The Day the Dam Broke

Introduction: The $150 Billion Illusion

On the eve of May 14, 2018, the American map of sports betting was a work of fiction. On its face, wagering was an activity penned into the neon corridor of Nevada and forbidden nearly everywhere else by federal law. Underneath that surface ran a river of unregulated cash that touched every state in the country, moved by neighborhood bookmakers and by offshore websites that treated the federal ban as a routing problem rather than a prohibition (Meer; “What Was PASPA”; “PASPA Explained”).

How large was that river? The figure nearly every account of the era reached for was $150 billion a year. Others used $380 billion. In 2014, NBA commissioner Adam Silver put it at $400 billion in an op-ed arguing for legalization. Follow those numbers back to their source and they collapse into a single sentence spoken at a 1999 meeting of the National Gambling Impact Study Commission, where a commissioner estimated illegal sports wagering at “somewhere, depending on whose guesstimate you take, within $80 to $380 billion.” Reporters who later went looking for the study behind that range found the commission’s own footnote pointed to a news story rather than to any research (Weissmann). The American Gaming Association’s $150 billion was the low end of that off-the-cuff remark, carried forward to 2017 dollars by the growth of gross domestic product, and the association itself called it a conservative estimate. An economist who instead benchmarked American behavior against the United Kingdom, where the wagering is legal and therefore countable, arrived at something closer to $67 billion (Zagorsky).

The point is not that one of those estimates was correct. The point is that the United States spent a quarter of a century regulating, and then deregulating, a market it had never measured, using a number nobody had ever calculated. That is the first mirror in this story. It will not be the last, because the industry that grew in the gap between the guess and the reality would go on to manufacture a great many more.

The law responsible for that gap was the Professional and Amateur Sports Protection Act of 1992, known ever after by its acronym, PASPA. For twenty-five years it stood not as an impenetrable wall against sports wagering but as a structurally unsound dam. It held back a reservoir of commercial ambition, state fiscal desperation, and steadily shifting public opinion. Yet the dam was poured along a known constitutional fault line, and its concrete was cracked before it set. It was always going to break. The only open questions were when, and how much would come through the breach.

The fall of PASPA was not merely a legal event. It was the starting gun for a frantic, multi-front gold rush in which state legislatures, corporations, and the very leagues that had lobbied for the ban raced to claim a piece of a newly legitimized frontier. This article deconstructs the flawed law that held the illusion in place, chronicles the decade-long legal rebellion that toppled it, and maps the landscape that emerged. It is a landscape that, eight years on, is already being redrawn again by a rival market the Supreme Court never contemplated.

A Law Built on a Constitutional Fault Line

PASPA was a product of the early 1990s, a moment when anxieties about the integrity of sport and the morality of gambling still carried real political weight. The bill was championed by a figure of unimpeachable sporting credibility: Bill Bradley, ten seasons a forward for the New York Knicks and by then a United States senator from New Jersey (“Professional and Amateur Sports Protection Act of 1992”; Moore; “What Was PASPA”). The statute is still called the Bradley Act. Its stated purpose was to protect the integrity of the game and to prevent the moral erosion that its proponents believed state-sanctioned gambling would bring (Holden and Ehrlich; Moore; “What Was PASPA”).

Testifying before a Senate Judiciary subcommittee in June 1991, the era’s most powerful commissioners made the case in person. David Stern of the NBA, Paul Tagliabue of the NFL, and Fay Vincent of Major League Baseball argued that widespread betting would irrevocably undermine public confidence in the authenticity of their product (“Professional and Amateur Sports Protection Act of 1992”; Moore; “What Was PASPA”). The Senate Judiciary Committee report that accompanied the bill, issued under the chairmanship of Joseph Biden, put the argument for federal action plainly: “Sports gambling is a national problem. The harms it inflicts are felt beyond the borders of those States that sanction it” (S. Rep. No. 102-248). President George H. W. Bush signed the act on October 28, 1992, and it took effect on January 1, 1993, freezing the national map of sports wagering in place (Professional and Amateur Sports Protection Act; “Professional and Amateur Sports Protection Act of 1992”; “PASPA Explained”).

