Insiders can and do trade on prediction markets, and traders have already been fined, frozen, banned, and in two cases criminally charged with it. The more useful question is where the line actually sits, because it is written in three different places that do not perfectly overlap.
The Quick Answer
On a regulated US exchange like Kalshi, insider trading is banned in writing: the exchange's CFTC-filed rulebook prohibits anyone who can influence an event, or who holds material nonpublic information about it, from trading contracts on that event. Known athletes, sports personnel, and referees are screened out of markets on leagues they are involved in, and Kalshi says its systems now aim to block candidates from markets on their own races. Federal law is narrower than the exchange's own rules, so a real gray zone survives between what the rulebook bans and what a regulator can actually prosecute. The exact rulebook text, the $35,000 George Santos settlement, the first criminal cases federal prosecutors have now brought, and the market classes where inside knowledge still slips through are all below.
The Fear Behind The Question
Spend a week reading prediction market forums, as I do more of than I should admit, and you will find the worry stated plainly. One of the year's top posts on the Polymarket subreddit put it this way: "Someone Keeps Predicting Trump's Travel Plans With Surprising Accuracy … Either they've done exceptional research, or they know something the rest of us don't." Another, on the prediction markets subreddit, flagged "Campaign staffers making 'thousands' betting on their own candidates using insider information."
Both posts are asking the same question this article answers, and the stakes are not niche anymore: per a March 2026 Congressional Research Service review, sports contracts alone account for more than 85% of trading volume on Kalshi, and a Kalshi-style event contract settles at $1 for the side that called it and zero for the side that did not. A contract trading at 60 cents is conventionally read as the market pricing a 60% chance, so anyone who already knows the answer is buying dollars for dimes from whoever is on the other side. That could be you. Notably, though, none of the enforcement cases you are about to read involve sports; the insider problem concentrates in the smaller slice of the volume, and we will get to why.
The honest answer to the fear has a shape worth stating up front. Every event contract is a race to the resolving fact, and three separate layers of rules decide who is allowed to run: the exchange's own rulebook, the CFTC's civil fraud authority, and the criminal law federal prosecutors have now started to use. Each layer catches something the others miss, and the gaps between them are exactly where the travel-plans trader lives. Hold onto that three-layer picture, because we are going to walk down it from the strictest layer to the loosest, and then through the cases that show each layer firing.
What Kalshi's Rulebook Actually Bans
Kalshi is a CFTC-regulated exchange, and its rulebook is filed publicly with the regulator. Under Prohibited Transactions and Activities, the KalshiEX LLC Rulebook bans insider trading in terms that reach past the standard federal courts apply, because breach of a duty is not an element of it. The operative line is worth reading in full:
"If a Trader is a decision maker, either directly or indirectly, or has any influence, either directly or indirectly, no matter the scale and importance of the influence, on the outcome of the Underlying (event) of any Contract, that Trader is prohibited from attempting to enter into any trade or entering into any trade, either directly or indirectly, on the market in such Contracts."
The clause immediately above it closes the other half. It bars any "Insider that has access to material non-public information that is the subject of an Underlying of any Contract," and defines an insider as "any person who has access to or is in a position to have access to material nonpublic information before such information is made publicly available." The same clause separately bans anyone who is "an employee or affiliate of a Source Agency" from trading the contracts that agency's data settles, which is why a staffer at the office that publishes the number cannot trade the market built on the number.
Notice what is missing from all three prohibitions: any requirement that you stole the information or betrayed anyone to get it. If you can influence the outcome, or you simply know it early, the market is closed to you. No other element required.
Those prohibitions sit inside the same filed rulebook that governs everything else about how Kalshi markets run. On top of the written rule, Kalshi runs a screening program that blocks whole categories of people before they trade. When it expanded in March 2026, the exchange said in its own guardrails announcement that "individuals involved in college and professional sports (including athletes, personnel, and referees) will be preemptively blocked from trading markets associated with sports in affiliated leagues they are involved in," using screening lists built with the integrity firm IC360, and that after an enforcement action against a candidate who traded his own election, "our systems will now aim to preemptively block these kinds of trades."
