Two Rules, One Vacuum: The CFTC's Quiet Bid to Own Event Contracts
The docket appeared without ceremony. No press conference. No coordinated tweet thread. No friendly anchor teeing up the talking point. Just a line item on a regulatory calendar that most of the market will never read: two proposed rules, routed through the Office of Information and Regulatory Affairs, the White House's quiet gatekeeper for anything the executive branch deems significant. The subject line was unremarkable. The implications were not.
I have spent enough hours inside regulatory calendars to know that the interesting signal is rarely the headline. It is the routing. When a rule reaches OIRA, it has stopped being a press strategy and started being a legal artifact in motion. Someone at the Commodity Futures Trading Commission decided that prediction markets, the business of wagering on whether a specific future event will or will not occur, deserved not one rule, but two. And the White House, rather than waving it through on a routine clock, chose to weigh in.
Two rules. Two questions. Zero published answers. That is the entire information content of the event, and it is enough to move a multi-billion-dollar question: who, precisely, gets to decide what an event contract is?
Whale tails flicker in the NFT gallery shadows. Here the whale is a federal agency, and the gallery is a jurisdiction nobody has finished claiming.
To understand why two rules matter, you have to understand what a prediction market actually is once you strip away the wisdom-of-crowds language. Underneath the marketing, it is a derivative. A contract whose payoff is contingent on a binary real-world outcome. Did the candidate win? Did the central bank cut? Did the bill pass? Did the team score? The mechanics are as old as futures, a price that aggregates dispersed beliefs about the future into a single tradable number. The novelty is not the concept. The novelty is the settlement rail.
There are two architectural families, and they resolve differently. The first is the centralized, licensed model, exemplified by Kalshi, which operates as a designated contract market under CFTC oversight, settles in fiat, and answers to a compliance department before it answers to its users. The second is the on-chain model, exemplified by Polymarket, which settles in USDC on Polygon and delegates the question of whether the event happened to an optimistic oracle, UMA in Polymarket's case, where token holders stake capital to assert outcomes and dispute them.
The legal term that binds them is event contract. That phrase is not a marketing label. It is a jurisdictional claim in miniature. Call a product an event contract and you are arguing it belongs to the CFTC, the federal derivatives regulator. Call it a wager and you are arguing it belongs to state gaming commissions, the bodies that have policed betting since long before the internet existed. The naming is the entire fight. Whoever wins the naming wins the revenue, the rulebook, and the right to say no.
This is not a new argument. It is an old argument with a new costume. The CFTC has been circling event contracts for more than a decade. In 2012 it effectively shuttered Intrade, the most prominent prediction market of the pre-crypto era, by asserting that its contracts were off-exchange futures. For a decade afterward, the category went quiet in the United States, migrating offshore and, eventually, on-chain. Kalshi was founded in 2018 explicitly to fit inside the regulatory perimeter rather than fight it, and it spent years accumulating licenses before it dared to list anything controversial. Polymarket launched in 2020 with the opposite instinct, permissionless by design and offshore by necessity. In 2022, Polymarket settled with the CFTC for a seven-figure penalty over offering unregistered swaps and agreed to block US users. The two platforms are not just competitors. They are two theories of how to survive American regulation.
Then 2024 made the fight unavoidable. During the US election cycle, on-chain prediction volumes exploded into the billions, turning a niche crypto curiosity into a mainstream political instrument that cable news anchors referenced without irony. Kalshi, meanwhile, won a federal court ruling that let it list election contracts over the CFTC's objection, a decision that cracked the agency's posture of blanket prohibition. The result was an awkward equilibrium: prediction markets were demonstrably popular, legally contested, and structurally unresolved.
That is the backdrop for the two-rule proposal. When the CFTC routes rules through OIRA, it is doing more than drafting text. It is attempting to convert a patchwork of enforcement actions, court losses, and informal guidance into a durable federal framework, one that would preempt state gaming law by declaring event contracts a federally regulated derivative class. And when the White House weighs in, it signals that the question has escaped the agency and entered the political layer, where elections, moral panics about gambling, and inter-agency turf wars all collide.
The code whispered what the whitepaper hid: a prediction market is not a casino, and it is not a stock exchange. It is an information-pricing layer that happens to resemble both, which is precisely why no single regulator wants to let it go.
