There is a version of the Polymarket story that gets told at conferences and it goes something like this. A young founder builds a prediction market, the regulator shuts him out of his own country, he spends four years in exile, and then he buys his way back in, wins the licence, rings the bell, and the industry is legitimate at last. It is a good story. It is also, as of the middle of this year, quite badly out of date, because the licence turned out to be the beginning of the legal problem rather than the end of it.
The short history is worth setting out, because the sequence matters. In 2022 the Commodity Futures Trading Commission fined Polymarket USD 1.4 million for running an unregistered facility for trading commodity options, and as part of that settlement the platform agreed to stop serving customers in the United States. The main exchange moved offshore and stayed there. Then in July 2025 the company bought QCEX, a licensed derivatives exchange and clearing house, for a reported USD 112 million, which gave it something it could not build from scratch, namely an entity already holding Designated Contract Market status. That entity became Polymarket US. In November 2025 the CFTC issued an amended order of designation, and by the end of that year Americans could trade again, legally, on a federally supervised venue with surveillance obligations, clearing procedures and reporting duties attached.
That is a genuine achievement and it cost real money. What it did not do, and what almost nobody outside the industry’s own legal teams anticipated with any clarity, was settle the question of whether Polymarket is allowed to operate in any particular place in the United States.
The Ground War Nobody Priced In
Within weeks of the relaunch the letters started arriving, and they did not come from Washington. Nevada’s gaming regulator brought a civil enforcement action on the basis that sports prediction contracts are simply sports betting with a different label, and the platform geo-blocked the state. Tennessee’s sports wagering body issued cease and desist letters in January, requiring open contracts to be voided and deposits refunded. Massachusetts moved, and Polymarket responded by suing the state’s attorney general and its gaming regulators in federal court. Michigan followed. So did Kentucky and Rhode Island. On the first of July the company filed against New Mexico, asking a federal court to declare that state gambling law cannot reach a federally authorised market and to enjoin the state from proceeding against it. By that point something in the region of ten states had become entangled in litigation over prediction markets inside a matter of months.
The federal regulator has taken the industry’s side, and has done so aggressively, suing states in its own name to assert exclusive jurisdiction over event contracts. That is an unusual posture. It is not often that a regulator goes to court to defend the people it regulates against another arm of government, and it tells you how much of the sector’s future is riding on a single legal question.
Everything Turns on What the Contract Is
The question is deceptively narrow. Is a contract on whether a particular team wins a particular game a swap, which is to say a derivative traded on a federally regulated exchange, or is it a wager, which is to say gambling, which is to say a matter that has always belonged to the states?
If it is a swap, the Commodity Exchange Act occupies the field and the states have nothing to say. If it is a wager, a federal licence is a permission to run an exchange and not a permission to run an exchange in Nevada.
The courts have now given both answers. In April the Third Circuit held that sports event contracts are swaps under the Commodity Exchange Act and that federal preemption applies, which the industry treated, understandably, as vindication. A district court in Ohio went the other way, finding that these contracts are not swaps at all. And on the seventh of July, in what was probably the sector’s most significant legal setback of the year, Judge Analisa Torres in the Southern District of New York refused to enjoin New York’s gambling enforcement, holding that the preemption argument fails on all three of the available theories — express, field and conflict — in substantial part because the Commodity Exchange Act contains a rule that expressly preserves state authority over gaming. The Ninth Circuit has heard Nevada’s appeal and observers came away from the argument thinking the panel was not persuaded by the industry’s position.
A circuit split of that kind does not resolve itself quietly. Our earlier analysis of the SEC-CFTC regulatory framework flagged exactly this fault line. The chairman of the CFTC has said publicly that this looks like a Supreme Court question, and the platforms’ own markets have been pricing a meaningful probability that the Court takes a case before the end of the year. A decision would then be somewhere between six and eighteen months behind that. In the meantime the map of where an American can lawfully trade is drawn by geo-blocking software that has to be updated every time a district judge rules.
The Irony in the Insider Trading Case
There is a detail in the enforcement docket that captures the whole problem better than any brief. The CFTC has brought what is understood to be its first insider trading case involving event contracts, against a serviceman alleged to have used non-public government information to take a large position on whether Maduro would be removed from power, and to have made somewhere north of USD 400,000 when the contract resolved.
The defence is that the CFTC has no authority over him, because the things he traded were not swaps. They were bets.
That is the industry’s own preemption argument turned exactly upside down and pointed back at the regulator. If prediction contracts are derivatives, the states are shut out but the federal enforcement apparatus, including insider trading liability, market manipulation rules and position reporting, applies in full. If they are wagers, the federal apparatus recedes and fifty separate gaming regimes come forward. There is no version of the answer that leaves a platform lightly regulated, and any structuring exercise that assumes otherwise is building on sand.
What This Means if You Are Building One
Most of the projects we see are not Polymarket. They are smaller, they are often built outside the United States, and they are frequently designed by people who assume that a licence somewhere is a licence everywhere. The Polymarket year is a useful corrective, and the lessons transfer well beyond prediction markets — the same characterisation problem runs through the questions we set out in our earlier piece on CFTC prediction market regulation.
The first is that authorisation and permission are different things. A licence tells you that a regulator has accepted you into a framework. It does not tell you that every other authority with a claim over the same conduct agrees that the framework applies. Gaming regulators, consumer protection bodies, advertising standards authorities and tax authorities all have their own hooks, and none of them is bound by the characterisation the financial regulator adopted.
The second is that characterisation is the whole ball game. Whether an instrument is a derivative, a wager, a security or something else determines which regime applies, which is why the drafting of the contract specification, the resolution mechanism and the marketing copy matters far more than founders expect. Promotional material that describes a product as betting has a way of reappearing in a state regulator’s exhibit bundle, which is presumably why Polymarket has been auditing its own promotional content.
The third is that the offshore structure is not the escape hatch it once was. Polymarket ran its principal exchange outside the United States for years and still spent that period in litigation, under investigation, and eventually applying to come back in through the front door. Distance buys time and it buys optionality, but it does not buy immunity, and the cost of returning later is almost always higher than the cost of structuring properly at the outset.
Where We Come In
At Esquare Legal we advise digital asset and event contract businesses on exactly these questions, working across the United Arab Emirates, El Salvador, the British Virgin Islands and Panama, and the conversation we have most often is the one about characterisation: what the product is, legally, before it is a product commercially; which regulators can plausibly claim it; and what happens to the licensing analysis if a court somewhere decides the instrument is something other than what the term sheet calls it. It is the same discipline that shaped our view of where prediction markets are headed globally.
Those are unglamorous questions and they are much cheaper to answer in a structuring memorandum than in a preliminary injunction hearing. If you are building something in this space and the answer is not yet written down anywhere, that is the piece of work to commission first.
Esquare Legal advises prediction market and digital asset platforms on licensing and cross-border structuring. Contact us to discuss a characterisation review.
Author: Haider Ali, Associate, Esquare Legal.