The exchange acquired gaming companies, and the data war ensued.

The parent company of the New York Stock Exchange just completed a $2 billion bet on Polymarket. Kalshi raised $1 billion at a $22 billion valuation and generated fee income that most exchanges envy. Seven bills in Congress want to ban such markets completely. Wall Street is not gambling on the prediction market; it is buying the probability layer of the financial system, and that difference explains everything.

Abstract

Intercontinental Exchange, the parent company of the New York Stock Exchange, completed a $2 billion commitment to Polymarket in March, of which $1 billion was invested in October, plus a new round of $600 million, and the platform is currently discussing a new round of financing at a valuation of nearly $15 billion. Kalshi raised more than $1 billion this spring at a $22 billion valuation, which nearly doubled in a few months, with its June transaction volume reaching $31.5 billion, compared with Polymarket's $10.8 billion, with annualized expense revenue estimated at $850 million to $1.5 billion. The key is deal structure: ICE purchased global distribution rights for Polymarket event data and within months launched a probabilistic data stream for institutions, an investment that Chairman Jeffrey Sprecher positioned as a new level of financial intelligence, rather than venture capital. Signs of integration are everywhere: CEOs of two rival companies jointly funded a $35 million industry venture fund, Kalshi reached institutional distribution through Tradeweb, Robinhood's event contract revenue exceeded its crypto business, and banks expect the industry to achieve $10 billion in annual revenue by 2030. It all comes against a backdrop of peak legal hostility: at least seven bills in the field in 2026, a bipartisan bill aimed at banning sports contracts outright, and the 50-state jurisdictional battles that this newspaper previously reported. This contradiction is being viewed by valuations as a temporary phenomenon.

Wall Street has a reliable signal: Watch what exchanges buy. Exchanges are the rent-collectors of the market; they make profits through their activities without taking on risks. When an exchange operator writes a ten-figure check, it has decided that a new type of activity is long enough to warrant taxation.

In March, the Intercontinental Exchange, which operates the New York Stock Exchange and twelve other regulated venues and has a market capitalisation of US$80 billion, just completed this judgment: it invested US$600 million in the crypto-native prediction market Polymarket, completed a total commitment of US$2 billion, and its valuation climbed from US$9 billion to its next round target price of US$15 billion. A few weeks ago, Polymarket's regulated archrival Kalshi raised more than $1 billion at a $22 billion valuation, double its winter valuation, and estimated annual revenue between $850 million and $1.5 billion. The founders of the two companies subsequently jointly set up a venture fund for their own industry, the corporate equivalent of a rival general co-founded military academy. At the same time, Robinhood's financial report quietly showed that its event contract revenue has exceeded its crypto business. Congress, which has witnessed all this, has proposed at least seven bills to restrict or ban this area.

What did ICE buy?

The most analyzed deals in the industry are also the most misunderstood because reviews price them as bets on betting, while the transaction structure tells a different story. ICE's $2 billion is not mainly used to purchase a share of transaction fees. It bought, in addition to equity, the global distribution rights of Polymarket event-driven data, and follow-up actions were underway within months: Polymarket Signals and Sentiment, launched in February, packages real-time predictive market pricing into a structured data stream for institutional customers, sold through the same set of ICE data machines that distribute bond pricing and commodity curves to every terminal around the world. Chairman Jeffrey Sprecher's statement was clear and clear enough to be understood word for word, rather than the rhetoric of a trading announcement: the investment adds a new level of financial intelligence. Translation in the language of exchange operators: Event markets produce a product that exchanges have never had before-the probability of continued pricing of the world's discrete outcomes (elections, interest-rate decisions, wars, product launches). Companies that have these probability distribution rights have something similar to what Bloomberg has in the reference data field. Trading is the factory; data is the product; and ICE, whose entire modern history has been transforming exchanges into data companies, is performing its signature operations on the latest existing exchange class. The tokenization partnership accompanying the deal, and Polymarket's acquisition of DeFi infrastructure startup Brahma to strengthen its on-chain architecture, paint a picture of the platform being adapted to the organization's underlying infrastructure rather than adding weight to the retail business. Read this way, other developments in the industry no longer look like bubbles, but more like an assembling supply chain. Kalshi's dealings with Tradeweb point in the same direction: Forecast pricing is distributed into institutional macro workflows, where the Fed's decision-making market is not a casino, but a hedging tool with a clearer definition of events than any interest rate option. The $35 million venture fund jointly supported by the two CEOs is an infrastructure signal: sowing the tool layers, Bloomberg terminals, and risk systems needed for mature asset classes. And Citizens Bank's forecast-a $3 billion industry revenue run rate and a credible path to $10 billion by 2030-is sellers starting to cover a category that can finally be modeled, because fee-based business based on transaction volume is something Wall Street knows how to value.

