Robinhood bought a license. Kalshi had built a business

For a year Robinhood was Kalshi’s largest distributor. Then it bought a CFTC-licensed exchange off the shelf, put Susquehanna behind the order book, and began routing its own flow to itself. The World Cup was the proving ground, the migration is under way, and the lesson is the one every platform eventually teaches its suppliers: the license was never the moat.

There is a sequence that plays out in every platform business, and the companies on the wrong end of it almost never see it coming, because the early years feel like partnership. A distributor takes a supplier’s product to its customers. The product succeeds. The distributor learns the economics, the operational requirements, and above all the size of the margin flowing past it to someone else. Then the distributor builds or buys the supplier’s function and keeps the margin. Amazon ran it on the merchants who taught it which products sold. Netflix ran it on the studios whose licensing bills it was paying. And this year Robinhood ran it on Kalshi, the prediction-market exchange it spent a year introducing to a hundred million retail accounts. The vehicle is Rothera, a CFTC-licensed derivatives exchange and clearinghouse that Robinhood and Susquehanna International Group acquired and rebranded, and the migration is already visible in the tape: the World Cup’s core markets routed to Rothera in June, the chief financial officer says most flow follows, and analysts report Robinhood customers now account for a shrinking share of Kalshi’s volume. This piece is the anatomy of that sequence, what it says about where value actually sits in prediction markets, and why the newest development, Robinhood reportedly negotiating to add a third party’s contracts alongside its own, is the most revealing detail of all.

What Rothera is, and what it cost to become one

The first fact worth internalizing is how ordinary the hard part turned out to be.

Rothera was not built. It was purchased: MIAXdx, previously known as LedgerX, a derivatives exchange and clearinghouse that already held Commodity Futures Trading Commission registration, acquired in a majority stake by Robinhood alongside Susquehanna International Group and renamed. That single sentence contains the entire strategic insight of this story. The regulated status that Kalshi spent years and a federal lawsuit securing, the designated contract market license and the clearing infrastructure that constitute the legal right to list event contracts in the United States, was available for purchase from an existing holder. Licenses are assets. Assets have prices. And a company with Robinhood’s balance sheet can buy in one transaction what a startup treats as its defining achievement.

The complementary piece was liquidity, and Susquehanna supplied it. One of the world’s largest quantitative trading and market-making firms serves as Rothera’s day-one liquidity provider, with both Susquehanna and Robinhood holding advisory-board seats. New exchanges usually fail at exactly this point, because thin books produce bad fills, bad fills drive traders away, and the absence of traders keeps the books thin. Starting with a top-tier market maker committed to the venue removes the failure mode that kills most new exchanges before their first quarter closes.

So the assembled package is license plus clearing plus institutional liquidity plus, critically, a customer base that already exists inside an app those customers open every day. Rothera’s contracts are also expected to carry lower fees for Robinhood users than third-party alternatives, which is the natural consequence of removing an intermediary’s margin from the chain. Everything a prediction-market exchange needs, in other words, except the years.

The migration, contract by contract

The rollout has been methodical enough to read as a case study, and the sequencing shows a company managing risk, not making a statement.

The first step was the quiet one: Rothera self-certified a baseball outcome event contract in mid-May with an intended listing date on or after May 20, and Robinhood began routing select Major League Baseball and economic-data contracts through it in a late-May soft launch. Small markets, unglamorous categories, minimal customer visibility, exactly where a platform tests new plumbing.

The second step was the World Cup, and the choice of venue was not incidental. Rothera’s tournament contracts were self-certified on May 27, and when the competition opened on June 11 across the United States, Canada, and Mexico, Robinhood routed the core markets, individual match outcomes, tournament winner, spreads, and totals, through its own exchange. A hundred and four matches over a month, with the largest event-contract volumes of the year attached to them, is the most demanding load test available, and Robinhood ran it on the venue it owns.

