On September 25, 2025, the Sixth Circuit Court of Appeals ruled against Kalshi. It held that the exchange's sports event contracts are not swaps. It held, further, that even if they were swaps, federal commodities law does not preempt the gambling statutes of Ohio and Tennessee. Three days later, on September 28, the Commodity Futures Trading Commission delivered two draft rules to the Office of Management and Budget for pre-publication review.
The first is a proposed rule. It would expand the definition of “swap” to capture event contracts, pulling them into the CFTC's exclusive jurisdiction. The second is an interim final rule. It would carve “casino-style” products out of that same definition, pushing them back to the states.
The docket timestamps are the finding. A regulator that loses an appellate ruling and files corrective text three days later is not clarifying a statute. It is rebuilding a jurisdictional claim before the window closes. Hype evaporates; receipts remain.
This is not a technology story. There is no contract to decompile here, no sequencer to inspect, no emissions schedule to model. What makes it worth a forensic read is that a swap definition is itself an instruction set. It decides which regulator's machine executes. Everything downstream — licensing, tax treatment, institutional capital access, the ability to sell into one national market or fifty state ones — is a function of where the line falls.
The line is not drawn yet. And the document that draws it has not been published.
Prediction markets are platforms where users trade binary contracts on the outcome of discrete events. “Will the Fed cut in December.” “Will a bill pass.” “Will a team win.” The contract settles at one or zero. Revenue comes from transaction fees and spread capture.
There is no token in this sector. That matters for how the sector should be read. There is no emissions schedule subsidizing activity, no reflexive asset whose price depends on the price, no treasury manufacturing volume to decorate a chart. The cash flows are real, they are transaction-based, and they are small relative to the narrative weight placed on them.
Kalshi is a CFTC-registered Designated Contract Market. That registration is the whole of its business case: a federal license to operate a national market. CME is a legacy derivatives exchange that already sits inside the CFTC's core jurisdiction and lists financial event contracts. Polymarket is the crypto-native venue, offshore, previously penalized by the CFTC, settling on-chain. State-licensed sportsbooks sit outside all of this and would benefit directly if event contracts were classified as gambling rather than as financial instruments.
The instrument in dispute is the swap definition, a product of Title VII of the Dodd-Frank Act. Title VII was drafted in 2010 to bring the over-the-counter derivatives market under supervision. It left the boundary between a financial swap and a wager under-specified. That under-specification is now the operating system of a multi-billion-dollar industry, and it is the reason two federal agencies and at least nine states are litigating over the same word.
The stakes are not only jurisdictional. They are fiscal. A contract classified as a swap generates capital gains treatment and settles inside a federal clearing perimeter. A contract classified as a wager generates wagering or excise treatment, settles under state licensing, and carries a compliance surface measured in fifty jurisdictions. For an institutional allocator, that difference is not a footnote. It is the difference between a market that can be sized and a market that has to be re-underwritten every time a state legislature convenes.
The two rules are not redundant. They are a hedge. The proposed rule runs the standard notice-and-comment path: months of process, adversarial comment, and a final rule that can be litigated before it ever binds anyone. The interim final rule is the fast lane: effective on publication, with comment collected afterward. Pairing them lets the agency move on two fronts at once — pull the financial end of the market inward with the slow rule, cut the gambling end loose with the fast one.
The core insight is that the CFTC is not defining a product. It is choosing which products it wants to keep.
A regulator that wanted to define would publish a standard. A regulator that wanted to choose would publish an exclusion and leave the boundary to enforcement discretion. The second approach is cheaper, more flexible, and harder to attack on its face. It is also what a party that has just lost in court would build.
The exclusion is where the text will be contested. OMB received the package without a published definition of “casino-style product.” The filing does not state the criteria. There is no metric, no threshold, no enumerated list, no resolution-source test.
An undefined exclusion term is not a drafting oversight. It is discretion wearing the grammar of a standard.
Consider the operational consequences. A sports contract and a macroeconomic-indicator contract are structurally identical. Binary payoff. Discrete resolution source. Defined settlement date. The only difference is subject matter. If the agency's line runs along subject matter, it is not applying a technical test. It is applying a policy preference. That is legitimate as policy and fragile as law, because it invites a reviewing court to ask what the standard is — and the record contains no answer.