The scoreboard and why it is inconsistent with valuations

Now look at the numbers because they contain a real puzzle: the market leader by volume is not the platform bought on the world's largest exchange, and valuations conflict with operating data. Kalshi is ahead in operations and has obvious advantages. The transaction volume in June was close to US$31.5 billion, while Polymarket was US$10.8 billion, a ratio of about three to one; revenue estimates, whether it is a conservative fee figure of US$850 million or an annualized estimate of US$1.5 billion, are enough to make mid-end traditional exchanges envy; entering 2025, it will increase 994% year-on-year to US$260 million; the regulatory status of the CFTC makes it the only completely local and fully legal event trading venue in the United States; Distribution of retail machines through Robinhood, which has billion-dollar accounts; and the ongoing international expansion in Brazil. Its $22 billion valuation doubled in a few months, pricing continued leadership. Polymarket's operating story is even more undulating: The brand is bigger and the crypto native is more liquid. Through the acquisition of QCEX, it has obtained CFTC permission to return to the U.S. market from which it was expelled. In February, the trading volume reached US$23 billion due to the World Cup popularity, but in June it was only Kalshi's trading volume. One-third of its revenue is still in the process of starting, and the fee for taking orders has just begun to be collected this spring. The valuation is US$9 - 10 billion after ICE shares, and is now hitting US$15 billion. Despite stronger backing, But it is still one-third lower than its competitors. Solving this puzzle is the core argument of this article. Kalshi is treated as an exchange valuation: transaction volume, fees, growth, multiplied by a multiple. Polymarket is treated as an infrastructure valuation: the relationship with ICE prices not its current expense revenue, but its status as the probability layer that ICE intends to distribute, just as index providers value not based on their own revenue, but based on their irreplaceability among other products. Two different theories of predicting what the market is, with two different capital holdings, produce valuations inconsistent with the volume tables, and this inconsistency is the industry's living experiment: whether the money is made by running casinos or having odds. The honest answer, as can be seen in ICE's own history, is that the latter usually wins over decades, while the former funds wars during this period.

Paradox: simultaneous admission and prosecution

All of this is happening at a time when the U.S. legal system is escalating its hostility, a contradiction that is not a footnote but a defining condition in the industry. The litigation schedule, which has been detailed in our fifty state war report, will only become more intensive: at least seven bills targeting the forecast market have been proposed in 2026, led by the bipartisan bill proposed by Senators Schiff and Curtis, the Prediction Market is Gambling Act, which would directly ban CFTC regulated exchange-listed sports contracts; Nevada issued a temporary restraining order on Kalshi in March; more than a dozen state gaming regulators pressured, claiming that sports event contracts are bets dressed in derivatives; As well as intensified insider trading discussions surrounding suspicious transactions on geopolitical events on Polymarket's offshore books, they have sparked congressional questioning. The CFTC has stepped in multiple times on federal priority, supporting the position that the exchanges it oversees are accountable only to them, the precise structure of federally licensed versus state police powers, and the constitutional outcome of what we previously reported called the industry's real bet. Sports are the focus of war because sports are the focus of revenue, and the world of Schiff-Curtis bills would cut off the largest branch of the industry. So why are checks still cashed? Because institutional capital has concluded that conflicts will be resolved one way, and its reasoning deserves frank presentation rather than being dismissed as bluff. First, federal issues are being sued in areas where industry typically wins: the priority principle has historically biased in favor of federally licensed markets, and every court that allows CFTC venues to continue to operate has intensified the ratcheting effect. Second, the political economy changes with each acquisition: when the NYSE's parent company has odds tiers, when Robinhood's earnings rely on event contracts, and when Tradeweb distributes pricing, the ban is no longer a consumer protection bill, but an attack on the exchange infrastructure, which has the best lobbyists money can buy. Third, data arguments have legal jurisdictional exemptions, while transaction arguments do not: even a U.S. prediction market with sports banned and state governments troubled generates probability data about everything else (elections, interest rates, geopolitics), and data products purchased by ICE can be circulated globally regardless of where the transaction is licensed. When read correctly, these valuations do not ignore the legal battle; they are pricing the outcome: a federally integrated, agency-owned category whose bumpy state litigation is the cost of the moat being built, because the same legal complexities that plague existing businesses also deter new entrants. Whether this pricing is correct is a question of the past decade. And it is pricing, there is no doubt about that. November's midterm elections, the industry's largest trading volume event in history and its largest legal battle in history, will simultaneously serve as the first comprehensive test of both arguments.

