If a contract pays out when a football team wins, is it a financial instrument or a bet? Two courts have now said the same thing, and the answer matters more than it sounds.
The Sixth Circuit ruled that Kalshi’s sports-event contracts are not swaps under the Commodity Exchange Act, which means they do not sit inside the CFTC’s exclusive jurisdiction. Strip out the statutory language and the practical effect is this: states get to keep arguing that these products are gambling under state law. That is the crux of the prediction markets vs betting debate, and as of the 28 September 2026 reporting cycle, the exchanges are losing the part of it that happens in appellate courtrooms.
Meanwhile, Brazil moved to dismantle its regulated online betting market less than two years after it launched. One jurisdiction is fighting over what counts as a bet; another has decided it does not want the licensed version at all. Both stories point the same way for anyone building or using these products.
The two models, weighed against each other
Prediction markets and sportsbooks can produce a near-identical experience for the person clicking the button. You pick an outcome, you risk money, you get paid if you are right. The differences sit underneath.
A prediction market is an exchange. Participants trade contracts that settle at a fixed value if an event happens and at zero if it does not, and the price floats with demand. A contract trading at 40 cents on a dollar implies roughly a 40% chance, the same way decimal odds of 2.50 imply 40% (1 รท 2.50). The exchange is not your counterparty; another trader is. The venue earns transaction and trading fees rather than a margin baked into the price.
A licensed sportsbook does the opposite. It sets the price, takes the other side, and builds in an overround so the book’s implied probabilities add up to more than 100%. That margin is the house edge in betting form. It is transparent if you know how to read a price, and it is the reason the operator profits over time regardless of who wins a given match.
| Feature | Prediction market (event contracts) | Licensed online sportsbook |
|---|---|---|
| Legal framing claimed | Financial contract on an event outcome | A wager under state or national gambling law |
| Regulator claimed | CFTC, via designated contract markets | State or national gaming regulator, licence by licence |
| Counterparty | Another trader on the order book | The operator itself |
| How the venue earns | Trading and settlement fees | Margin (overround) built into the odds |
| Price formation | Market-driven, moves with order flow | Set by the operator’s odds compilers and models |
| Footprint | Single federal registration, national reach claimed | Jurisdiction by jurisdiction, with local tax and rules |
| Player protection regime | Financial market rules, not gambling-specific tools | Licence conditions: self-exclusion, deposit limits, RG obligations |
That last row is where regulators keep pressing. A licensed sportsbook has to carry state-mandated responsible gambling machinery. A federally registered exchange, on its own account, does not, because it was never designed as a gambling product. Whether that gap is acceptable is a policy question, not a legal one, but it colours every legal argument in the file.
What the Sixth Circuit actually decided
The appeal consolidated two contradictory lower-court outcomes. A Tennessee court had granted Kalshi a preliminary injunction against state officials in February. An Ohio judge had refused one. The Sixth Circuit affirmed the Ohio decision and vacated the Tennessee injunction, which is about as clean a split decision reversal as the exchange could have feared.
“For the foregoing reasons, we hold that Kalshi’s sports-event contracts do not constitute swaps as defined in the CEA and thus do not fall within the scope of the CFTC’s exclusive jurisdiction,” the court said.
The word doing the work is “swaps”. Kalshi’s preemption argument rests on the idea that its contracts are federally regulated derivatives, so state gambling law cannot touch them. Take away the swap classification and the exclusive-jurisdiction shield goes with it. The full opinion is on the public docket for anyone who wants the reasoning rather than the summary.
Notably, the Sixth Circuit lines up with the Ninth, which reached the same conclusion on sports-event contracts. Two circuits agreeing narrows the room for a “one bad panel” reading. It also makes the eventual question of how far these products can travel less a matter of federal registration and more a matter of fifty separate state positions, which is precisely the model prediction markets were pitched as escaping.