The law’s fatal flaw was not its purpose but its mechanism. PASPA created no federal crime and carried no criminal penalty of any kind; enforcement ran entirely through civil injunctions sought by the Attorney General or by an aggrieved sports league (“Supreme Court Strikes Down Ban”; “U.S. Supreme Court Strikes Down PASPA”). What the statute made unlawful was for “a governmental entity to sponsor, operate, advertise, promote, license, or authorize by law or compact” a betting scheme based on competitive games (“28 U.S. Code § 3702”; Professional and Amateur Sports Protection Act). Read that sentence carefully and the strangeness surfaces. Congress did not regulate the conduct of bettors or bookmakers. It issued instructions to state legislatures about which laws they were permitted to keep on their own books. That is the essence of commandeering, a violation of the Tenth Amendment, which reserves to the states every power the Constitution does not hand to Washington (Boswell; Shannon; “Anti-Commandeering Doctrine”).

The design was a calculated risk rather than an oversight. The Supreme Court would later observe, in the very opinion that dismantled the act, that “Congress can regulate sports gambling directly, but if it elects not to do so, each State is free to act on its own” (Murphy v. National Collegiate Athletic Association; Sykes, “Sports Gambling”). A straight federal ban was available and constitutionally sound. Congress declined it, because a direct prohibition would have meant creating a new federal crime, funding its enforcement, and owning the politics of both. Nor was the anti-commandeering doctrine some obscure academic theory at the time. The Court had announced its modern form in New York v. United States on June 19, 1992, roughly four months before the president signed PASPA (New York v. United States; Sykes, “The Supreme Court Bets”). Congress passed the bill anyway, choosing a route it had reason to know was vulnerable and pushing the cost of the policy onto the states (“What Was PASPA”).

The statute did preserve the status quo where wagering already existed, though it never named a single state. Instead it exempted schemes according to when they had been authorized or operated, which in practice protected Nevada’s full single-game sports pools, Oregon’s NFL parlay lottery, Delaware’s parlay card scheme, and a narrow category of Montana sports pools and calcutta-style wagering; a separate provision carved out parimutuel racing and jai alai (“28 U.S. Code § 3704”; “Supreme Court Strikes Down Ban”). Buried in the same section was a clause written with Senator Bradley’s home state in mind. A casino betting scheme was permitted in any municipality where commercial casino gaming had operated throughout the ten years ending on PASPA’s effective date, provided the scheme was authorized within one year of that date (“28 U.S. Code § 3704”). Only Atlantic City qualified. New Jersey had until January 1, 1994 to take the offer, and in what would prove one of the more expensive miscalculations in the state’s legislative history, it let the window close (“Professional and Amateur Sports Protection Act of 1992”; “Murphy v. National Collegiate Athletic Association”).

The Garden State Rebellion

Two decades after letting that opportunity lapse, New Jersey was in no position to be fastidious. Atlantic City was hemorrhaging revenue, squeezed by new casinos in Pennsylvania, New York, and Maryland that had not existed when PASPA passed (“U.S. Supreme Court Strikes Down PASPA”). The state went looking for a lifeline and found it in the one enormous revenue stream federal law had placed off limits. In November 2011, New Jersey voters approved a binding constitutional amendment permitting sports wagering at Atlantic City casinos and at the state’s racetracks. The margin was not close: 671,797 votes in favor to 379,339 against, roughly 64 percent to 36 percent (New Jersey, Division of Elections; “Professional and Amateur Sports Protection Act of 1992”).

The legislature moved quickly, passing the Sports Wagering Act of 2012 and building a full state regulatory apparatus around it (“Murphy v. National Collegiate Athletic Association”). The response from the sports world was just as quick. The NCAA and the four major professional leagues, the same entities that had lobbied PASPA into existence, sued. They argued, correctly, that New Jersey’s law was a flagrant violation of the federal statute. The federal courts agreed, struck the 2012 act down, and enjoined the state from implementing it. New Jersey’s fallback argument, that PASPA was itself unconstitutional, was rejected out of hand by the lower courts (Honan).

Defeated but not deterred, the state’s lawyers returned in 2014 with something far more sophisticated. Rather than authorize sports betting, the new law simply repealed New Jersey’s own prohibitions on it, and repealed them only in part. The repeal reached wagering at licensed casinos and racetracks, by people twenty-one and older, and it pointedly did not reach wagers on New Jersey college teams or on any collegiate event held in the state (Murphy v. National Collegiate Athletic Association; Sykes, “The Supreme Court Bets”).

This partial repeal was a trap, and an elegant one. The leagues sued again, and to win they had to argue that a selective repeal amounted to a de facto authorization under PASPA. New Jersey answered that it had merely stopped forbidding something inside its own borders, and that requiring a state to keep its own criminal prohibitions on the books was commandeering in its purest form (Murphy v. National Collegiate Athletic Association; “Murphy v. National Collegiate Athletic Association”). The 2014 law weaponized the statute’s internal contradictions. It maneuvered the leagues and the federal government into insisting that a sovereign state lacked the power to repeal its own laws, which is close to a textbook definition of an unconstitutional command.