The taxonomy problem is that no clean boundary exists. Three candidates present themselves. The first is subject matter: financial events versus sporting and cultural events. The second is resolution source: contracts settled by an index or official statistic versus contracts settled by a game outcome. The third is settlement mechanics: cash-settled versus physically-referenced. Each produces a different map. Subject matter cuts CME in and Kalshi out. Resolution source cuts macroeconomic contracts in and league contracts out. Settlement mechanics cuts almost nothing, because every event contract in this market settles in cash. The agency has not said which of the three it uses. Until it does, the exclusion is a placeholder.
I have audited classification systems of exactly this shape. In 2021 I dissected a major marketplace's on-chain royalty enforcement and found that the mechanism was bypassable by a simple wallet switch. The code looked robust. It passed casual review because reviewers checked the surface, and the surface was correct. The failure was one layer down, in the part nobody read.
Same failure mode here. A rule that says “casino-style products are excluded” passes the first read. The second read asks who defines casino-style. The filing provides no second read. I do not quote a project's marketing without independently validating the contract, and I do not quote a regulator's summary without the regulator's text. The text is not available.
Procedural exposure follows from the same choice. The Administrative Procedure Act requires that a final rule be a logical outgrowth of the record and that interested parties receive a meaningful opportunity to comment before the rule binds them. An interim final rule asserts that prior comment is impracticable, unnecessary, or contrary to the public interest. That finding must be justified. The justification is not in the OMB filing.
An interim final rule is a procedural admission of urgency. Urgency is a question of fact. The record does not yet contain one.
Regulatory feasibility is where I have spent the last year. Under MiCA, I audited the compliance infrastructure of three exchanges operating in Stockholm and found that only one ran cryptographically verifiable, zero-knowledge proof-of-reserve rather than a periodic attestation. The lesson transferred directly: a regime's credibility is a function of whether its tests are verifiable, not whether its intentions are stated. An exclusion term with no defined test is not a compliance standard. It is an attestation without a proof. The CFTC has published the intention. The verifiable test is missing.
Now the circuit map. Three circuits have ruled on the same question and produced three answers.
In April, the Third Circuit sided with Kalshi against New Jersey, supporting federal preemption. In August, the Ninth Circuit sided with Nevada, supporting state authority. On September 25, the Sixth Circuit sided with Ohio and Tennessee — and did so twice over. It rejected the swap classification, and it rejected preemption even assuming the classification held.
The second holding is the more damaging one. It removes the foundation under the rule before the rule is filed.
If a federal appellate court has held that federal commodities law does not displace state gambling statutes, then expanding the swap definition does not deliver what the agency needs. It can bring event contracts into CFTC jurisdiction and still leave them subject to fifty state licensing regimes. The proposed rule would then be formally successful and functionally inert.
Circuit splits are the standard trigger for Supreme Court review. New Jersey has already petitioned. The docket has not been granted. Until it is, the operative condition is a jurisdictional vacuum: the judiciary has narrowed the federal claim, the executive is attempting to widen it, and neither has final say.
Preemption, not the swap definition, is the terminal question. The definition is the vehicle. The destination is federalism.
Strip the doctrine and the game is legible. Three players. The CFTC wants a single national market under its supervision, because a single national market is the object it can regulate. The states want to preserve gambling tax receipts and licensing authority, because those are real cash flows with real beneficiaries. The platforms want whichever regime yields the larger addressable market at the lower compliance cost, and they will accept either regulator if the arithmetic favors it.
Ledger balances do not lie; they only wait. State gambling receipts are on the ledger. They are the reason nine states are in litigation rather than negotiation.
The CFTC has sued at least nine states. That number is the market-size signal. Regulators do not litigate across nine jurisdictions over a rounding error. They litigate when the tax base and the licensing base are worth defending, or worth capturing.
Add the SEC. In June, the CFTC and the SEC issued a joint request for comment on event contracts. SEC Chairman Paul Atkins has said the clarification is overdue and includes event-based products. CFTC Chairman Michael Selig has attributed the problem to ambiguities in Title VII and argued that ambiguity suppresses innovation.
Two independent agencies, two chairmen, one direction. The administrative branch is not ambivalent about where event contracts should live. It has said, at the principal level, that they belong inside the federal financial perimeter. Selig's framing is worth noting on its own terms: locating the defect in a fifteen-year-old statute is the standard move for an agency that intends to clarify by rule rather than wait for Congress to legislate.
