When Baltimore City filed a lawsuit against Kalshi and Polymarket in early 2025, the market yawned. Another regulatory nuisance, another headline quickly buried under AI token mania. But strip away the noise, and this is not a routine enforcement action. It is a first-principles collision between federal derivatives authority and state gaming law, and it will determine whether the entire prediction market sector can exist in the United States without becoming a patchwork of 50 separate licensing regimes.
I have spent the last decade auditing protocols and tracing on-chain failures—from the 0x integer overflow that nearly drained an exchange, to the Compound flash loan vector I modeled weeks before the exploit. This case does not involve a smart contract bug, but it shares the same structural flaw: a mismatch between the system's design assumptions and the legal environment's default state. The code is not the law here; the capital flows are. And capital is king.
Context: Two Platforms, One Legal Trap
Kalshi is a CFTC-regulated designated contract market (DCM) that lists event contracts on economic and political outcomes. It won a landmark federal lawsuit against the CFTC in September 2024, securing the right to list political event contracts. Its entire business model is built on federal preemption: the idea that a CFTC license grants nationwide authority to operate event markets.
Polymarket is a crypto-native prediction market built on Polygon, using UMA oracles for settlement and a hybrid off-chain order book. It handled over $3.5 billion in trading volume during the 2024 U.S. election cycle, but in January 2025 it settled with the CFTC for approximately $250 million and agreed to block U.S. users from non-compliant markets. It has no federal license; its defense is technological globalism.
Baltimore's lawsuit accuses both platforms of operating unlicensed sports betting in Maryland. The city argues that predicting the outcome of a football game or a basketball championship is gambling, not a derivative contract. The legal question is deceptively simple: can a state reclassify a federally regulated financial product as gambling?
Core: The Asymmetric Damage Function
I modeled the impact of this lawsuit using a simple stress test: what happens if the state prevails? The answer depends on the platform's legal architecture, not its technology.
Kalshi's vulnerability is existential. Its entire value proposition is that a CFTC license is a universal pass. If a state court rules that the pass does not cover sports events, Kalshi must either obtain a sports betting license in every state—a costly, multi-year process—or remove all sports-related contracts from its platform. Sports contracts account for a significant portion of its trading volume post-election. The company's federal preemption defense is strong, but it has never been tested at the state level. The best-case scenario is a lengthy appeals process that drains cash reserves. The worst-case is a ruling that effectively revokes Kalshi's right to operate in the majority of U.S. states.
Polymarket's exposure is more contained. It already blocked U.S. users after the CFTC settlement. The lawsuit can only force it to strengthen geo-fencing and prove that no Maryland residents are accessing its markets. The financial penalty is a risk, but the operational impact is limited to legal costs and reputational damage. However, the reputational damage is not trivial. Polymarket's brand is built on being the premier global prediction market, and a state-level court labeling it an illegal gambling platform will scare off institutional partners and potential acquirers.
But the deeper insight is not about the two platforms individually. It is about the precedent. If Baltimore wins, every state with legalized sports betting—currently 38 states plus D.C.—will have a template to sue prediction platforms. The legal costs alone could kill the sector in the U.S. The CFTC's authority would be de facto circumscribed by state law, creating a regulatory patchwork that no centralized or decentralized platform can navigate efficiently.
Contrarian: The Bulls Might Be Right About One Thing
The conventional bear case is that this lawsuit accelerates the death of prediction markets in America. I disagree with the direction of the impact, but I agree with the mechanism. The lawsuit does not kill prediction markets; it accelerates the convergence of the sector toward institutional-grade compliance infrastructure.
Kalshi's legal team has already won a federal case against the CFTC. They know how to litigate. The Baltimore case gives them an opportunity to establish federal preemption at the appellate level, potentially creating a binding precedent that protects all CFTC-regulated DCMs. If Kalshi wins, the attack vector from state regulators is neutralized for years. Polymarket, meanwhile, can use the case to accelerate its pivot to non-U.S. markets and develop a licensing model similar to sportsbooks—paying for state licenses where it operates.
What the bulls miss is the cost. Legal battles are expensive. Even if Kalshi wins, it will burn millions in legal fees. The opportunity cost of not building product features during the litigation is real. And the threat of copycat lawsuits from other cities—Baltimore is one city out of 19,000—means that the company will need to invest in a permanent state-level compliance team, which is a fixed cost that reduces margin.
From my experience auditing the FTX collateral cross-contamination, I learned that the most dangerous risks are not the ones that appear in a single incident, but the ones that create a cascading compliance burden. The Baltimore lawsuit is a single node. The cascade comes when every state attorney general asks: "Why not us?"
Takeaway: The Federal vs. State Question Is the Ultimate Protocol
Prediction markets are not going to disappear. They are too useful for information aggregation, and the global demand for event-based trading is secular. But the Baltimore lawsuit forces a reckoning: the U.S. market is not a single jurisdiction; it is 50 independent jurisdictions with overlapping gaming laws. The only way to survive is to either secure a federal preemption ruling that is ironclad, or to treat each state as a separate market with separate licensing. The latter is prohibitively expensive for most startups.
I have seen this pattern before. In the 0x audit, the vulnerability was hidden in a function that seemed safe because it was never called with edge-case inputs. The Baltimore lawsuit is an edge case that the market chose to ignore. It is not a black swan; it is a slow-motion collision between two legal systems. The final judgment will be written in court dockets, not in whitepapers. And capital, as always, will follow the most certain legal path.
Code is law, but capital is king. Hype is leverage in reverse. The market is now pricing in a 30-50% probability that prediction markets in the U.S. face a structural reset. The next 18 months will tell us whether the reset is a correction or a collapse.