The Combination-Punch Hypothesis
When a regulator files two rules on the same subject, the instinct is to treat them as redundant. That is a mistake. Two rules usually mean two distinct legal levers, filed together because they reinforce each other. One is substantive. One is procedural. Together they form a frame.
The substantive rule is the territorial claim. It would define what an event contract is, which events are permissible, and, critically, whether the CFTC has exclusive authority over them. This is the rule that matters most, because it is the rule that decides whether a state gaming commission in Nevada or New Jersey has anything left to regulate. The procedural rule is the administrative machinery: registration requirements, reporting obligations, capital thresholds, data-sharing mandates. Drier, but decisive. A substantive claim without procedural scaffolding is a press release. Procedural machinery without a substantive claim is bureaucracy in search of a subject. Stacked, they become a jurisdiction.
My base case, and I want to be explicit that this is inference rather than fact, is that the substantive rule targets the legal boundary of event contracts while the procedural rule governs how platforms register and report. That is the standard pattern for building a federal regime from zero: claim the territory, then install the plumbing. I would assign moderate confidence to this reading, because it is the path of least institutional resistance and the one most consistent with how the CFTC has historically expanded its perimeter.
The OIRA routing is the second signal, and it is arguably the more revealing one. OIRA review is not a formality. Under the executive order framework that governs regulatory review, a rule reaches OIRA's desk only when an agency has determined it is significant, typically because it carries substantial economic impact, usually measured against a threshold in the hundreds of millions of dollars annually. The fact that these rules crossed that bar tells us the CFTC itself believes event contracts are economically material. It is a self-declaration of stakes. And the fact that the White House is actively weighing in, rather than letting the review lapse on a routine clock, tells us the political layer has been engaged. This is no longer a technical filing. It is a policy negotiation.
Let me be precise about what the White House weighs in can mean, because the phrase is doing a lot of work. OIRA sits inside the Office of Management and Budget, inside the executive branch. Its review is a coordination mechanism: it circulates a proposed rule to other agencies that might be affected, solicits their objections, and can send a rule back for revision. When OIRA takes its time, it is often because another part of the government has something to say. In this case, the most plausible interlocutors are the agencies with a stake in the outcome, the Department of Justice, the Securities and Exchange Commission, and the executive's own political advisors who understand that prediction markets is now a phrase that can appear in a campaign ad. I would read the delay as evidence of genuine disagreement inside the executive branch, not merely as bureaucratic slowness. Confidence: moderate. But the direction of the inference, that friction exists, is more reliable than any specific guess about who is objecting.
There is one more structural detail worth extracting from the fact that there are two rules rather than one. A single rule is a signal of confidence. An agency that knows exactly what it wants files once, cleanly. Two rules filed together suggest an agency managing internal disagreement by splitting the question into pieces it can win separately. If I am right, one rule is the one the CFTC is confident about, likely the procedural machinery, and the other is the one it is still negotiating, likely the substantive boundary. Watch which one emerges first. The order of publication is a tell.
The Legal Genealogy of Event Contract
Words in regulation are not descriptions. They are claims. The phrase event contract was engineered, not discovered, and its genealogy explains the current fight better than any policy brief.
For most of American history, betting on the outcome of an event was gambling, full stop. It lived in the domain of state gaming law, which meant licensing, taxation, and moral oversight. Then derivatives markets developed, and a new category emerged: contracts whose value depends on an underlying reference, settled through clearinghouses rather than bookmakers. The line between the two was never philosophical. It was institutional. Whoever could call a product a derivative could claim it for the financial regulators; whoever could call it a wager could claim it for the gaming regulators.
Event contracts sit exactly on that line, and the term itself is an attempt to move the product across it. Call something a contract and you invoke contract law, commercial exchange, and the CFTC. Call it a bet and you invoke chance, vice, and the states. The two-rule proposal is the CFTC trying to make the first word stick.
The genealogy matters because it determines which body of precedent applies. If event contracts are derivatives, then decades of futures regulation govern them: registration, capital, segregation of customer funds, market surveillance. If they are wagers, then a completely different body of law applies: state licensing regimes, gaming integrity standards, and the moral vocabulary of gambling control. These are not interchangeable. They impose different costs, attract different participants, and produce different market structures. A derivative is a financial instrument. A wager is a bet. The same cash flow can be either, depending on which word wins.