| Who | Blocked From | How It's Enforced |
|---|---|---|
| Athletes, Sports Personnel, Referees | Markets on leagues they are involved in | Preemptive screening lists built with IC360 |
| Political Candidates | Markets on their own races | Rulebook prohibition plus account screening |
| Anyone Who Can Influence An Outcome, At Any Scale | That event's contracts entirely | Rulebook ban, trade surveillance, account freezes |
| Anyone Holding Material Nonpublic Information | That event's contracts until the info is public | Rulebook ban, surveillance, referral to the CFTC |
The last row is the one most traders underrate. It covers people no screening list could ever name in advance, like a staffer who saw a schedule or an editor who saw a video before it posted, and it is enforced after the fact by surveillance rather than up front by a blocklist. Which raises the obvious next question: if the exchange's rule is that broad, why does anyone describe insider trading on these markets as a gray area at all?
Because the second layer, federal law, is narrower.
Where Federal Law Draws A Narrower Line
The federal rule that governs these markets is CFTC Regulation 180.1, a fraud rule modeled on the securities world's Rule 10b-5. Under it, trading on material nonpublic information is illegal when the information was obtained through fraud or deception, or traded in breach of a duty owed to the information's source. That covers the classic cases: an employee trading against confidential knowledge from work, a contractor trading on a client's secrets.
What federal commodities law does not do is ban informed trading outright. The Congressional Research Service's March 2026 review, Prediction Markets and Insider Trading Law, notes the CFTC's own position that "derivatives markets have long operated in a way that allows for market participants to trade on the basis of lawfully obtained" material nonpublic information. There is no parity-of-information requirement, no rule that every trader must know the same things. The same CRS review concludes that Kalshi's rulebook "appear[s] to extend beyond" the federal misappropriation standard, and that whether the extra conduct Kalshi bans could ever trigger federal liability "is unclear."
The exchange bans more conduct than the government can clearly prosecute. Every hard question in this subject lives inside that one sentence.
That gap is the legal heart of this whole subject. Separate statutes patch pieces of it. The STOCK Act amended the same commodities law to make it unlawful for members of Congress, congressional employees, executive branch employees, and judicial officers and employees to use nonpublic information from their positions to trade futures, options, or swaps. Title 18's wire and mail fraud statutes reach further still, though the CRS review flags two Second Circuit decisions, Blaszczak and Chastain, as a potential obstacle to Title 18 prosecutions generally. Congress has since taken runs at the gap directly, with the Public Integrity in Financial Prediction Markets Act of 2026 (H.R. 7004), the End Prediction Market Corruption Act (S. 4017), which would flatly bar the President, Vice President, and members of Congress from trading event contracts, and the Event Contract Enforcement Act (H.R. 7840) all introduced. Meanwhile federal prosecutors in Manhattan have brought the first criminal insider trading cases in the space, which we will get to shortly. The perimeter is tightening, but it remains a patchwork, and the rulebook's reach still outruns the prosecutor's.
None of which is theoretical. The policing layer has names, dollar figures, and public orders attached, and one settlement shows the exchange and the regulator working in sequence.
The Santos Case: A Worked Example Of Enforcement
Kalshi listed a contract titled "Who will attend the State of the Union?" It covered whether former Rep. George Santos would attend, and between February 12 and February 25, 2026, he traded it. Per the CFTC's order, he bought and sold positions in that market while posting on social media about whether he would attend, and those posts contained "a series of material misrepresentations and omissions," after which prices moved in favor of his positions. NBC News reported the shape of it: he talked up his plans to be in the crowd in the run-up, then said on X, minutes into the speech, that he had been waylaid at the airport.
Follow what happened next, because the first two layers fired in order. Kalshi's surveillance flagged the pattern, the exchange froze the account, and it referred its own customer to the CFTC and the Justice Department, per NOTUS. On July 31, 2026, the CFTC settled the matter: Santos disgorged $17,569.98 in profits, paid a $17,500 civil penalty, about $35,000 all in, and accepted a three-year trading ban. His attorney said the resolution "should not be mistaken for an admission of any wrongdoing, because it is not one."
Two details in that order matter more than the dollar figure. First, the agency charged manipulation, not insider trading: the theory was that Santos controlled the underlying event and misled the public about it, which is a different offense from trading on a secret. The most famous case in this space is not actually an insider trading case. Second, it ended in a negotiated order rather than a verdict, and the CFTC called it its first manipulation action in an event contract, which tells you how new this ground is. What is not contested is the arithmetic: the trades cost him roughly double what they made.
