options, and crypto leveraged products to event contract exchanges regulated by the CFTC. Beneficiaries include trading and distribution platforms like Kalshi, Polymarket, Robinhood, and Coinbase, as well as exchange groups like Cboe, CME, and ICE that are positioning themselves in event contracts. Traditional sports betting operators like DraftKings and FanDuel, state governments relying on state betting taxes, and some retail options and crypto contract brokerage businesses are under pressure.
Supplementary data: Since 2026, the global prediction market's average monthly trading volume has exceeded $37 billion, approximately three-quarters of the total trading volume for 2025. In August alone, the trading volume of Kalshi's commodity contracts was over 200 times that of the entire year of 2025.
Source: Public Information
ABAB AI Insight
The prediction market in the U.S. has reached its current state through litigation and fines. In 1988, the University of Iowa launched the Iowa Electronic Markets, operating within an academic framework thanks to a non-enforcement letter from the CFTC. Dublin's Intrade was sued by the CFTC in 2012 and closed in 2013. PredictIt received a non-enforcement letter in 2014, which was revoked by the CFTC in 2022. Polymarket paid a $1.4 million fine to the CFTC in 2022 for operating without registration and blocked U.S. users; in November 2024, founder Shayne Coplan's residence was searched by the FBI, and after the federal investigation concluded in July 2025, it returned to the U.S. by acquiring the licensed exchange QCEX for $112 million. Kalshi has taken a compliant route: it obtained a CFTC Designated Contract Market (DCM) license in 2020, sued the CFTC in 2024, and won the election contract case, opening the door to political and sports contracts.
From a capital perspective, traditional financial infrastructure is directly investing. In October 2025, the parent company of the New York Stock Exchange, ICE, announced an investment of up to $2 billion in Polymarket. Kalshi secured continuous financing at the end of 2025, raising its valuation to the $10 billion level. Robinhood launched a prediction market section through Kalshi in March 2025, subsequently partnering with Susquehanna to acquire LedgerX (later renamed MIAXdx), and is building its own exchange and clearinghouse. CME has partnered with FanDuel to launch event contracts, and Coinbase has also connected with Kalshi. Their motivations are similar: to use the CFTC-regulated "futures" identity to bypass state betting licenses, turning sports, macroeconomic, and crypto price predictions into nationally sellable derivatives, capturing exchange fees and clearing revenue.
The closest historical analogy is 0DTE options. Since 2022, Cboe has expanded S&P 500 index options to daily expirations, and 0DTE options quickly accounted for about half of SPX options trading volume, essentially turning options into intraday binary bets. A counterexample is the standardized binary options launched by Cboe in 2008, which failed due to a lack of retail distribution channels. This time, retail entry points like Robinhood and Coinbase are filling that gap. The industry is currently in a phase of "single leader expansion plus traditional exchange group entry": Kalshi's share has risen from 35% to 60%, while Cboe, CME, and ICE are entering through clearing, cooperation, and equity paths, likely leading to a period of mergers and integration ahead.
Essentially, this is a transfer of pricing power. In the past, the risk pricing power of retail investors was dispersed among two types of institutions: sports bookmakers pricing based on odds and vig, and brokers and options market makers pricing based on complex derivatives spreads. Event contracts unify these two types of products into standardized contracts of "yes/no" plus price from 0 to 1, transferring pricing power to the exchange order book and professional market makers, with platforms only collecting fees and no longer taking positions. This shift occurs for three reasons. First, the CFTC's federal jurisdiction effectively covers state betting laws, spreading compliance costs nationwide. Second, the learning curve for binary contracts is much lower than for options, leading to a higher retail conversion rate. Third, the around-the-clock volatility of crypto assets provides high-frequency trading scenarios. Bernstein's adjustment of the sports share down and financial share up reflects a bet that these contracts will transition from "entertainment betting" to the mainstream form of "retail derivatives".
ABAB News · Cognitive Laws
The most profitable is not the one who bets correctly, but the one who prices the bets.
The simpler the financial product, the more retail funds it attracts.
Regulatory arbitrage wins first, compliant players reap later.