The Howey Misdirection
There is a temptation, whenever crypto meets a regulator, to run straight to the Howey test. Is this a security? Does it satisfy the four prongs: investment of money, common enterprise, expectation of profit, efforts of others? It is a useful reflex for token analysis. It is the wrong frame here, and understanding why is the whole game.
Run the test anyway, briefly, because the exercise is instructive. Investment of money: yes, users stake capital to take a position. Expectation of profit: yes, the entire point of a wager is a payout. Common enterprise: doubtful, because the two sides of a prediction market contract are adversaries, not partners, and common is a stretch when your counterparty is betting against you. Efforts of others: doubtful, because the outcome depends on a real-world event, not on the managerial exertions of an issuer. Two prongs satisfied, two in doubt. The composite reads more like a derivative or a wager than a security, which is exactly why the SEC is a supporting actor here, not the lead.
The real contest is between the CFTC and the states. The CFTC regulates derivatives, futures, options, swaps, and it wants to fold event contracts into that category because doing so gives it exclusive federal authority. The states regulate gambling, and they want to keep event contracts classified as wagers because doing so preserves their licensing revenue and their authority over betting. The CFTC's argument is essentially: this is a financial instrument, and financial instruments are ours. The states' counter is: this is a bet dressed in financial clothing, and bets are ours.
Here is the structural problem the CFTC cannot easily resolve. Its claim to exclusivity runs headlong into a legal doctrine called preemption, which determines when federal law displaces state law. Preemption is clean when a federal statute clearly occupies a field. It is messy when a federal agency is trying to preempt a field, gambling regulation, that Congress has historically left to the states and that the states have regulated for more than a century. The CFTC is not merely saying we have a rule. It is saying our rule erases yours. That is a claim courts will scrutinize, and it is a claim the states will litigate.
I have watched this movie before. In 2017, I spent four months reverse-engineering the smart contract logic of a failed ICO, reading through more than fifty thousand lines of C++ to trace where the money actually went. The lesson that stuck was not about that specific contract. It was about the gap between what a system claims to be and what its code enforces. Regulation has the same gap. A rule on paper is a claim. A rule that survives litigation is an enforcement. The CFTC has filed a claim. It has not yet survived anything.
The Oracle Problem: Where the Real Fragility Lives
Everyone is arguing about who regulates event contracts. Almost no one is arguing about the part that actually breaks: resolution.
A prediction market has exactly one hard problem, and it is not matching buyers with sellers. It is answering the question did the event occur objectively, in a way that no participant can manipulate. This is the oracle problem, and it is the technical heart of the entire asset class. Get resolution right and a prediction market is a pricing engine. Get it wrong and it is a mechanism for extracting money from anyone who trusted the outcome.
On-chain markets solve this with optimistic oracles. The pattern is elegant and fragile. Someone proposes an outcome and posts a bond. If nobody disputes it within a challenge window, the outcome is accepted. If someone disputes it, the question escalates to a token-holder vote, where the majority, weighted by stake, decides the truth. The security assumption is economic: lying should cost more than telling the truth. The vulnerability is that this assumption can invert when the value at stake in the market exceeds the value of the bond securing the resolution. That is the recursive failure mode I have been tracking since I built a Python pipeline to monitor fifteen thousand daily transactions across Uniswap, Compound, and Aave during DeFi Summer in 2020. I mapped the implicit dependencies between those protocols and found a liquidity contagion risk that nobody had priced, a recursive collateral cascade. The same logic applies to oracles. A resolution mechanism is only as strong as the ratio between what it protects and what it costs to corrupt.
Now layer regulation on top. The regulator's central fear about prediction markets is manipulation, the worry that a well-capitalized actor takes a position and then manipulates the underlying event, or the resolution, to cash out. This fear is legitimate in principle and, on-chain, it is technically real. A thinly capitalized oracle with a short challenge window is a target. A centralized platform with an internal resolution committee is a target of a different kind: it substitutes a single point of human failure for a distributed economic one.
