On 12 June 2026 the Commodity Futures Trading Commission published a notice of proposed rulemaking titled Prediction Markets; Public Interest Determinations, which would rewrite Regulation 40.11 and add a new Appendix F to part 40 of the Commission's rules. Much of the commentary read it as the regulator tightening its grip. That is the least interesting thing about the document.
The proposal does something the category has been asking for since the first state cease and desist letter landed. It says, in ordinary regulatory language, what the statutory word "gaming" means, and when an event contract "involves" one of the activities Congress singled out. For the first time an exchange can take its own list of live markets, apply a published test, and reach a defensible answer before a regulator applies that test for them.
A definition constrains you. It also measures you. Our reading is that the second use is worth far more to a serious operator than the first one costs.
What the proposal is, and where it stands
The document is a proposed rule, not a final one. It carries regulatory identification number 3038-AF65, was issued by the Commission on 10 June 2026 and published in the Federal Register two days later. The comment period closed on 27 July 2026, and as of 17 August 2026 no final rule has been published. The Commission's own announcement quotes Chairman Michael S. Selig saying the CFTC "will protect the integrity of our regulated markets without standing in the way of responsible innovation." The voting summary attached to the document records that no Commissioner voted against it.
The rulemaking did not appear from nowhere. The Commission proposed rules on the same subject in 2024, withdrew them in 2026, and issued an advance notice instead, which drew roughly 3,500 public comments. A Congressional Research Service legal sidebar walks through how the current proposal fits the statutory scheme. The Commission also puts the category in proportion in its own economic analysis: total trading volume across CFTC registered prediction markets exceeded 25 billion dollars in 2025, against a regulated futures market with a notional value of roughly 31 trillion dollars.
Two things follow. The proposal is early enough that its language may still change, so nothing here should be treated as settled law. And it is late enough that operators who wait for a final rule before checking their book will be checking it under time pressure.
"Involve" now points at the event, not at the trade
Section 5c(c)(5)(C) of the Commodity Exchange Act, usually called the Special Rule, lets the Commission review event contracts that involve unlawful activity, terrorism, assassination, war, gaming, or activity the Commission determines to be similar. The statute never explained what involvement means, and the gap did real damage.
Proposed section 40.11(a)(3) closes it in one sentence. Contracts involve an activity, the text says, "if their settlement is determined by an occurrence, extent of an occurrence, or contingency in the activity." The subject of the inquiry is the underlying event in the world, not the act of trading.
That sounds technical and is not. The Commission had previously argued to a federal court that gaming was simply a synonym for gambling, meaning the practice of betting without limitation on what is being bet. The court rejected the reading, observing that under it "all event contracts would be subject to review under the special rule because they all involve purchasing (and thus risking money on) some contingent event with the hope of receiving a payoff." The Commission now concedes the point in its own words in the proposal, describing its earlier position as an error and its earlier order against a political contract as a misapplication of the Special Rule.
An agency writing down where it went wrong is unusual, and it is the reason the rest of the document can be used as a tool. A test that would capture every contract measures nothing.
The definition of gaming, in plain terms
Proposed section 40.11(b)(1) states that gaming means any activity that: participants typically engage in for recreation or to entertain others; is governed by rules; and includes measurable occurrences or outcomes that depend on the participants' luck, skill, or athletic ability during the activity.
Three elements, all of them about the activity being referenced rather than about the wrapper it is sold in. A football match satisfies all three. So does a poker tournament and a video game final. An election does not, and neither does an award decided on merit over time: the proposal treats those as contests rather than games, a distinction law firm analyses of the proposal flagged immediately as one of its most consequential moves. The Commission is explicitly asking for comment on the harder middle cases, including game shows, reality competitions and pageants, which is a fair signal that the boundary there is not fixed.
Note what the definition does not do. It does not declare sports contracts off limits, and it does not make gaming a synonym for prohibited. Falling inside the definition means a contract enters a review, not that it fails one. That is a distinction worth repeating to anyone who reads the proposal as a ban in slow motion.
Twelve factors that turn a definition into a checklist
The part of the proposal with the most immediate operational value is the factor framework. Proposed section 40.11(a)(5) sets out considerations that apply to every contract caught by the Special Rule: whether it provides meaningful hedging or price basing utility, yields economically useful information, or promotes responsible innovation; whether it presents particular risks of manipulation, settlement integrity deficits, or information leakage and exploitation of material non public information by insiders; and whether trading it would challenge the registered entity's own self regulatory tools and compliance infrastructure.
For gaming specifically, proposed section 40.11(a)(6)(iii) then splits into two lists. Six factors weigh against a finding that a contract is contrary to the public interest:
- settlement based on the aggregate outcome of professional or collegiate games, including final scores, point differentials, win loss results, tournament advancement, or season long performance metrics;
- settlement based on an individual's aggregate statistical performance over the course of a game;
- settlement referenced to publicly reported, league verified or otherwise objectively determinable data, rather than inherently subjective determinations;
- an underlying game subject to an established integrity framework, meaning a recognised governing body, a monitoring function, published rules of competition and disciplinary procedures;
- formal information sharing or coordination arrangements between the exchange and the relevant league, governing body or integrity monitor;
- appropriate surveillance and trading prohibitions maintained by the exchange itself.
Six weigh in favour of such a finding: games depending entirely on random chance; contracts settling solely on a participant's injury, including its duration, severity or medical diagnosis; contracts settling solely on officiating calls; contracts settling solely on a discrete action or occurrence in a game; contracts settling solely on physical altercations subject to penalty or discipline; and contracts settling solely on games below the collegiate level.