The third front: retail distribution and the Robinhood wedge

Between Kalshi's regulated exchanges and Polymarket's crypto-native ledgers, there is a third front that this newspaper tracks from its earnings perspective, without which the industry landscape would be incomplete: brokers, who have customers that everyone is spending money to reach. Robinhood's numbers speak volumes before any theory was formed: single-quarter event contract revenue was $147 million, an increase of 320%, exceeding the company's entire crypto trading business. Based on 8.8 billion contracts, this distribution economy is unmatched by any independent platform, because marginal customers already own the app, account and balance. The company's response to its own data is vertical integration, the Rothera Exchange Project, transforming Robinhood from Kalshi's largest distributor to a future competitor, a sequence that economists on every platform recognize: distribute, learn profit margins, and then own a trading venue. The World Cup quarter set an industry record, Kalshi settled more than $31 billion in a month, and retail brokerage traffic was the main source, which bodes well for steady state: Forecasting volume becomes a standard feature of every retail trading app, just as options become a standard feature, while trading venues compete for channels like bettors. The retail front has also redefined the stakes of the legal war. Seven bills and state actions target trading venues, but the wedge that really brings event contracts into American homes is broker integration: regulated companies holding licenses that need protection, distributing CFTC listed contracts to mass-market accounts, and any legislative bans must therefore pass both the broker lobby and the exchange lobby, an alliance that has historically been the most effective in financial politics. It has also intensified criticism of insider trading and consumer protection, because retail customers buying election contracts in stock trading apps are the very participants the gambling framework fears, and the industry's response-self-monitoring mechanisms, contract design standards, position limits-will be formulated in this spotlight. The industry's integrity framework captures all three fronts simultaneously: the exchange war for trading volume, the data war for institutional relevance, and the distribution war for retail default options, and the same November stress tests will apply to all three fronts.

Things to Pay Attention to

Vote count for the Schiff-Curtis bill. The Sports Ban Act is the only legislative tool that can change revenue algorithms rather than compliance algorithms. Its committee's progress, and whether the imprint of the exchange lobby appears in the amendment, will reveal how much the political and economic landscape has shifted. Polymarket's U.S. re-entry mechanism. The conversion of the QCEX license into real-time, fee-for-service U.S. operations will be the first time that brings two competitors to the same regulatory arena and comparable income statements. Valuation gaps will be adjudicated there. Adoption of data products. Signals data stream on terminals, Tradeweb integration indicators, the first hedge fund letters to cite forecast market pricing as input: ICE's argument is falsifiable, and the evidence will be subscription revenue, not transaction volume. Mid-term elections in November. Record trading volumes are certain; the tests will be operational integrity at peak loads, insider trading headlines under maximum scrutiny, and whether state legal wars produce bans that actually disrupt trading activity. The industry's institutional era will either survive its first U.S. election as infrastructure or fail. Finally, the final calibration of the industry landscape opened up by this article: honest short views, stated without valuation optimism. The industry's revenue is concentrated in sports, which is exactly what a living bipartisan bill would ban; its growth statistics are exaggerated by the World Cup and faces a calmer 2027; its two leaders are burning venture capital on customer acquisition, locked in a cost war that has generated zero-fee promotional pairs; its insider trading surface, the event market in which participants can influence or predict outcomes, is worse structured than the stock market and has attracted congressional attention; Its institutional data arguments, no matter how elegant, currently generate only a fraction of transaction revenue, which means valuations are based on the parts most influenced by law and least proven under pressure. If the Schiff-Curtis framework is passed, if a state supreme court breaches the priority shield, or if a massive manipulation scandal occurs in November, the $22 billion and $15 billion valuations will be re-priced, becoming the same over-optimistic product of every previous wave of financial innovation when it was priced at its legislative peak. The bullish view, as noted above, is that exchanges have made the category too systemically important to kill. Bearishness is something that Congress has strangled systemically important before, and for a century, the legal normalization of betting has stumped savvy investors. This series of reports will hold both views at once, and it is the only honest way to report on an industry whose defining transactions happen to be binary contracts for its own survival. This is educational analysis, not investment advice.