Santos is not an isolated file. Here is the exchange-flagged side of the public record in one place; the criminal cases come in the next section:
| Case | Market Class | What Happened | Outcome |
|---|---|---|---|
| George Santos | His own State of the Union attendance | CFTC found he traded an event whose outcome he controlled while posting misleading plans | $17,569.98 disgorged, $17,500 penalty, 3-year ban, no admission |
| Political Candidate (Feb 2026 CFTC Advisory) | Contracts on his own candidacy | Kalshi enforcement action; CFTC said he potentially violated Rule 180.1 | Fined and suspended by the exchange |
| Employee Of A Company Affiliated With A YouTube Channel (Same Advisory) | Contracts on the channel's videos | Kalshi enforcement action; CFTC said the same | Fined and suspended by the exchange |
| White House Teleprompter Operator | Mention markets on speeches he allegedly could read early | Kalshi's surveillance flagged atypical trades and froze the account | Out of the post; settlement talks with regulators reported, allegations unresolved |
The February 2026 CFTC enforcement advisory covering the middle two rows is blunt about the division of labor: exchanges have "an independent duty pursuant to the core principles of the [Act] to maintain audit trails, conduct surveillance, and enforce rules against prohibited practices," and the agency keeps "full authority to police illegal trading practices" on top of them. The row I keep coming back to is the last one, reported by CBS News in July 2026, because mention markets are the purest test of the whole system: a market that settles on whether specific words get said, traded by someone who allegedly had the script in advance. Reporting put the alleged profits above $100,000, with more than $90,000 of it frozen on the platform when the account was locked, and the allegations remain allegations. That reporting also surfaced the quietest guardrail in the whole system: per the same coverage, Kalshi now requires users to disclose their place of employment, and its policy bars trading on information gained through your job. A job becomes a screening input instead of something surveillance has to infer after the money is already down. Notice the pattern across every row: the exchange's own monitoring, not a whistleblower or a journalist, threw the first flag.
So the policing is real. It is also strongest where the trader's tie to the information is easy to draw, which brings us back to the travel-plans trader.
What Slips Through: The Honest Gray Zone
Start with the two complaints that opened this article, because they describe two different problems.
The campaign staffer case is the clearer one legally and the harder one practically. A staffer trading on internal polling may well breach a duty to the campaign, and a government employee doing the same runs into the STOCK Act. But the screening lists stop at the principals. When Kalshi's blocklist expanded, Rep. Alexandria Ocasio-Cortez called it "absolutely not enough", naming the "staff, advisors, consultants, cabinet secretaries, spouses, and more" a pre-trade screen cannot list. Kalshi CEO Tarek Mansour answered that the exchange "already bans, monitors, and enforces against all the groups you mentioned," and that the update "adds pre-trade screens to block people from participating at all — even the stock market doesn't do this yet."
Both of them are describing the same wall from different sides. A screen stops the people a list can name in advance; everyone else is a surveillance problem, which means the trade happens first and the catching happens after. That distinction is the whole practical difference between the two complaints, and enforcement against the unlisted still requires connecting an anonymous-looking account to a specific job.
When that connection can be drawn, though, it now gets drawn in an indictment. In April 2026, federal prosecutors in Manhattan brought the first criminal insider trading case in prediction market history against an active-duty Army Special Forces soldier who, per the Justice Department, helped plan the operation that captured Venezuelan President Nicolás Maduro in January 2026, then traded the outcome. The CFTC filed a parallel civil action the same day, and between the two filings the alleged mechanics are specific: a VPN used to reach Polymarket's offshore book, roughly $33,934 spread across Maduro-related contracts, and more than $400,000 out when the operation succeeded. A month later, per TechCrunch, federal prosecutors charged a Google engineer with making $1.2 million on Polymarket markets tied to Google's own Year in Search reveal, allegedly using confidential internal data. Every charge in both cases remains an allegation until a court rules.