The regulatory-capture question is not rhetorical. When two agencies coordinate to draw a line that separates one class of products from another, the question is which class survives the cut.
The economics being allocated are not capital. They are market access. An event contract classified as a swap trades inside a federal perimeter: one rulebook, one licensing standard, institutional counterparties that can transact without a state-by-state legal opinion. An event contract classified as gambling trades inside a fragmented perimeter: state licenses, state exclusions, state tax treatment, and an institutional buyer that must model fifty regimes before sizing a position.
The valuation gap between those two states is the entire asset. Prediction-market platforms are not being repriced on volume. They are being repriced on which of two regulators gets to say what they are.
The classification is the product. Everything else is throughput.
That is why the effective-on-publication design matters economically, not only procedurally. Participants cannot restructure products around a standard they have not seen. A carve-out that lands on publication day imposes an immediate compliance shock on the excluded segment while the included segment accrues a legitimacy premium in the same instant. The asymmetry is the intent.
The competitive mapping follows. If the exclusion runs along subject matter and separates sports from financial events, CME's existing financial event contracts are reinforced while Kalshi's growth engine — sports — is pushed into state-by-state compliance. If the exclusion runs along some other axis, the outcome changes. The filing does not say which axis it uses.
There is a third pending rule, on crypto asset trading and markets, still in the OMB pipeline. Its presence suggests event contracts are a component of a broader agenda rather than a standalone item. Both rules arrived at OMB without an economically-significant designation, which shortens the review cycle and, possibly, reflects a deliberate choice not to frame the package as major policy.
Not marking a rule economically significant shortens the review. It does not shorten the litigation.
Three outcomes are live. In the first, the Supreme Court takes the case and holds for the states. The CFTC rules are then largely academic and prediction markets revert to state-by-state licensing. In the second, the Court takes the case and holds for preemption. The federal perimeter closes and the included platforms gain a national franchise. In the third, the Court declines review. The split persists, the rules operate unevenly across circuits, and cross-state operation remains blocked for the foreseeable term. The second outcome is the one the bulls are pricing. The third is the one the docket currently implies.
The consensus bear read is that the CFTC's rule is a dead letter. Filed into a court that has already ruled against the theory. Filed three days after the loss. Filed with no published standard and a procedure built to be challenged. On that read, the rule is theater.
That read is half right and strategically wrong.
The bulls are correct about the thing that matters most. Prediction markets have real, transaction-based revenue. There is no emissions schedule, no reflexive token, no treasury subsidizing activity to manufacture a chart. Based on my audit experience, that absence is the single most important fact in the file. In 2022 I ran comparative game-theory models on algorithmic stablecoin designs before Terra collapsed. The failure was structural: the peg depended on a subsidy loop that required continuous new entrants. Nothing here depends on new entrants in the same way. Fees are fees.
The bulls are also correct that the current patchwork is genuinely costly. A platform that must satisfy fifty gambling regulators cannot serve institutional capital, and a market that cannot serve institutional capital cannot price risk efficiently. The demand for federal clarity is not manufactured. It is the demand of any market that wants to scale past its licensing map.
Where the bulls are wrong is in treating administrative action as the resolution. The Sixth Circuit did not rule on the swap definition in the abstract. It ruled on whether a federal classification can displace state law. No CFTC rule answers that. A rule can expand a definition. It cannot rewrite the Supremacy Clause analysis. The bulls are buying clarity from an agency that does not control the variable they need clarified.
Administrative clarity is not judicial finality. The market is pricing the first and needs the second.
There is a second blind spot. The same package that grants federal legitimacy to financial event contracts can strip it from sports contracts in the same filing. A rule that picks winners is not a rising tide. It is a reallocation, and the reallocation is not disclosed in the text because the text is not public.
Watch one date: the Federal Register publication of the interim final rule. That is when the exclusion's actual boundary becomes readable, and it is the largest un-priced event in this file. Everything before it is speculation about a document no one outside OMB has seen. Volatility is not risk; opacity is. The volatility on publication day is the symptom. The opacity before it is the disease.
The forward question is not whether event contracts are swaps. The forward question is whether a federal agency can define its way past a circuit split. If it cannot, the next two years are a state-by-state licensing exercise with a Supreme Court opinion at the end of them. If it can, the same filing that expands federal jurisdiction also decides which platforms remain inside it. Either way, the line is being drawn by parties who have not disclosed the standard. Read the rule when it prints. Not the press release.
