Election-Staff Trading Bans Put Prediction-Market Money to the Test

Maricopa County, Arizona, banned about 13,000 of its employees from trading on prediction-market contracts tied to elections. Delaware County, Pennsylvania, barred roughly 2,200 workers from the same. In Los Angeles, the county registrar described threats and "aggressiveness from observers" after live market odds clashed with early vote counts. On the surface this is a pile of local ethics housekeeping — the sort of thing county boards do when a new fancy seems risky. Read it that way and you will miss the point. These bans are the first open stress test of a business that in the span of a year went from a $2 billion valuation to one near $40 billion, and that is now preparing to sell itself to the public.

To see why, you have to remember what a prediction market is. It is a machine built to pay the people who know things to put money on what they know. Odds on a contract only move when someone with an edge trades it, and that is precisely why the odds are information: they are the collective weight of people betting with access. On an election contract, the people with the earliest edge are not analysts in pundit chairs. They are the people who process ballots and tabulate returns days before the public sees them. The bans are not sermons about impropriety. They are an attempt to protect the product's integrity by removing the only participants who can break it — a structural fix, with the irony that the offending groups are the very reason the market has value in the first place.

A weekend-business problem becomes a public-company problem

This used to be a hobby market, and it no longer is. The main U.S. platform, Kalshi, booked more than $2 billion in annualized revenue, roughly double its run rate from three months earlier, and reported $16.81 billion in trading volume for a single month. About 85% of its recent transactions are tied to sports contracts. Kalshi raised $1 billion at a $22 billion valuation in May 2026 — a tenfold jump from a year earlier — and by July was reported to be in talks at a $40 billion figure. Its CEO says the company is thinking about going public, just not this year. That trajectory is the whole reason the integrity question now has a price tag on it: an institution that wants Wall Street's money and an eventual listing must first persuade investors that the market running on its platform is clean.

That case keeps getting harder to make. In August, Kalshi permanently banned former Representative George Santos for life and fined him over $71,000, finding he had bet on the State of the Union address using non-public information about whether he would attend. It settled separately with a former White House teleprompter operator who returned more than $100,000 in profits and paid a $65,000 fine for trading on advance word of presidential speeches. The highest-profile cases run through the crypto-native platform Polymarket: federal prosecutors allege a special-operations soldier wagered about $33,000 across 13 contracts on the operation to capture Nicolás Maduro — using classified knowledge, per the indictment — then cashed out more than $400,000 when the news broke.

None of this is accidental. The scandals are not exceptions to a well-functioning market; they are the same incentive running at its logical extreme. Prediction markets are valuable to the degree that informed people trade them, and the most informed people in any domain are, by definition, the ones whose information is not yet public. That is the structural contradiction at the center of the whole industry, and election contracts concentrate it because the set of people who genuinely know is small, defined, and — crucially — vulnerable to being walled off.

The wall was the only tool the governments had left

Watch who is doing the banning and it becomes clear why this is a governance story rather than an ethics story. Individual states largely cannot regulate these platforms directly. The CFTC, which designated Kalshi as a designated contract market back in 2020, takes the position that event contracts are financial products under exclusive federal jurisdiction and has sued states that try to block them. A federal appeals court even ruled Nevada could ban Kalshi as illegal gambling, creating a circuit split the Supreme Court may have to settle. So the one lever a county or a state actually controls is its own workforce: it cannot stop the exchange, but it can stop its poll workers and tabulators from trading on it.

That explains the shape of the response. Roughly 32 states already ban election betting in some or all circumstances, according to a Pew Research Center analysis of National Conference of State Legislatures data, and 44 states signed a letter calling prediction markets "a new form of casino." Several governors have issued executive orders barring state workers from insider-style betting. Maryland asked its state prosecutor to investigate whether election contracts violate state wagering law. But federal preemption keeps most of this at the level of employee bans and unenforceable bluster, which is why the platforms can shrug at states while their own self-policing — the fines and the lifetime bans — has become the real enforcement mechanism.

The result is a rulebook that protects a $22 billion market by exclusion, not by enforcement. Kalshi has no authority over the soldier who bets on Polymarket's offshore rails, or over a foreign intelligence service reading a surge in strike-date contracts as a signal that an operation is imminent. It can only chase the cases it can see and hope the rest never surface. That is a fine place for a privately owned startup to be. It is a very different place to be the day an S-1 lands and public-market investors start asking whether the clean-market covenant justifies the multiple — and whether the 85% of volume sitting in sports contracts, the segment most exposed to state gambling lawsuits, survives the courts at all.

What the reader should take from the odds

There is a personal corollary here for anyone who has looked at a 98% "win will happen" number on a prediction app and read it as truth. That number is not neutral polling; it is the output of a market that pays people to trade on what they know, and whose biggest watchers are exactly the insiders the industry is now trying to keep out. The institutions that justify Kalshi's valuation are only coming if the market can be shown to stay clean in front of it — yet the cleanest it can currently get is a patchwork of county employee bans, platform fines, and unsettled federal preemption, none of which reaches the offshore crypto rails where the worst cases have ridden.

So the election-worker bans deserve a second look not because they are a big story on their own but because they are the spot where the whole edifice is being decided. A prediction market's value is a bet that it can aggregate private information without being destroyed by the people who have it. Right now that bet is being secured by telling the poll workers they cannot play. The unresolved question — inside Kalshi, inside every investor weighing a future IPO — is what happens to the product, and to the people leaning on its odds, when the easiest insiders to ban have all been banned and the hardest ones remain.