And even the flagship case is not settled law. The soldier has moved to dismiss the CFTC's civil complaint on the ground that the Polymarket contracts he traded were not "swaps" at all, which would put them outside the rule being used against him, and a judge stayed that case while the argument plays out. Read that against the CRS review's warning and the shape is consistent: the conduct is easy to describe and the jurisdiction is still being litigated.
The gap did not close, it moved. Polymarket's offshore book also saw a sharp uptick in large purchases of contracts predicting US military strikes on Iran shortly before those strikes occurred in February 2026, and per the CRS review, some members of Congress have voiced concerns that those trades may have been based on nonpublic information. As of that review's March 2026 publication, no charges had been announced. The soldier was reachable because his job tied him to the event, and the engineer because the data came from his employer. Anonymous wallets with no institutional footprint remain far harder to reach.
The travel-plans trader is the deeper problem, because that trader may not be breaking any rule at all. Someone who studies flight trackers, motorcade patterns, and press pool logistics is doing research. Someone in the scheduling office is an insider. From the order book, the two are indistinguishable, and the Redditor's own framing, exceptional research or inside knowledge, is exactly the line the law struggles to draw. Federal rules allow trading on lawfully obtained information; the exchange bans trading by anyone positioned to know early; and the enforcement record so far shows the system catching people who traded markets about themselves or their own workplace, the cases where prosecutors or exchange surveillance can plausibly tie the trader to the information source.
From the order book, exceptional research and inside knowledge look identical: a trader who keeps being right early. The law needs a breached duty to act; the screen just shows a price moving.
What separates the exposed markets from the resilient ones is how the resolving fact gets made:
| Market Class | Who Makes The Resolving Fact | Insider Exposure |
|---|---|---|
| Elections, Game Outcomes | Millions of voters, or everyone on the field in front of cameras | Low on foreknowledge, nobody holds the answer early; non-zero on influence, which is why clause (z) and the screens exist |
| Macro Prints From A Published Series | A source agency, on a scheduled release the rulebook already fences off | Low to moderate: the ban names the agency's staff directly |
| Mention And Speech Markets | The speaker and whoever handles the script | High: the answer exists in writing before it is said |
| Travel, Appearance, Announcement, Release-Date Markets | A scheduling office or a product team | High: a small circle knows the date first |
Read the bottom two rows against the case list and the pattern is total. Santos, the teleprompter allegations, the Maduro operation, the Year in Search reveal, every one of them lived in that second class. The exposure is structural rather than a policy failure, and no rulebook fully closes it: the CRS review's own bottom line is that banning everything the exchanges ban may take new legislation.
Now apply that test to sports, because sports is where most of this audience actually trades and where more than 85% of Kalshi's volume sits. A game result is the safest thing on the board by this standard: it is manufactured live, in public, by everyone on the field, and nobody holds it in advance, which is exactly why not one enforcement case on the public record is a game-outcome case. But sports contracts are not only game outcomes. Player availability is decided in a training room and a coach's office days before it is announced, and a scratch, an inactive, or a late-week practice designation is a fact a small circle owns first. That is the same structure as a travel market wearing a jersey. It is also why the sports screening runs through IC360, the integrity firm the leagues themselves use.
The sports rule of thumb that falls out of the table: who wins is a public fact, and who plays is a private one until somebody announces it. The first carries less insider-information risk; the second deserves the same caution as an announcement market.
There is a counterargument worth naming, and it is the reason nobody has simply banned informed trading outright: an insider's dollar drags the price toward the truth, which is exactly why a prediction market's price is worth reading in the first place. As a forecast consumer, you benefit from that. As the counterparty holding the other side, you are the one funding it. Both are true at once, which is why the practical question is not whether the system is fair in the abstract but how you position around it.
How To Trade With Insiders In The Pool
The defenses here cost nothing but discipline, and they follow directly from the structure above.
- Weight your activity toward publicly resolved markets. Elections, game outcomes, and macro prints resolve on facts produced in public. Mention markets, travel markets, and announcement markets resolve on facts a small circle holds early. Most of my own volume stays in that first bucket, and it is not close; the most contested markets cluster in the second one for a reason.