Judgment Day: The Supreme Court Dismantles the Dam

After the Third Circuit Court of Appeals sided with the leagues once more, the Supreme Court granted certiorari in Murphy v. National Collegiate Athletic Association (Honan; “Murphy v. National Collegiate Athletic Association”). Taking the case signaled that the question had changed. It was no longer whether New Jersey had violated PASPA. It was whether PASPA could survive the Tenth Amendment.

On May 14, 2018, the dam broke. Writing for the Court, Justice Samuel Alito struck PASPA down, and the opinion is worth reading for its bluntness. The provision at issue “unequivocally dictates what a state legislature may and may not do,” Alito wrote, and “a more direct affront to state sovereignty is not easy to imagine” (Murphy v. National Collegiate Athletic Association; “Supreme Court Decides Murphy”). The vote is usually reported as seven to two, and on the commandeering question it was: seven justices agreed that Congress could not forbid a state from authorizing sports wagering, with Justice Ginsburg, joined by Justice Sotomayor, disagreeing. The Court divided differently on what should happen to the rest of the statute. There the split was six to three, with Justice Breyer joining the dissenters to argue that the remaining provisions could stand on their own (Murphy v. National Collegiate Athletic Association; Sykes, “The Supreme Court Bets”; Somin).

The Court also demolished the distinction on which the leagues had built their case. They had argued that commandeering occurs only when Congress affirmatively commands a state to act, not when it merely forbids state action. “This distinction is empty,” Alito answered, because “in either event, state legislatures are put under the direct control of Congress” (Murphy v. National Collegiate Athletic Association; Somin). Having found the core provision unconstitutional, the Court then declined to salvage the remainder, holding that no provision of PASPA was severable from the one at issue. The entire twenty-six-year-old act fell at once, and with it the federal barrier in front of every state in the union (Murphy v. National Collegiate Athletic Association; “Supreme Court Decides Murphy”).

The decision’s true legacy runs well past gambling. Murphy is among the most consequential affirmations of the anti-commandeering doctrine in modern American law, and the principle at its heart, that Washington cannot conscript state governments to carry out federal policy, is indifferent to subject matter. Commentators reached immediately for the parallels: federal efforts to stop states from adopting sanctuary policies on immigration, and the long-running friction between federal marijuana prohibition and state legalization (Boswell; Somin). A decade-long fight over point spreads had turned out to be the vehicle for a landmark ruling on federalism. The dam was never only holding back sportsbooks. It was holding back a broader conception of state power, and when it went, the water ran in several directions at once.

The American Gold Rush

What followed was a scramble for territory and treasure on three fronts at the same time: in state legislatures, on corporate balance sheets, and inside the headquarters of the leagues that had spent twenty-five years insisting none of this should ever happen.

The State Stampede

Freed from the federal constraint, state governments saw a new and lucrative line of tax revenue (Meer; Ruedas). Several had seen it coming. Pennsylvania and others had already passed trigger laws designed to legalize wagering automatically the moment PASPA fell (Ruedas; “Anyone’s Game”). The race was on before the ink on the opinion was dry.

Delaware, leaning on its grandfathered lottery scheme, was first out of the gate, taking full single-game wagers on June 5, 2018, three weeks after the ruling, at Dover Downs, Delaware Park, and Harrington Raceway (Ramsey, “Welcome To Delaware Sports Betting”). New Jersey, the state that had spent a decade making the day possible, followed nine days later, with the first tickets written at William Hill’s book at Monmouth Park and at the Borgata in Atlantic City (“Day One Of Legal NJ Sports Betting”). Then came the dominoes. Within a year, seven states had opened books, and the pattern of those first openings says something the later boom obscures: online betting, which now accounts for the overwhelming majority of American wagering, barely existed at the start. Six of the seven launched with retail counters only.

