Are Prediction Markets Legal in the US? The Federal-State Tug-of-War Explained

You can open an app, buy a contract on the Fed’s next rate move or a congressional election, and watch the price tick in real time like you’re trading a stock. It’s legal. Try to make that same trade in Arizona, and the state’s attorney general might call it a gambling operation and sue the platform out from under you.

That’s the reality of prediction markets in the US right now: federally legal, increasingly regulated, and under attack in state courts. The reason why is a genuinely fascinating tangle of derivatives law, a 1992 no-action letter, a D.C. Circuit ruling, and a regulator that can’t quite decide whether it’s building a rulebook or just reacting to events. So, are prediction markets legal in the US? The honest answer is yes, with a patchwork of caveats that could rearrange itself at any moment. Here’s the full picture.

Wait, what exactly is a prediction market?

Before we dig into the legal weeds, let’s nail down the instrument itself, because the legal argument only makes sense if you understand what you’re actually buying.

A prediction market is a venue where you buy and sell contracts whose value settles on the outcome of a real-world event: an election, tomorrow’s weather, a sports game, an economic number. The contracts are structured as swaps under the hood, that’s the legal detail that makes them derivatives rather than bets, and it’s worth remembering because it comes up constantly in CFTC filings. The typical contract is a yes/no question with a fixed payout of one dollar. “Will the Fed cut rates in March?”

You buy “yes” for 70 cents, or you buy “no” for 30 cents. When the event resolves, winners get a dollar and losers get zero. The price you pay is the market’s probability estimate: a 70-cent “yes” contract means traders collectively think there’s roughly a 70% chance the outcome happens.

The classic mental model is “Will it rain tomorrow?”, if you buy “yes” at 70 cents and it rains, you get a dollar back. If it doesn’t, you lose your 70 cents. The profit is the difference between your entry price and the payout. Taxes and platform fees chip away at that headline number, so the dollar isn’t quite a dollar in practice, but the structure is that simple.

There are more complex variants, multiple-choice contracts, outcome ranges with partial payouts, bundled combinations of yes/no questions, but they all work from the same principle, and they all have thinner liquidity because fewer people trade them. Stick with the simple yes/no stuff for now; that’s where the action is.

Are they just gambling? (The legal answer is no, and here’s why)

This is the question everyone actually wants answered, so let’s hit it head-on: under federal commodities law, event contracts are classified as derivatives, not wagers. That distinction is the entire ballgame.

The Commodity Exchange Act gives the CFTC exclusive jurisdiction over derivatives, which includes event contracts that derive their value from an underlying event. The legal logic works because these structures, swap agreements with binary payoffs, are financial instruments, not casino games. A daily fantasy sports site offers a bet. Kalshi offers a financial instrument that bets on a yes/no question. Same energy, different legal universe.

There’s also a structural difference that matters: prediction market transactions are peer-to-peer. There’s no house taking the other side of your trade. The exchange provides the venue and matches buyers with sellers, but it doesn’t hold the risk. In sports betting, the house absorbs your loss when you lose; in prediction markets, another trader does. That changes the regulatory analysis significantly.

Then there’s the hedging angle, which is where the “this is a legitimate financial product” argument gets its teeth. A citrus farmer in Florida can buy a weather contract that pays out if temperatures drop below freezing, effectively insurance against a freeze wiping out the crop. That’s not gambling in any meaningful economic sense; it’s risk offset. Speculation is the other side of that coin, you’re taking on risk to try to profit, and both are legitimate uses of the instrument.

The “gambling” label isn’t wrong so much as incomplete. These are speculative instruments with real financial risk. But the legal classification is structural, not moral, and the structure says derivative.

That’s also why prediction markets operate in 40+ states while sports betting is legal in only about half of them. The derivative classification created a different regulatory lane, and until recently, most states didn’t pay attention to it. That’s changing, more on that in a minute.