The third step is the one still under way, and its shape is the most informative part. Robinhood did not cut Kalshi off. Player-specific contracts, parlay-style combinations, and complex tournament props continued to route to Kalshi, with the company saying routing decisions depend on liquidity and resolution clarity per contract type. That is the textbook profile of a migration, not a rupture: keep the partner supplying the long tail that is expensive to build while taking the high-volume core that generates the revenue. Chief Financial Officer Shiv Verma has said publicly that most prediction-market flow is expected to migrate to Rothera over time, which converts the split from an operational nuance into an announced trajectory.

The numbers that made it inevitable

Understanding why Robinhood did this requires only the scale of what it was routing elsewhere.

Robinhood has processed more than 16 billion event contracts this year, against more than 12 billion across all of 2025, growth that made prediction markets one of the company’s fastest-expanding segments. The revenue line tells the same story from the other end: event contracts produced $147 million in a single quarter, exceeding the company’s cryptocurrency transaction revenue in the same period, an internal flippening this publication covered in its earnings analysis. A business generating that much revenue while paying an external exchange for the venue function is, from the platform’s perspective, a margin leak with a countdown attached, and the countdown ends whenever acquiring a license becomes cheaper than continuing to rent one.

Kalshi’s exposure is the mirror image. The exchange grew explosively on the strength of exactly this distribution, with Robinhood’s hundred-million-account retail machine supplying a large tributary of the volume that took Kalshi to roughly $31.5 billion in a single month and a $22 billion valuation. Analysts now report that Robinhood customers represent a shrinking share of that volume, and Kalshi’s own chief executive named Robinhood as one of its largest competitors in June, roughly a year after naming it a partner. Kalshi’s response has been to build directly toward its own users, launching a professional-tier product and expanding into perpetual-style contracts, which is the correct strategic answer, and also an expensive one for a company that until recently had distribution handled.

The asymmetry underneath is worth stating plainly, because it generalizes past this pair. An exchange’s assets are its license, its clearing infrastructure, its liquidity, and its distribution. Three of those four can be bought. The fourth, a customer base that opens your application every day, is the one that takes a decade and a brand, and it is the one Robinhood already had.

The Crypto.com signal: a shelf, not a store

Then, two days ago, the strategy revealed a further layer, and it changes what the whole exercise means.

Reports place Robinhood in talks with Crypto.com to offer that company’s prediction-market contracts inside the Robinhood application, alongside contracts already sourced from Kalshi, Interactive Brokers’ ForecastEx, and Rothera. A company that had just built its own exchange negotiating to carry a competitor’s products looks contradictory only if the goal was to own an exchange. It is entirely coherent if the goal is to own the shelf. Robinhood’s stated position is that it intends to work with multiple exchanges to give customers a broad and resilient marketplace, and read against the Rothera migration, that sentence describes a specific architecture: the platform routes each contract type to whichever venue offers the best economics or the deepest book, including its own, and captures the customer relationship regardless of where any individual trade clears.

That is a materially stronger position than vertical integration alone, and it maps onto the pattern our cluster coverage has been tracing from the other direction. The exchange operators bought their way toward the probability-data layer on the theory that owning the odds beats operating the casino. Robinhood is executing the third possibility neither of them centered: own the customer, and let the venues compete for the flow. In retail brokerage this is simply order routing, a business Robinhood understands intimately and has been litigated over before, and applying it to event contracts turns exchanges into interchangeable suppliers bidding for access to a distribution point they cannot replicate.

Which reframes the competitive question the whole sector is asking. The prediction-market war has been narrated as Kalshi versus Polymarket, regulated versus crypto-native, with a legislative overhang above both. The Rothera sequence suggests a different axis entirely: the venues are competing for volume that a small number of retail distributors control, and those distributors have every incentive to commoditize them. Kalshi’s $22 billion valuation prices continued category leadership. Robinhood’s build prices the possibility that leadership among venues is worth less than ownership of the front door.

The conflict nobody has priced yet

There is a structural problem inside this architecture that the competitive story tends to skip, and it is the one most likely to attract official attention: Robinhood now decides where its customers’ orders go, and it owns one of the destinations.