Appendix F explains the reasoning behind each one, and the reasoning is recognisable to anyone who has thought seriously about listing. Injury contracts are treated as creating perverse incentives around physical harm and as depending on medical information that ought not to leak. Officiating contracts are treated as resolving on a handful of discrete human decisions made under pressure. Discrete action contracts are treated as markets where a single player or coaching staff member can determine settlement. The Commission also notes that combat sports are not caught by the altercation factor, because contact is a sanctioned element of those events rather than misconduct.
Readers of our own responsible listing standard will notice how much of that overlaps with what we published in March 2026, before any of this text existed. We are not claiming authorship of a regulator's thinking. We are pointing out something more useful: an operator who had already applied a serious listing standard would find very little in Appendix F that surprises them.
Run the test on your own book
Here is the practical value. The twelve factors are written as questions an exchange can answer about its own contracts, using information it already holds. For every live market in the gaming category, an operator can ask:
- Does settlement turn on an aggregate outcome, or on one discrete act by one identifiable person?
- Is the settlement source publicly reported, league verified, or otherwise objectively determinable, and is that source named in the contract terms before the market opens?
- Does the underlying competition have a real integrity framework behind it, or only the appearance of one?
- Is there an actual information sharing arrangement with the governing body, documented, or just an intention to build one?
- Does surveillance cover this contract type specifically, including the trading prohibitions that go with it?
- Does anything in the book settle on injuries, officiating, altercations, or competitors below collegiate level?
An operator who runs that exercise learns two things at once. They learn which contracts would be hard to defend, which is the obvious benefit. They also learn where their documentation is thin, which is the benefit that pays off later, because the proposal's own procedure is built around producing documents under a clock.
This is Directive 05 in its most literal form. Standards get set against the jurisdiction that regulates next, not only the one that has not yet. The point of a published test is that you can sit it early.
The procedural half nobody reads
The proposal also rebuilds the process, and this is the part that a compliance team should read twice. Under the proposed text the Commission may begin a review only through a written determination, issued no later than 10 days after a contract is listed, which must identify the specific submission under review, the enumerated activity implicated, the specific contract terms at issue, and the factors the Commission considers to warrant review. That determination goes to the registered entity and is posted on the Commission's website.
From there the clock runs. The Director of the Division of Market Oversight must provide a written statement of the factual basis and legal theory within 15 days. The exchange may respond in writing within 30 days, with supporting data, economic analysis and proposed modifications to the contract. A staff recommendation, requiring the General Counsel's concurrence, may go to the Commission within 60 days and is provided to the exchange at the same time. The exchange may answer that recommendation within 70 days. An order, if any, comes at 90 days, and the review period can be extended only with the exchange's agreement or at its request. The Commission may also ask that trading be suspended while the review runs.
Three consequences are worth naming. The exchange now sees the reasoning against it in writing rather than inferring it. The exchange gets a formal opportunity to propose fixing the contract instead of only defending it. And because determinations are published, the category acquires a body of reasoned decisions it can learn from, which is exactly what has been missing while the argument played out through litigation. The 90 day figure in the proposal refers to this review clock, not to the comment window, and the two are easy to confuse.
What we would do before the rule is final
Nothing in the proposal obliges an exchange to act while it remains a proposal. That is precisely why acting now is the cheap option.
Audit the book against the twelve factors and write the results down. Not as a memo to the file, but as a document that could be handed to a regulator, an auditor or a counterparty. A contract that would score badly should be modified or delisted on the operator's own initiative and timetable, which is always less expensive than doing it on someone else's.
Fix the settlement documentation first. The objectively determinable data factor is the one most exchanges can improve fastest, and it happens to be the same discipline that reduces settlement risk for users. Naming the source hierarchy, the fallback and the dispute path before a market opens is Directive 01 and it is also a factor in your favour.
Build the integrity relationships before you need them. Information sharing arrangements with leagues and integrity monitors are listed as a favourable factor, and they cannot be assembled in the 30 days you get to respond to a written determination. Sports contracts remain the category's growth engine and its most exposed flank, and that exposure is not only regulatory.
The wider picture reinforces the point rather than softening it. The conflict between federal and state authority is unresolved: in July 2026 a federal judge in New York declined to shield Kalshi from state gambling enforcement, while days later a federal judge in Minnesota blocked that state's ban on preemption grounds, according to reporting on both rulings. The Commission's Enforcement Division had already issued a prediction markets advisory in February 2026. And the CFTC's Innovation Advisory Committee met on 20 August 2026 with prediction markets on the agenda, announced by the Chairman on 13 August 2026 and noticed in the Federal Register on 11 August 2026, with written statements from the public accepted until 27 August 2026.
None of that produces a stable answer for an operator deciding what to list next quarter. The proposed factor framework does, at least provisionally, and it is the most concrete guidance the category has been given. Legal analyses published in June 2026 reached broadly the same conclusion about its practical significance, even where they differ on individual elements.
The bottom line
A definition of gaming was always going to arrive. The question was whether it would arrive as a considered framework with published factors and a procedure attached, or as an enforcement action against whoever was standing closest when patience ran out.
What the CFTC proposed on 12 June 2026 is the first version. It is not final, parts of it will be argued over, and reasonable operators will disagree with individual factors. But it is written down, it is specific about what counts in your favour, and it can be applied by an exchange to its own contracts on a Tuesday afternoon without anyone's permission.
Markets that survive regulation tend to be the ones that were already behaving as though it had arrived. For the first time in this category, behaving that way does not require guessing what the standard is.