When the Trade Looks Like a Bet and the Bet Looks Like a Trade

The Massachusetts Securities Division objected to a feature of a brokerage app that let customers move up a waitlist by tapping a digital debit card, up to 1,000 times a day. When they ran out, the app told them, in the language quoted in a law firm’s account of the case, that they were “out of taps today” and should “Come back tomorrow if you’re feeling tappy.” The state called this gamification. In January 2024 the firm, Robinhood Financial, resolved the allegations with a $7.5 million fine and an agreement to overhaul its digital engagement practices.

The tapping card is a small thing. It is useful because it is so legible. Nobody mistakes it for a financial instrument, and nobody mistakes it for a slot machine either. It sits in the space between, which is where a growing share of young Americans spend their money, and where the surveys, read together, show that the usual vocabulary has stopped working.

What the surveys count

Start with the cleanest number. Pew Research Center asked 9,916 U.S. adults between July 8 and August 3, 2025, whether they had bet on sports in the past year. Twenty-two percent had, up from 19% in 2022. Among adults aged 18 to 29, the figure was 31%; among men in that age group, 36%.

The more revealing detail is where the increase came from. Betting with friends or family was flat at 15%. Betting in person at a casino, racetrack or kiosk was flat at 8%. The whole rise came from betting online, through an app, a sportsbook or an online casino, which went from 6% of adults to 10%. For men under 30 it went from 9% to 21%.

That is a story about a channel, not about a generation’s character. A wager that used to require a bookmaker, a trip or a friend with a spreadsheet now requires a phone. Pew’s data, collected on its American Trends Panel, cannot say why, but a qualitative study released this summer offers a hint. Researchers at NORC at the University of Chicago, funded by the Massachusetts Gaming Commission, interviewed 62 Massachusetts residents aged 18 to 25. According to GBH’s account, participants described mobile betting as “frictionless and always available in the palms of their hand,” and most recognized the risks “in theory,” in the words of NORC’s Katie Gallant, “but many of them viewed themselves and their immediate friends as relatively immune.” Most bets among active bettors were under $10.

A sample of 62 proves nothing about a country. It does describe a texture: small stakes, constant access, and a confidence that the harm is something that happens to other people.

The line moves up the stack

The same convenience that moved sports betting onto phones has moved other things next to it. The FINRA Foundation published results in December 2025 from the 2024 National Financial Capability Study investor survey, which covered 2,861 respondents with non-retirement investment accounts. Investors under 35 were far likelier than investors 55 and older to say they traded options (43% against 10%) or bought on margin (22% against 4%).

Options are legal, regulated and, for professionals, a tool for managing risk. They are also, in their shortest form, a contract that expires the day it is bought. Cboe reported that so-called zero-days-to-expiration options reached a record 62.4% of S&P 500 index option volume in August 2025, roughly 2.4 million contracts a day, with retail traders making up an estimated 53% of the volume. Cboe is an exchange and has an interest in volume; the figure is an estimate, not an audit. It still describes a market in which a large share of retail activity is a bet on what will happen before the close.

Then there are the contracts that drop the pretence of an underlying asset. In an April 2026 release of the American Sports Fanship Survey, from Siena Research Institute and St. Bonaventure University’s Jandoli School of Communication, 15% of Americans said they had bet on sports through prediction platforms such as Polymarket, Kalshi or Robinhood. Among men aged 18 to 49 the figure was 33%. (The report we read for this piece, from the trade site Covers, does not give the sample size or field dates, so treat the numbers as indicative.)

What do respondents call this? In a Harris Poll survey for the National Council on Problem Gambling, fielded February 19 to 23, 2026, among 2,072 U.S. adults, 30% said prediction markets were most similar to gambling, 24% said financial forecasting, and 18% said investing. No category won a majority. Respondents split three ways.

Where it hurts

If the distinction were only semantic, it would be a matter for editors and lawyers. The research suggests it is not.