This is where the two rails diverge most sharply, and where the regulatory debate quietly becomes a technical one. A licensed platform resolves events through a committee or a reference data provider, both of which can be audited, supervised, and sued. An on-chain platform resolves events through an economic game, which cannot be audited in the same way because its behavior is emergent rather than procedural. A regulator comfortable with supervision will always prefer the first, because supervision is what regulators do. The second is legible only to someone who thinks in incentives and attack costs. Most regulators do not.
The code whispered what the whitepaper hid, and here the whisper is that the regulatory debate and the technical debate are the same debate wearing different clothes. When the CFTC talks about manipulation risk, it is describing an oracle vulnerability in legal vocabulary. When a state gaming regulator talks about game integrity, it is describing the same thing. Neither has a technical fix. The fix, if it exists, is economic design: challenge windows long enough and bonds large enough that corruption is unprofitable. Regulation can mandate that design. It cannot substitute for it.
Two Rails, Two Regulatory Fates
The most underappreciated fact about the prediction market landscape is that the two leading architectures face opposite regulatory futures, and the two-rule proposal will decide which future wins.
Consider the on-chain rail. Polymarket settles in USDC on Polygon. It has historically restricted US users, settled with the CFTC in 2022 over offering unregistered swaps, and operated in a gray zone that its defenders call innovation and its detractors call evasion. Its architecture is permissionless: anyone with a wallet can trade, and resolution runs through a decentralized oracle. That permissionlessness is the source of both its appeal and its legal exposure. A permissionless system cannot easily run the KYC and AML checks that federal registration demands. It cannot easily restrict access by jurisdiction. It cannot easily answer to a compliance officer.
Now consider the licensed rail. Kalshi holds a designated contract market license, runs KYC, settles in fiat, and litigated its way to listing election contracts. Its architecture is centralized by design, which makes compliance tractable and decentralization impossible. It can restrict users by state. It can report to the CFTC. It can, in short, be regulated, because it was built to be regulated.
The two-rule proposal will allocate these rails differently. If the rules tilt toward strict federal registration with preemptive effect, they favor the licensed rail and squeeze the on-chain rail toward offshore operation or outright prohibition. If the rules tilt toward accommodation, recognizing that on-chain markets exist and trying to bring them inside a lighter framework, they could legitimize the on-chain rail and accelerate its adoption. The direction is unpublished. That is the vacuum.
There is an economic asymmetry that the compliance conversation tends to hide. A licensed platform pays for compliance directly, in legal fees, surveillance systems, and reporting infrastructure. An on-chain platform pays for compliance indirectly, by losing the US market entirely and operating where it cannot be served. Both bear costs, but only one bears them visibly. The visible cost becomes a line item that shapes pricing and market structure. The invisible cost becomes a geographic constraint that shapes who can participate. When the rules arrive, they will not create a level playing field. They will formalize an asymmetry that already exists.
And here is where my own prior work on institutional flows becomes relevant. In 2025 I built a real-time dashboard tracking institutional inflows into spot Bitcoin ETFs, analyzing roughly five million daily trade records to separate smart money accumulation from retail momentum. The single most consistent finding was that about seventy percent of institutional volume occurred during low-volatility windows. Institutions accumulate in calm, not in chaos. Apply that to prediction markets. Institutional capital does not enter a market whose legal status is unknown. It waits for the rules. The two-rule proposal is the gate. Until it opens, the sophisticated money stands outside, and the market is left to the retail speculators who happen to be comfortable with ambiguity.
The Composability Question
There is a version of this story that is about gambling, and there is a version that is about infrastructure. The version you believe determines what you think regulation should do.
Prediction markets, at their best, are an information-pricing layer. A liquid event contract is a continuously updating probability estimate, and probability estimates are inputs to everything. A DeFi protocol could hedge against an election outcome. A corporate treasury could buy protection against a regulatory decision. A structured product could pay out contingent on a macroeconomic print. This is composability, the property that lets one protocol's output become another's input, and prediction markets are unusually composable because their output is a number everyone can read.