- Treat an unexplained sharp move in an insider-prone market as information you do not have. In a thin book a snap is often just one large order doing mechanical damage, and the fastest venues, like Kalshi's 15-minute bitcoin markets, snap constantly without a whiff of inside knowledge, so check depth and volume first. But when a travel or mention market runs from 20 cents to 60 on no news, I stop asking what the crowd learned and start asking who does not need to learn anything. Fading that move is a donation. The working rule I hold myself to: in the mention, travel, and announcement classes I size to roughly half what the same read would get me in a game or election market, and I treat a move past about 15 cents with no public trigger as untouchable in either direction. Call it a rule of thumb rather than a measured insider tax, calibrated to the shape of the case record rather than to a study: every enforcement case on the public record sits in that class, and none sits in the other.
- Read the resolution terms before the trade, not after. Insider risk concentrates in markets with narrow, technical resolution criteria, because the people who wrote or execute the underlying event know exactly how the fine print lands. Knowing what happens at settlement is the cheapest edge available.
- Where comparable markets exist across venues, compare prices before you believe a move. The clearest illustration is the one the CRS review documents: the pre-strike Iran buying and the pre-capture Maduro buying both concentrated on Polymarket's offshore book, not across the whole event-contract world at once. Informed flow tends to land where the informed trader can actually reach, which means a spike on one venue that a comparable market elsewhere never confirms is a localized-information signal rather than a global repricing. Sports bettors already run this reflex when they shop the number across every major book on the live odds screen; pointed at event contracts, it is the fastest read available. And it argues for staying inside the regulated perimeter, where a filed rulebook, surveillance that has demonstrably referred its own customers, and a regulator with clear authority all exist, even though the Maduro charges proved an offshore book is no hiding place either.
The Bottom Line
So, can insiders trade on prediction markets? The people most obviously positioned to cheat, athletes, sports personnel, referees, and candidates, are the ones Kalshi now tries to stop before they place an order, and the written rule behind them is broader than stock market law: influence the event or know the outcome early and the market is off limits, full stop. The policing is real and has receipts, from a $35,000 settlement and a three-year ban for trading a market about your own attendance to an exchange that keeps reporting its own customers. And the gap is real too. Prosecutors reached the soldier and the engineer because their jobs tied them to the outcome, while the market classes where a small circle produces the resolving fact keep rewarding whoever is quietly early, and the line between research and inside knowledge there is one the law has not finished drawing.
Which lands you back at the travel-plans Redditor, who had it exactly right: exceptional research or something the rest of us don't know. On the markets built for public facts, that question has a good answer. On the markets built for private ones, it is the whole risk, and the only honest way to hold it is as a price rather than a grievance: half size in that class, hands off a 15-cent move with no public trigger, and the rest of the bankroll pointed at markets whose answer nobody owns in advance. If the mechanics underneath any of this are new to you, start with how prediction markets work and how these venues differ from sportsbooks, and weigh the full trust question with the receipts in hand.
FAQ
If someone on the other side of my trade turns out to be an insider, does my fill get cancelled? Almost certainly not, and that is the part worth internalizing. Kalshi's rulebook says that because of the nature of its contracts, "there will generally be no cancellation or adjustment of an erroneous trade except in extraordinary circumstances" — and that rule is about erroneous prints, not about a counterparty who turned out to know something. Nothing in the rulebook unwinds a fill on those grounds. Enforcement in the cases on the public record has run at the offending account instead: freezes, disgorgement, fines, suspensions, referrals. The teleprompter matter locked more than $90,000 in the alleged insider's own balance. Nothing in that process rewinds the counterparty's fill, which is why avoiding the exposure beats expecting a remedy.
Is it insider trading if I am simply better-researched than everyone else? No, and the CFTC has said so directly. It has said derivatives markets "have long operated in a way that allows for market participants to trade on the basis of lawfully obtained" material nonpublic information, and there is no parity-of-information requirement. The exchange rule is the stricter one: if you can influence the outcome, or you can reach the answer before it is public, Kalshi bars you regardless of how honestly you came by it. Flight trackers and press-pool logistics are research. A seat in the scheduling office is not.
This article is educational, not financial or legal advice. Enforcement actions described here ended in settlements or remain allegations unless a court has ruled. Event contract trading involves risk, platform rules change, and availability varies by state; always read the venue's current published rules. 18+. If it stops being fun, call 1-800-GAMBLER.