How prediction market regulation got here: a 40-year accidental timeline

Nobody designed this regulatory framework. It was assembled piecemeal over four decades, driven by academic experiments, enforcement actions, a financial crisis, and a court case that nobody saw coming. The timeline tells the story:

1988: The Iowa Presidential Stock Market (IEM) launches at the University of Iowa, the first modern prediction market. It’s an academic experiment, not a commercial venture, designed to see whether market prices could forecast elections better than polls.

1992: The CFTC issues a no-action letter to IEM, effectively a green light to operate without full regulatory approval, as long as it stays a not-for-profit research and education tool. The letter also allows up to 20 other universities to join the program. That no-action letter would come back to matter decades later as the intellectual precedent for “event contracts are different.”

2004: The CFTC approves Hedge Street Inc. as the first designated contract market for binary options, the first exchange to offer these as regulated products. It rebrands to Nadex in 2009, which you might recognize from the binary options world. Crypto.com acquires it in 2022. Wild ride.

2010: The Dodd-Frank Act amends the Commodity Exchange Act and gives the CFTC explicit authority to ban event contracts deemed “contrary to the public interest.” That’s the legal hook the agency would try to use against Kalshi years later.

2022: The CFTC fines Polymarket $1.4 million for operating an unregistered exchange and orders it to stop US operations entirely.

2024: Kalshi v. CFTC, the D.C. Circuit rejects the agency’s attempt to block congressional control contracts. The case that changed everything.

2025: Polymarket reopens in the US under a CFTC-licensed exchange called QCX LLC. The platform that was banned three years earlier is now back and regulated.

2026: The CFTC flips direction, withdraws its 2024 proposal to ban political and sports contracts, withdraws a September 2025 staff advisory that required pre-review of event contracts, and issues a new advisory plus an Advance Notice of Proposed Rulemaking signaling that it intends to build a comprehensive framework from scratch.

That last stretch is where the story gets genuinely interesting, so let’s zoom in.

What the CFTC actually does today (and why it’s about to change)

The Commodity Exchange Act doesn’t even define “event contract.” The term is a regulatory invention, and the CEA’s definition of a “swap” is deliberately broad enough to cover them. That’s the legal door that lets the CFTC claim jurisdiction, and also the door that leaves the whole framework feeling slightly improvised.

Designated market contracts (DCMs), the exchanges that offer event contracts, have to clear a stringent application process and pass periodic exams. The core principles that govern them are worth knowing because they’re the actual teeth of the regulation. Core Principle 3 requires that contracts not be readily susceptible to manipulation. Core Principle 4 requires exchanges to have the capacity to monitor and prevent manipulation. Core Principle 12 requires rules that protect markets from abusive practices. If a contract is obviously manipulable, say, tied to an event controlled by a single person, it shouldn’t get listed, but understanding how do prediction markets make money helps clarify why such rules matter.

Exchanges can list contracts two ways: self-certification, which lets them list quickly (at least one business day ahead) and is the default path, or prior CFTC approval, which takes longer but comes with explicit sign-off. Either way there’s a compliance explanation and supporting documentation. The CFTC can stay a self-certified listing if something looks wrong, which is exactly what happened with Kalshi’s congressional contracts in 2024.

That brings us to the current pivot point. In September 2025, the CFTC issued Staff Advisory Letter No. 25-36, which required exchanges to submit event contracts for Commission review before listing, a move that would have effectively killed self-certification for prediction markets. In February 2026, that advisory was withdrawn. Then on March 12, 2026, the CFTC issued a new advisory (No. 26-08) from the Division of Market Oversight alongside an Advance Notice of Proposed Rulemaking.

The advisory lays out the staff’s current views on how regulations apply to event contracts, it’s a roadmap, not a rulebook, and it explicitly notes that the same compliance rules apply across all event contracts. The ANPRM, meanwhile, asks 40 questions spanning core principle application, public interest determinations, whether “gaming” means the same thing as “gambling,” insider trading, wash sales, clearing house treatment, swap data reporting, and whether the old economic purpose test should make a comeback. Comments are due by April 30, 2026.