But composability is a double-edged sword, and I have spent years mapping its edges. When I diagrammed the dependencies between Uniswap, Compound, and Aave in 2020, the point was not that composability is good or bad. The point was that composability transmits risk as efficiently as it transmits value. A prediction market integrated into DeFi imports its oracle risk, its settlement risk, and its legal risk into every protocol that touches it. If an event contract's resolution is corrupted, the corruption does not stay contained. It propagates. A lending protocol that accepts a position as collateral inherits the oracle's failure modes. A stablecoin that settles the contract inherits the settlement layer's congestion risk. The dependencies compound, which is exactly what I found when I mapped that first DeFi graph and predicted the flash-loan vector before it materialized.
Regulation determines whether prediction markets become infrastructure or remain a novelty. A clear legal framework makes them composable, because developers can build on them without fear that the foundation will be declared illegal overnight. A murky framework keeps them isolated, because no serious protocol wants a legally radioactive dependency. The two-rule proposal is therefore not just a gambling question. It is a question about whether the next generation of financial products can be built on top of event contracts, or whether that entire branch of the design space stays dark.
Preemption and the CME Shadow
Two more forces deserve naming, because they will shape the outcome regardless of what the rules say.
The first is preemption. The CFTC's ambition to make event contracts exclusively federal collides with a century of state gambling law. States like Nevada and New Jersey have built regulatory apparatus around betting, and they will not surrender it quietly. The likely battleground is litigation: a state sues to defend its jurisdiction, or a platform sues to escape state oversight, and a federal court decides where the line falls. This means the rules, even if finalized, will not be final. They will be the opening brief in a longer case. I would expect the practical timeline for clarity to be measured in years, not months.
The second is the CME. The Chicago Mercantile Exchange is the incumbent in regulated derivatives, and event contracts are a product category it can list. If the CFTC consolidates jurisdiction, the CME gains a compliant pathway to compete directly with crypto-native platforms. This is the competitive threat that the crypto prediction market ecosystem underweights: regulation that legitimizes event contracts also legitimizes the incumbent exchanges that can list them at scale, with institutional distribution, without any of the permissionless baggage. A regulatory win for the category can be a regulatory loss for the crypto-native participants in it. The rising tide lifts the biggest boats.
I have seen this pattern in another form. When I analyzed the wallet clusters around a blue-chip NFT collection in 2021, the consensus was that the market was about culture. It was not. It was about early-stage venture capital distribution wearing an art costume. The incumbents who understood the distribution mechanics won, and the enthusiasts who mistook the narrative for the mechanism lost. The same dynamic is setting up here. The crypto-native platforms are the enthusiasts, treating regulation as the enemy. The CME is the incumbent, treating regulation as the mechanism. Guess which one is better positioned.
The State Regulators' Case
It is tempting to treat the states as a footnote in this story, a fading obstacle to federal clarity. That reading is wrong, and it underweights the most motivated party in the entire dispute.
State gaming commissions are not passive. They derive revenue from licensing and taxation, they derive authority from statute, and they derive legitimacy from a century of practice. For them, event contracts are not a novel financial instrument. They are a threat to their revenue base dressed up in jargon. If the CFTC wins exclusive jurisdiction, every dollar currently flowing through state-licensed sportsbooks into event-style products migrates to a federally regulated venue, and the states lose both the revenue and the authority.
This gives the states every incentive to fight, and they have the tools. They can bring suit. They can coordinate through associations of gaming regulators. They can lobby Congress, which has the power to clarify the boundary by statute, and which has historically been protective of state prerogatives in gaming. The CFTC's two rules are not fighting a paper opponent. They are fighting a coalition with money, motive, and a home-field advantage in the courts.
What the 2024 Volume Signature Actually Showed
Let me bring this back to data, because narrative is where analysis goes to die.
The 2024 election cycle produced a volume signature in on-chain prediction markets that is worth reading carefully. The headline was the peak, billions in notional volume concentrated around a single event. But the interesting information is in the shape, not the peak. Event-driven markets have a pulse profile: volume builds as uncertainty rises, spikes at resolution, and collapses afterward. This is structurally different from the volume profile of a trading venue, which is roughly continuous. It is also structurally different from the volume profile of a sportsbook, which is seasonal but recurrent.
The implication is uncomfortable for anyone building a business on prediction markets. If the core use case is event-driven speculation, then the market is a series of pulses, not a stream. Revenue is lumpy. Retention is episodic. Users arrive for the election and leave when it resolves. That is not a durable business model. It is a series of campaigns.
