No State Has a Good Legal Definition of Gambling. The Texas AG Just Made That Clear.
Texas says any element of chance makes it gambling. Pennsylvania says skill can save it. Neither test actually works nor is a perfect definition.
Texas Attorney General Ken Paxton issued a formal opinion this week concluding that skill-based gaming machines are illegal under Texas law if chance plays any role in determining whether a player receives something of value. The opinion addressed machines resembling slot games that allow players to recover losses through a Simon Says-style memory game, and concluded that the presence of a skill component does not exclude the machines from Texas gambling laws. “So long as chance plays any role in determining whether the player obtains something of value, the device satisfies the statutory definition,” Paxton wrote.
The opinion is not binding on Texas courts and does not change existing law. It is, however, a clean statement of the strictest approach to gambling definition in American jurisprudence, what legal scholars call the “any chance” test. And read alongside the broader landscape of state gambling law, it illustrates a problem that extends well beyond Texas skill machines: no state has developed a legal definition of gambling that coherently captures what people actually mean when they use the word.
States Have Opposite Defintions of What Gambling is, and Both are Legally Defensible
The United States does not have a uniform legal standard for what constitutes gambling. States apply one of three primary tests. The predominant factor test holds that if skill dominates the contest, the activity is legal even if chance is present. The material element test asks whether chance is present to any material degree, a lower threshold than predominance but stricter than the any-chance test. And the any chance test, the strictest of the three, holds that any element of chance at all is sufficient to classify an activity as gambling.
Texas applies the any chance test. The result, as Paxton’s opinion illustrates, is that virtually any competitive activity with a monetary component could theoretically qualify as gambling, since almost every game involves some element of chance, even if vanishingly small.
Pennsylvania applies the predominant factor test and has reached precisely the opposite conclusion on nearly identical facts. The Pennsylvania Commonwealth Court unanimously ruled that Pace-O-Matic skill game machines were legal because skill predominantly determined the outcome, even though a chance-based component was present. The machines at issue in that case had meaningful structural similarities to those Paxton addressed this week. The legal outcomes could not be more different.
Neither state is obviously wrong given its own statutory framework, and in there lies the problem. Two states using two different tests produce two opposite results on what constitutes gambling, and both results are legally defensible. The test chosen, not the underlying activity, determines legality.
Which Activities Fall Under Which Tests to Determine What Gambling Is?
The larger issue at hand is that neither extreme maps cleanly onto the intuitions that gambling law is supposed to protect.
The any chance test captures too much. Under a strict application, a golf tournament with prize money is technically gambling. A poker league with entry fees is gambling. A stock options trade is also a form of gambling. And a prediction market contract is a form of gambling under the any chance test as well. Most people’s intuitive sense of what deserves the social and legal stigma associated with gambling does not extend to golf tournaments, even if the winner’s margin of victory depends partly on a favorable bounce on the 14th hole. Is there chance involved on which way that ball bounces? Yes. Does anyone believe that makes the golf tournament a form of gambling? Not really.
The predominant factor test captures too little. Under that standard, a slot machine with a rudimentary skill overlay, memorize this pattern and press this button, can potentially escape the definition of gambling entirely. Pennsylvania’s court found that a machine designed primarily to extract money from players was legal because a skilled user could memorize patterns to recoup losses. That outcome strikes most observers as a regulatory evasion rather than a principled distinction.
The broader problem is that the skill-chance binary was designed to address a relatively simple landscape: card games, dice games, lotteries. It was never equipped for the modern gambling environment, where the same underlying activity, wagering on a sports outcome, can be presented as a prediction market contract (federally regulated financial instrument), a daily fantasy sports contest (game of skill under most state tests), or a traditional sports bet (gambling, regulated state by state). The activity is functionally the same. The legal category depends almost entirely on how the product is framed and which test the relevant jurisdiction applies.
The Prediction Market Problem Is the Same Problem
This definitional fragility is not academic. It is directly relevant to the regulatory fights that have consumed state attorneys general, the CFTC, and federal courts for the past two years.
Attorney general opinions do not change the law and are not binding on Texas courts, but they serve as advisory interpretations intended to guide enforcement agencies and lawmakers. Paxton’s any chance standard, applied consistently, would make prediction markets illegal in Texas. Sports event contracts settle based on outcomes that are meaningfully influenced by chance, regardless of how much research and analysis a trader brings to the position. Under the strictest reading of Texas law, that would be considered gambling.
And yet Paxton’s office has conspicuously declined to take action against prediction markets, did not respond when asked to join multistate amicus briefs against them, and has not explained its reasoning publicly. The office that just issued an opinion saying any chance makes it gambling has quietly declined to apply that standard to the fastest-growing wagering product in the country.
Pennsylvania, applying its more permissive predominant factor test, has taken the opposite institutional posture: the Pennsylvania Gaming Control Board has expressed profound opposition to prediction markets, with Executive Director Kevin O’Toole arguing that the CFTC has abandoned its historical mandate by allowing sports event contracts to proliferate. The state with the more lenient definition of gambling is the one fighting hardest to restrict the new product.
That inversion is not a coincidence. It reflects how thoroughly the skill-chance framework has decoupled from the underlying policy questions. The test a state applies tells you very little about that state’s actual regulatory appetite. What matters is institutional incentives, existing industry relationships, and political will, not whether the jurisdiction uses the predominant factor test or the any-chance test.
The Legal Definition of Gambling Doesn’t Appear Any Closer to an Agreement Between States
Legal scholars have proposed alternatives. One approach, sometimes called the gambling instinct test, asks not whether chance is present or dominant but whether the activity appeals to the gambling instinct, defined as the desire to obtain something of value through risk rather than through productive activity. The modern chance-based test for gambling is fundamentally flawed, descriptively inaccurate, difficult to apply, and easily circumvented, and the gambling instinct test offers a competing framework grounded in the underlying psychological phenomenon rather than the mechanical form of the activity.
That approach has its own problems, not least that “gambling instinct” is not a legally precise standard and would invite significant litigation over what it means in any given case. But it at least points toward the right question. The issue is not whether a game contains any chance, or whether chance predominates. The issue is whether an activity is structured to extract value from participants through risk-taking that resembles wagering, regardless of its label.
No state has fully operationalized that standard. Until one does, the legal landscape will continue producing outcomes like this week’s: Texas issuing an opinion that logically applies to prediction markets while declining to enforce it against them, and Pennsylvania fighting prediction markets under a legal standard permissive enough to legalize slot-adjacent machines. The definitions are clearly not working. The Paxton opinion is a useful reminder of how long that has been true.
Players trust our reporting due to our commitment to unbiased and professional evaluations of the iGaming sector. We track hundreds of platforms and industry updates daily to ensure our news feed and leaderboards reflect the most recent market shifts. With nearly two decades of experience within iGaming, our team provides a wealth of expert knowledge. This long-standing expertise enables us to deliver thorough, reliable news and guidance to our readers.
Latest News