The signal is clear: the CFTC is done reacting and is moving toward building a comprehensive rulebook. The 2024 proposal that would have banned political and sports contracts outright was formally withdrawn in February 2026, citing state actions and litigation. But the ANPRM suggests the agency hasn’t abandoned the idea of tighter rules, it’s just trying to build a foundation that can survive court review this time.

Kalshi v. CFTC: the case that broke the dam

You can’t understand the current legal landscape without understanding this case. In 2024, the CFTC tried to block Kalshi from offering congressional control contracts, contracts that would pay out based on which party controls the House and Senate after the election. The agency argued these were “gaming” contracts contrary to the public interest.

The D.C. District Court disagreed, ruling under 7 U.S.C. § 7a-2(c)(5)(C)(i)(V) that the contracts were not “gaming” and were permissible. The CFTC appealed and asked for a stay pending appeal. The D.C. Circuit rejected the stay, and the reasoning is what matters: the court said the CFTC hadn’t provided sufficient evidence that the contracts constituted gaming, it had just asserted it. That’s a significant legal shift.

The agency can’t simply wave the “public interest” flag and expect courts to nod along. Under this precedent, the CFTC needs to demonstrate actual harm or a statutory basis for banning a given contract.

The practical effect was immediate and enormous. The ruling opened the door not just for Kalshi’s congressional contracts but for an entire category of election-based markets, which are a key part of what prediction markets are and how they operate. It’s the only definitive legal victory for prediction markets so far, and every platform currently operating in the US is standing on the foundation of that one court decision.

The state vs. federal patchwork (where things get messy)

Here’s where “it’s legal” gets qualified. The federal government says prediction markets are derivatives and falls under CFTC jurisdiction. Several states disagree, and they’re doing something about it.

The most aggressive moves have come from Arizona, whose attorney general filed a complaint in 2026 accusing Kalshi of operating a gambling site. Mississippi, Ohio, Nevada, New Jersey, and Pennsylvania have also sued Kalshi, Polymarket, or both. States with legalized sports bettingNevada, New Jersey, and Massachusetts, have filed cease-and-desist letters against prediction platforms, uncomfortable with the idea that a product that looks like gambling is operating outside their regulatory regimes. Utah and Minnesota have banned prediction markets outright.

Washington banned certain sports prediction and pop culture contracts effective August 2025, and New York has been openly hostile as well. Connecticut also appears in the list of states pushing back.

The core legal question underneath all of this is preemption: does the Commodity Exchange Act’s federal jurisdiction override state anti-gambling statutes? That question remains essentially unlitigated at the appellate level. The Kalshi case settled the federal question, but it didn’t touch state law, and it sidesteps the deeper philosophical question of are prediction markets gambling, which pits the wisdom of crowds against the house edge. The smart reading is that this patchwork eventually ends up in the Supreme Court, because you can’t have a platform regulated at the federal level and simultaneously accused of being an illegal gambling operation by half a dozen states without the conflict escalating.

The in-between states. California and Texas among them, are worth noting. Both have banned sports betting, but prediction markets operate there without explicit state action against them. The legal status is a legal labyrinth, and the only thing you can confidently say is that the ground is shifting.

Where can you actually trade right now?

If you’re in a state that hasn’t taken action, here’s the current landscape of platforms, and the differences between them matter more than you might think.

Kalshi. Launched 2021, offers sports, politics, weather, and entertainment markets. It’s the platform that won the pivotal court case, and it’s aggressively consumer-friendly: 3.25% APY on balances above $250 and a generous new-user promo. If you want a guided, CFTC-compliant experience, this is the default recommendation.

Polymarket. The crypto-native platform, founded in New York, famously fined and banned by the CFTC in 2022. It regained US access in May 2025 by acquiring QCX LLC, a CFTC-licensed exchange, and now operates entirely under that registered entity. It also landed official deals with MLB, MLS, and NHL, a sign of the legitimation the industry is hoping to ride. Probably the most interesting contracts and the deepest liquidity outside the binary sports markets.

Nadex / Crypto.com. The first DCM for binary options, going back to Hedge Street in 2004. It’s the institutional veteran, now under Crypto.com’s umbrella. If you want the longest track record, this is it.