An NBER working paper, first issued in November 2024 by Scott Baker, Justin Balthrop, Mark Johnson, Jason Kotter and Kevin Pisciotta, used household transaction data to study what happened after states legalized online sports betting. It found that savings fall significantly while other gambling and consumption do not change, because risky bets “crowd out positive expected value investments.” The effects concentrate in financially constrained households, which show more credit card debt, less available credit and more frequent overdrafts. The paper is a study of households, not of Gen Z. It is nonetheless direct evidence for a claim that sounds like rhetoric: the money a young person puts into a sportsbook is, in part, money that was not going into an index fund.

The harm shows up in clinical data as well. A Fairleigh Dickinson University poll of 801 registered voters, conducted August 17 to 20, 2024, used the Problem Gambling Severity Index and found that 10% of men aged 18 to 30 scored at a problem level, against 3% of all voters. Of online sports bettors, 68% reported at least one problem behavior. The poll’s executive director, Dan Cassino, said: “Online sports betting and slots are creating a generation of problem gamblers.” That is a strong sentence from a pollster, not a finding. The subsample of young men in an 801-person poll is small, and the margin of error on the full sample is 3.5 points.

The helpline numbers are not a sample at all, which is both their strength and their weakness. The National Council on Problem Gambling’s annual report, published in May 2026, said that more than 31,000 people a month contacted the national helpline in 2025. Men were 70% of contacts. Nearly half, 49.48%, were between 18 and 34. Concerns about online and app-based gambling rose to 31% of contacts, from 23% the year before, while concerns about slot machines and electronic gaming fell from 36% to 31%. A helpline measures people who call it; it cannot say how many do not.

The skeptics have a case

Three objections deserve a hearing.

The first is that the line was never clean. The NCPG’s February 2026 poll found that 65% of adults 21 and older had gambled before turning 21: 40% on lotteries or scratch-offs, 37% in home games for money. Only 23% had placed a sports bet. Gambling by the young is not new. What the same survey adds is that sports betting before 21 was reported by 33% of respondents aged 21 to 44, against 11% of those 55 and older. The old generation had the lottery and the poker night. The new one has a phone. That may be a difference of degree; whether a difference of degree matters is the argument, not its refutation.

The second is that participation in markets is a good thing and the evidence of excess is selective. FINRA’s own data point in this direction. The share of investors who had begun investing within the previous two years fell to 8% in the 2024 survey, from 21% in the 2021 survey, and the share holding crypto, 27%, was unchanged from 2021. The rush of new entrants has slowed. Anyone claiming that a generation is being herded into speculation has to explain why the number of new entrants collapsed. A defender of retail trading would add that access to options and fractional shares, however frothy, is a large improvement over a world in which only wealthy clients could trade at all. That argument rests on principle rather than on the surveys above, but it is a serious one and the data do not refute it.

The third is that self-reports are noisy. People misremember bets, undercount losses, and answer a pollster’s category differently depending on how the question is framed. Pew’s finding that 47% of men under 30 now call legal sports betting bad for society, up from 22% in 2022, can be read as growing concern or as a question wording effect on a small subgroup. The NORC study has 62 people. The FDU poll surveyed registered voters, not all young people. The Siena figure on prediction markets comes to us through a trade publication. Each of these is a reason not to quote one number as a verdict.

The turn

Pew’s finding on attitudes is where the skeptics’ case meets its problem. Among men under 30, 47% said legal sports betting is a bad thing for society, up from 22% in 2022. In the same survey, 36% of men aged 18 to 29 had bet on sports in the past year. Many young men are, in effect, telling a pollster that they dislike a thing a large minority of them do.

That is a position with a long history, and it is not hypocrisy. People smoked when they thought smoking was a bad idea. But it complicates the usual stories. The young are not being tricked into believing betting is harmless; the NCPG found that Americans, as a whole, cannot decide what to call it. Nor are they simply reckless; in the Massachusetts study, most wagers among active bettors were under $10. The more precise reading is that a generation is using, daily and with open eyes, products built so that the difference between investing, trading and betting is a design choice rather than a fact about the world. Digital confetti after a customer’s first trade was among the practices the Massachusetts regulator cited.

Whoever draws the line next, whether a legislator, an exchange or a user, will be drawing it in a place where the data say it has already moved. The question the surveys cannot answer is the one the tapping card put plainly: out of taps today, come back tomorrow, if you are feeling tappy.

Not financial advice.