The company frames routing as an operational judgment based on liquidity and resolution clarity per contract type, which is a reasonable description of how any multi-venue router should work. It is also, precisely, a description of discretion exercised by a party with a financial interest in one outcome. When Robinhood routes a World Cup match contract to Rothera instead of Kalshi, the economics of that decision accrue to Robinhood twice, once as the distributor and once as part-owner of the venue and its clearing, and the customer has no visibility into the comparison that produced the choice. This is not a novel problem. It is the same structure that made payment for order flow the most litigated question in retail brokerage, produced a nine-figure settlement for this same company over disclosure of its routing economics, and remains a standing item on the regulatory agenda for equities and options. Applying the model to a newer product category does not make the question newer.

The mitigating facts are real and worth stating. Event contracts are not equities, best-execution obligations in derivatives markets work differently, and Rothera is a CFTC-regulated designated contract market with a clearinghouse, subject to that agency’s oversight instead of operating in a gray zone. Lower fees for Robinhood users, if they materialize as expected, are a genuine customer benefit that a vertically integrated venue can deliver and an arm’s-length partner cannot. A regulator examining the arrangement would find a licensed exchange, a licensed broker, disclosed common ownership, and a market maker with a public role, which is a considerably cleaner picture than the offshore venues occupying much of this category.

But the incentive asymmetry does not disappear because the entities are licensed, and the category’s regulatory environment makes scrutiny likelier and not less likely. Event contracts already face a bill that would ban sports markets outright, active litigation from a dozen state gaming regulators, and a congressional oversight probe into platform surveillance practices, all of which this publication’s cluster coverage has mapped. A retail platform routing customer orders to its own exchange, in a product category legislators are already inclined to treat as gambling, is a headline waiting for its hearing. The most valuable thing Robinhood could do about it is the thing platforms almost never do voluntarily: publish routing statistics per venue, per contract type, with the fee differential attached. Its absence will be noticed eventually, and the notice will come from somewhere less friendly than a competitor.

What to watch

Kalshi’s volume composition. The single decisive number: what share of Kalshi’s monthly volume originates from Robinhood accounts, and how fast it declines. Kalshi does not break this out, but its total volumes against Robinhood’s contract counts allow a serviceable estimate, and a sharp divergence between the two series would confirm the migration is more than tactical.

Whether the Crypto.com deal closes. Reports note there is no guarantee of an agreement. A signed deal confirms the shelf strategy explicitly; its collapse would suggest Robinhood prefers vertical integration after all, which is a meaningfully different future for every exchange in the category.

Rothera’s fee schedule. Lower fees for Robinhood users were the expected consequence of removing an intermediary. Whether the savings reach customers or stay with the platform is both a competitive variable and, given the company’s history with order-routing economics, a likely subject of eventual regulatory attention.

November’s routing. The midterm elections will produce the category’s largest political volumes ever, and where Robinhood routes those specific contracts, to its own venue, to Kalshi, or split, will be the clearest available statement of how far the migration has progressed under maximum load and maximum scrutiny.

One historical note completes the picture. LedgerX, the entity now trading as Rothera, was itself a landmark: the first federally regulated venue for physically settled crypto derivatives, later absorbed into a bankruptcy estate and sold, then sold again. Its license has now outlived two owners and two business models, and it arrives at its third life as the instrument through which a retail brokerage disintermediates the exchange that taught it the category. That is a fair emblem for where prediction markets sit in 2026: the regulatory permission that once looked like the industry’s scarcest asset has become a durable, transferable good, changing hands between owners with entirely different plans for it, while the genuinely scarce thing, an audience that shows up daily, was never for sale at any price.

Disclaimer: This article is for information and educational purposes only and does not constitute financial or investment advice. It describes commercial arrangements and reported negotiations that may change or fail to conclude, and figures reflect company statements and third-party reporting available at the time of writing. Nothing here is a recommendation regarding any company or contract. Always do your own research. Information is accurate as of July 26, 2026.

Frequently Asked Questions