ProphetX. Smaller player, peer-to-peer, and notably presents probabilities as American odds rather than decimal prices. Not available in all states.

The new entrants. Robinhood, Underdog, Crypto.com, Fanatics, FanDuel, and DraftKings have all launched prediction market products. For FanDuel and DraftKings, the prediction markets live inside the same sportsbook apps, which is a wild integration of two legal universes that courts are still trying to reconcile.

Your rights as a customer (and how to protect yourself)

Before we get into the how-to, let’s cover the protections that exist. You’re not completely on your own out here.

CFTC-regulated platforms are required to give you clear, complete information about risks, commissions, fees, and penalties. You’re entitled to accurate account statements and timely access to your funds. If something goes wrong, you can submit a complaint through the CFTC’s website, file with the National Futures Association, or go through arbitration. Whistleblowers have financial incentives under CFTC programs, and the agency’s Office of Proceedings exists to review disputed actions.

But, and this is the part nobody likes to talk about, registration isn’t a guarantee against losing money. The CFTC’s mandate is to protect markets from manipulation and abuse, not to protect you from your own bad trades or the inherent uncertainty of predicting future events. You can lose everything on a contract that seemed like a lock. The regulation makes the market fairer; it doesn’t make it safe.

Practical guidance is pretty straightforward. Do monitor positions closely, use stop-loss orders where available, and only trade with money you can genuinely afford to lose, the industry calls it “risk capital,” and it’s the right framework. Don’t trust promises of free money, celebrity endorsements, or “guaranteed” outcomes. Verify that any app you use is registered with the CFTC, the agency publishes a list of registered entities, and a quick check takes thirty seconds.

There’s also the insider trading angle, which is the sleeper risk in this whole industry. Prediction markets, by design, reward people with superior information. A soldier was indicted by the DOJ for using confidential military intelligence to make more than $400,000 in prediction markets on a military operation. That case defined the risk profile: prediction markets are uniquely vulnerable to anyone with non-public information about the event being traded, and the enforcement framework for that is still being built.

How to make your first trade (legally, obviously)

If you’re in a state that allows it, here’s the process. It’s easier than you’d expect.

You need to be 18 or older with a valid ID. Know-your-customer requirements mean a Social Security number or a live selfie verification, depending on the platform. Choose a CFTC-regulated platform. Kalshi, Polymarket, or Nadex/Crypto.com are the main options. Check your state’s status first: if you’re in a state with an active ban or lawsuit, you’ll likely be blocked at signup, but it’s worth checking before you put money in.

Funding is standard: Apple Pay, ACH transfer, or debit card. The minimums are modest, typically around $10. From there it’s a buy-and-hold game: pick a contract, buy “yes” or “no,” and either trade out before expiration if the price moves in your favor, or let it settle and collect your dollar if you’re right. You can also close losing positions early to salvage some value.

Taxes are the part everyone forgets. Your winnings are taxable income, and the IRS expects you to track it. The platforms provide statements, and there’s no way around reporting it. If the intersection of prediction markets and tax law is a rabbit hole you want to descend into, that’s a topic for another piece, but the short version is: keep records of every trade.

The bottom line on prediction market legality

So, are prediction markets legal in the US? Yes, nationwide, under federal law, where they’re classified as derivatives regulated by the CFTC. That’s not a loophole or an accident; it’s the product of forty years of regulatory evolution and a landmark court case. But “legal” and “accessible” are different things. State enforcement is real, active, and currently aimed at the two biggest platforms in the country.

Utah and Minnesota have outright bans. Arizona is suing. A dozen other states have joined or taken enforcement actions.

The current state of play is a tug-of-war between federal and state law, and it won’t resolve until the preemption question reaches the Supreme Court, or until the CFTC completes its new rulemaking and creates a framework that survives appellate review. Until then, the practical guidance is simple: check your state’s position, use a platform registered with the CFTC, and never put in money you can’t afford to lose. The prediction market era has arrived. The regulation has not yet caught up with it.

People Also Ask