Table 1.3.1The First Wave: Every State to Take a Legal Sports Bet in the Year After Murphy

State	Date of First Legal Bet	First Books to Take Wagers
Delaware	June 5, 2018	Delaware Lottery, at Dover Downs, Delaware Park and Harrington Raceway (retail only)
New Jersey	June 14, 2018	William Hill at Monmouth Park; Borgata, Atlantic City (retail only)
Mississippi	August 1, 2018	MGM Resorts, at Beau Rivage and Gold Strike (retail only)
West Virginia	August 30, 2018	William Hill at Hollywood Casino at Charles Town Races (retail only)
New Mexico	October 16, 2018	Santa Ana Star Casino, Pueblo of Santa Ana, under tribal compact (retail only)
Pennsylvania	November 17, 2018	William Hill at Hollywood Casino at Penn National (retail only)
Rhode Island	November 26, 2018	Rhode Island Lottery, operated by IGT and William Hill at Twin River (retail only)

Source: Compiled from launch-day reporting (Ramsey, “Welcome To Delaware Sports Betting”; “Day One Of Legal NJ Sports Betting”; “MGM Resorts First to Market”; Ramsey, “Hollywood Casino Launches West Virginia Sports Betting”; Ramsey, “Welcome State No. 6”; “William Hill Takes First Legal Bets in Pennsylvania”; “Rhode Island Becomes 8th State”).

From there the curve steepened into something without a real precedent in American consumer finance. Through June 2026, Americans had legally wagered roughly $699 billion since the Murphy decision, producing about $61.6 billion in operator revenue and $12.8 billion in state taxes (Altruda). The single year 2025 accounted for $166.9 billion of that handle and $16.9 billion of the revenue in the commercial market alone, which also generated $3.7 billion in state gaming taxes (“Commercial Gaming Revenue Hits $78.7 Billion”; State of the States 2026). Thirty-nine states and the District of Columbia now permit sports betting in some form, with roughly thirty-two of them plus the District offering it online. Missouri, whose voters approved wagering in 2024, launched on December 1, 2025 and became the thirty-ninth state (“Missouri Legalizes Sports Gambling”). Wisconsin followed in April 2026, when Governor Tony Evers signed a bill authorizing statewide mobile wagering through tribal compacts that have yet to be renegotiated (Phillimore). Eleven states still hold out, among them the two largest prizes in the country, California and Texas.

There is a quieter figure inside those numbers that matters more than the handle. In 2024, sportsbooks kept about 9.2 percent of every dollar wagered. In 2025 they kept roughly 10.2 percent, and revenue grew at nearly twice the rate of handle (“Commercial Gaming Revenue Hits $78.7 Billion”; Altruda). Americans did not simply bet more. They bet worse, shifting toward the multi-leg parlays and same-game combinations on which the house edge is several times what it is on a straight side. The state stampede is usually told as a story about access. It is at least as much a story about the composition of the bets, and that is a subject this series will return to repeatedly.

The Corporate Land Grab

While the states wrote laws, the operators wrote checks. Two companies began the race with an advantage nobody else could buy. The daily fantasy sports operators DraftKings and FanDuel already had national brands and, more valuably, databases holding millions of American sports fans who had been giving them money for a quasi-gambling product for years (“Online Gaming Companies”; “Huddle Up”; Bussmann). On the morning of May 15, 2018, they were not starting from zero. They were starting from a mailing list.

What followed was one of the most aggressive customer acquisition campaigns in American corporate history. DraftKings and FanDuel poured hundreds of millions, and before long billions, into television advertising, celebrity endorsement, podcast sponsorship, and the risk-free bonus offers whose mechanics a later article in this series takes apart in detail (Moore; “Normalization of Sports Betting”; “Sportsbooks’ TV Ad Spend”). Profitability was not the objective. Market share was, on the theory that the first company to acquire a bettor in a newly legal state would hold that bettor for years (“Huddle Up”).

The theory held. By April 2026, DraftKings and FanDuel between them took roughly 67 percent of all money wagered nationally and about 71 percent of the revenue, with DraftKings and FanDuel each holding more than three times the share of their nearest competitor (Smiley). The gap between those two percentages is itself informative: the duopoly earns a disproportionate share of the revenue relative to the volume it handles, which is what a superior product mix and a stickier customer base look like in a spreadsheet.

For everyone else the market has been a graveyard. By the end of 2024, more than twenty companies had shut down at least some of their American digital sportsbooks, a list that includes Betway, WynnBet, Fubo Sportsbook, TwinSpires, SuperBook, MaximBet, and Unibet, alongside earlier casualties such as FOX Bet and Barstool Sportsbook (Dorson). The retreat did not stop there. Betfred closed its last American online operation in July 2025 (“Betfred Ends US Online Operations”), and in November 2025 Penn Entertainment and ESPN mutually terminated ESPN Bet barely two years into a ten-year, $1.5 billion agreement, retiring the brand that December (Douglas). A market that looked in 2019 like an open frontier turned out to have room for two national winners, a handful of survivors, and a long column of write-offs.