The bear market makes this worse, because in a bear market the marginal speculator leaves entirely, and what remains is the pulse without the baseline. Strip out the event-driven spikes and what is left is a thin, illiquid residual that struggles to support market makers. This is the survival question the sector has not answered. It is easy to look impressive during an election. It is hard to look alive in a quiet quarter.
This is the signal I would flag to anyone trying to value the sector. The question is not how big the 2024 volume was. The question is what the volume looks like in a non-election quarter, in a bear market, with no catalyst. The two-rule proposal will not answer that. Only the tape will. And the tape, in a bear market, is unforgiving.
The Blind Spot: Correlation Is Not the Story
Here is where I part ways with the consensus reading.
The consensus says the two-rule proposal is a binary event: either it tightens, which is bearish for prediction markets, or it clarifies, which is bullish. Traders want to know which way it cuts. Analysts want to know when the text drops. Everyone is waiting for a direction.
That framing is wrong, and it is wrong in a way that matters. The blind spot is the assumption that a single federal rule resolves the jurisdiction question. It does not. The two-rule proposal does not end the federal-state conflict; it escalates it. It does not settle the courts; it invites them. It does not eliminate uncertainty; it institutionalizes it into a longer, more expensive process. The market is treating the proposal as an answer. It is a question that will be litigated.
There is a deeper contrarian point, and it is about causation. The volume explosion of 2024 is widely cited as the reason regulation is now on the table. But that gets the causality backward, or at least sideways. Regulation did not arrive because prediction markets grew. Regulation arrived because prediction markets grew in a way that touched a political nerve, election outcomes, and because a court told the CFTC it could not simply prohibit what it disliked. The volume was the symptom. The trigger was a legal loss and a political exposure. If you model the regulatory timeline as a function of volume, you will misforecast it. Model it as a function of litigation and electoral cycles, and the timeline becomes legible.
I learned this lesson the hard way. When I analyzed NFT whale behavior, the consensus was that the market was about culture, which pictures people loved, which communities they joined. I pulled the wallet clusters and found that a small number of entities, roughly a dozen percent of supply held by a few dozen wallets, bought systematically into dips. The market was not about art. It was about early-stage venture capital distribution wearing an art costume. The causation everyone assumed was backward. Correlation between cultural enthusiasm and price was real. Causation ran the other way.
The same trap is set here. The correlation between prediction market volume and regulatory attention is real. The causation is not that growth invites regulation. The causation is that legal vulnerability invites regulation, and growth made the vulnerability expensive to ignore. Those two models predict different things. The first says regulation tracks volume, so watch the tape. The second says regulation tracks litigation and politics, so watch the courts and the calendar. I would stake on the second.
And one more blind spot, specific to the crypto-native audience. The reflex is to read the two-rule proposal as an attack. It may be the opposite. A federal framework that legitimizes event contracts is the single most bullish structural development available to the sector, because it converts a legally radioactive product into an investable one. The discomfort is that the framework will almost certainly favor the licensed, centralized rail over the permissionless one. The crypto-native audience is about to discover that the enemy of its enemy, federal clarity, is not its friend. It is a competitor. The bear market taught everyone to fear regulation. The bear market also means the survivors will be the ones regulation decides to keep.
There is a final inversion worth naming, because it is the one the market will get wrong longest. The sector has been pricing prediction markets as a growth story, a category whose value scales with adoption. The regulatory reality is that prediction markets are a permission story, a category whose value scales with clarity. Growth without permission is fragile, because the entire franchise can be voided by a single enforcement action. Permission without growth is at least durable. The two-rule proposal is the market's first real attempt to convert growth into permission, and the conversion is going to be expensive, slow, and contested. Anyone who models the sector as a pure growth curve is modeling a business that can be switched off.
The signal to watch is not the rules. It is the routing.
If the two-rule proposal clears OIRA and appears in the Federal Register, the text will tell you which rail wins: the licensed or the permissionless. If it stalls or returns for revision, the political layer is still fighting, and the vacuum persists. Either way, the next real catalyst for prediction markets is a document or a judgment, not a product.
Four years of ledgers never lie, only distort, and the ledger here says the same thing it always says during a jurisdictional fight: the market will price the rumor long before the regulator publishes the truth. Position accordingly, or do not position at all.