The CFTC is drafting rules regardless
The CFTC has indicated new sports-event contract rules could land within two months. That timing turns a courtroom argument into a rulemaking argument. Courts are deciding what these contracts are under existing statute; the regulator is deciding what it will permit going forward.
For operators the sequencing is awkward. Product roadmaps built on the assumption of national access via federal registration now sit between two unfinished processes: appellate litigation that has gone against the exchanges twice, and a rule text nobody has read yet. Anyone budgeting on the basis of one or the other is guessing.
Brazil takes the other route entirely
While US courts argue over definitions, Brazil moved to shut down the regulated online betting market it built. Less than two years from launch to dismantling is short by any regulatory standard, and it upends the assumption that the direction of travel for online betting is always one-way toward licensing.
The two stories are not the same fight, but they rhyme. In the US, the classification of a product determines who regulates it. In Brazil, a political decision determines whether a legal market exists at all. Either way, the licence or the classification you relied on last quarter is not a permanent asset. Operators who treated Brazil as a growth line in a 2027 model now have a hole in it, and players in an unlicensed market lose the complaint routes and protections that came with the regulated one. Prohibition rarely removes demand; it usually just moves it somewhere with fewer safeguards.
Where the definitional fights are spreading
The same weekend brought two smaller signals that the “is it gambling?” question is now the whole regulatory agenda rather than a side issue:
- Florida widened its crackdown on sweepstakes casinos, another category built on the argument that it is not technically gambling.
- Pennsylvania put forward a sweeping gambling bill covering sports betting, iGaming and fantasy contests in one package.
Sweepstakes casinos, fantasy contests and event contracts arrived by different routes, but each occupies space next to licensed gambling while claiming a different legal identity. Regulators are closing those gaps one product at a time, and the burden of proof has clearly shifted onto the product.
The verdict
On the narrow legal question, licensed betting wins the argument for now. Two federal appeals courts have said sports-event contracts are not swaps, which leaves prediction markets exposed to state gambling law rather than shielded from it. The commercial pitch of prediction markets, one registration instead of fifty licences, is exactly the part the courts have declined to endorse.
On the product question, prediction markets still have a real case. Exchange pricing, fee-based revenue and peer-to-peer liquidity are a genuinely different structure from an operator-set price with an overround, and traders who understand implied probability can see the difference. That structural argument survives the ruling untouched; only the jurisdictional one lost.
For readers, the practical takeaways are unglamorous. Access to event contracts in any given state may change with litigation or with the CFTC’s rule text, so treat availability as provisional. Understand which regime you are actually inside before depositing, because the consumer protections differ. And if you are trading sports outcomes rather than betting on them, the label does not change the maths: prices carry a cost, outcomes stay uncertain, and no structure removes the risk of losing what you stake.
Frequently asked questions
Are prediction markets gambling?
Legally it depends on the jurisdiction, and that is precisely what is being litigated. The Sixth Circuit did not rule that event contracts are gambling; it ruled they are not swaps under the Commodity Exchange Act, which removes the federal preemption argument and lets state gambling law apply.
Does the ruling stop Kalshi operating?
The decision affirmed Ohio’s denial of an injunction and vacated the Tennessee injunction, which strengthens those states’ enforcement positions. It does not settle the position in every state, and the CFTC’s forthcoming rules on sports-event contracts could reshape the picture again.
Why does the “swap” classification matter so much?
Because exclusive CFTC jurisdiction over swaps is what would block states from applying their own gambling laws. Without that classification, each state can take its own view of the same product.
What does Brazil’s ban mean for players there?
The regulated market is being dismantled less than two years after launch, so the licensed framework and the protections attached to it are being withdrawn. Players should be cautious about unlicensed alternatives, where complaint handling and safer gambling tools are typically weaker or absent.
Gambling and event-contract trading both carry a real risk of loss and are never a source of income. Set deposit and time limits, use self-exclusion tools if you need them, and seek support from a national helpline if your play stops feeling like a choice.