Frequently Asked Questions

What exactly did ICE invest in Polymarket? A total commitment of US$2 billion: US$1 billion will be invested in October 2025, an additional US$600 million will be completed in March 2026, and up to US$40 million in secondary market shares will be purchased from existing holders. The core of the transaction structure is the global distribution rights and equity of Polymarket event data. In February, ICE launched Polymarket Signals and Sentiment, a data-streaming product for institutions, with tokenization cooperation. How do Kalshi and Polymarket currently compare? Kalshi leads operations: June transaction volume was approximately US$31.5 billion compared to Polymarket's US$10.8 billion, annualized fee income is estimated to be between US$850 million and US$1.5 billion, regulated by the CFTC, has Robinhood distribution channels and Brazilian expansion plans, and is valued at US$22 billion after raising US$1 billion this spring. Polymarket has a larger brand, crypto-native liquidity, ICE partnerships and a U.S. re-entry path licensed by the CFTC acquisition, with a valuation of US$9 billion to US$10 billion and is hitting US$15 billion. Why does the parent company of the New York Stock Exchange want a prediction market? For the sake of data, this is the view supported by the transaction structure. Forecasting that the market continues to price the probability of discrete events (elections, interest-rate decisions, geopolitical outcomes), a data product never produced by traditional exchanges, while ICE's modern business is both data distribution and transactions. Chairman Jeffrey Sprecher positioned the investment as a new level of financial intelligence, and the February launch of Institutional Data Stream implemented this argument. Is the forecast market really profitable? Leaders 'data strongly suggests yes, but it needs to be viewed with caution. Kalshi's revenue estimates range from $850 million in fees to an annualized $1.5 billion, an increase of 994% in 2025;Polymarket only started charging take-in fees this spring, with early annualized forecasts of approximately $300 million;Robinhood's event contract line reached $147 million in a single quarter, exceeding its crypto revenue. Citizens Bank predicts that the industry is operating at US$3 billion, with a path to US$10 billion by 2030. What are the specific legal threats? It's multi-level. At least seven bills targeting the field in 2026, led by the Schiff-Curtis bipartisan Prediction Markets is Gambling Act, will ban the trading of sports contracts on CFTC regulated exchanges, the industry's largest source of revenue. Gaming regulators in more than a dozen states are suing claiming event contracts are unlicensed bets, Nevada won a temporary restraining order against Kalshi in March, and the CFTC backed federal priority, a constitutional conflict painted in previous reports by this newspaper. Why does institutional funds continue to flow in in the face of the bill? Because it is pricing the finale that has been resolved: the priority principle has historically favored federal licensing, every acquisition changes the political economy, it is easier to ban gamblers than the NYSE's data layer, and probability data arguments can survive even under adverse rulings because data is distributed globally, no matter where transactions are licensed. Valuations view the legal battle as the cost of building a moat, consolidating existing companies while discouraging new entrants. What are the competitor-backed joint venture capital funds? A $35 million venture capital fund for the prediction market space, backed by the CEOs of Kalshi and Polymarket, is used to seed investment in event market infrastructure such as analytical tools and Bloomberg terminals. Rival operators jointly fund their industry's supply chains, which is the clearest signal yet that both sides view the industry's growth and its institutional legitimacy as shared assets that precede their competition. What should observers focus on next? Four things: Committee progress on the Schiff-Curtis Act, the only tool for changing revenue rather than compliance;Polymarket's QCEX-licensed U.S. operation goes online, putting two competitors on the same stage; evidence of adoption of institutional data streams, where ICE's arguments can be falsified; and the November midterm elections, when the industry's largest transaction volume event collided with the largest legal battle, the first comprehensive stress test in the institutional era. This is educational analysis, not investment advice.