The Unholy Alliance

The most complete reversal in the gold rush belonged to the leagues. Having spent decades and tens of millions of dollars in legal fees defending the ban on the ground of integrity, they pivoted almost overnight (“A Look at the Growing Ties”; “What Was PASPA”). The same organizations that had sued New Jersey to preserve PASPA became commercial partners of the companies they had fought. The NFL, NBA, MLB, and NHL signed multi-year deals naming DraftKings, FanDuel, BetMGM and others as official sports betting partners, agreements worth billions that licensed official league data and team marks for use inside betting apps and wove betting content directly into game broadcasts (“Sports Betting Partnership Tracker”; “If You Can’t Beat ’Em, Join ’Em”; “A Look at the Growing Ties”).

The pivot exposed the real content of the leagues’ long opposition. It was never a moral position; it was a property position. So long as betting happened outside their control and paid them nothing, it was a threat to the integrity of the game. Once they could take a share, first through proposed integrity fees and then through sponsorship, the objection dissolved (“What Was PASPA”; “Anyone’s Game”).

Eight years on, that alliance is proving less permanent than it looked. In November 2025, ESPN made DraftKings its exclusive sportsbook and odds provider after unwinding its arrangement with Penn (Douglas). In April 2026, the NFL’s deals with FanDuel, DraftKings, and Caesars all expired without renewal, leaving the most valuable league in American sports without an official sportsbook partner for the first time since 2021, reportedly over the price of the official data feed the sponsorship requires (“NFL Reportedly Without an Official Sportsbook Partner”). The partnerships were never marriages. They were licensing agreements, and licensing agreements are repriced.

The three gold rushes did not run in parallel for long. They converged into a single self-reinforcing system. States needed operators to book the action and remit the tax. Operators needed league partnerships for legitimacy, for official data, and for access to fans. The leagues justified the arrangement by pointing to engagement, and in one student survey fans said they would engage more with the games if betting were legal (Labuda et al.; “How Betting Odds Drive Fan Engagement”; “A Look at the Growing Ties”). An industrial complex with deep financial and political roots was assembled in under five years, and every participant in it had a reason to want the volume to go up.

Conclusion: A New Frontier

Breaking the PASPA dam did more than legalize an activity. It released a torrent of capital, technology, and marketing that has re-engineered the American relationship with sport. In a handful of years, betting moved from a vice associated with back rooms and Las Vegas to a mainstream entertainment product stitched into the fabric of fandom. Its vocabulary, the spreads and the parlays and the player props, is now the ordinary language of sports radio, pregame television, and the group chat.

There is one more thing the dam metaphor gets right. Water does not stop at the breach. It finds the next low ground. Murphy handed the states the power to authorize sports wagering, and within eight years a competing market appeared that does not ask the states for anything at all. On federally regulated exchanges, a bet on a football game is written as an event contract, a derivative traded under the Commodity Exchange Act and supervised by the Commodity Futures Trading Commission rather than by any state gaming board. In April 2026, a divided panel of the Third Circuit held that such contracts fall within the commission’s exclusive jurisdiction and preempt New Jersey’s gambling law, the same New Jersey whose sovereignty Murphy had vindicated eight years earlier (KalshiEX, LLC v. Flaherty; “A Divided Third Circuit”). By then the federal regulator was suing states, the Ninth Circuit was weighing the opposite conclusion, and DraftKings and FanDuel had each launched prediction market products of their own within three days of one another in December 2025 (“DraftKings Debuts Predictions App”; “FanDuel and CME Group Launch FanDuel Predicts”). Reported trading volume on prediction markets passed the total handle of American sportsbooks during the 2026 World Cup, although as the next article will show, volume and handle are not the same unit and the comparison flatters the newcomers considerably (Radde; “H2GC: Prediction Markets”).

That is the subject of article 1.4, The Exchange Loophole. For now it is enough to notice the shape of the thing. The legal and financial structures erected during the gold rush of 2018 built the House of Mirrors that modern bettors have to navigate, and the industry’s response to every new constraint has been to find the room next door. The articles that follow examine what was built inside that house: the psychological machinery, the media apparatus, the information economy, and the question this series exists to answer, which is where in all of it a real edge actually lives.

Sources

This is 1.3 of House of Mirrors: A Bettor’s Guide to the Illusions of Modern Sports Gambling, a serialized investigation of the American sports betting industry. Next up is article 1.4, The Exchange Loophole.