Is Day Trading Gambling? What the Law and the Data Say

Futures trading and gambling are not the same category in US law: one of the four elements in the CFTC's own definition of a futures contract is that it is used to assume or shift price risk, and the Commodity Exchange Act's "gaming" category belongs to a listing review written for event contracts rather than for price-based futures. Whether it is gambling for the person doing it is a separate question, and the published research on individual day traders answers it far less comfortably. In the two large trader-level studies below — one of Brazilian index futures, one of Taiwanese stocks — most participants lost money, and in the Brazilian data the share who profited fell the longer they persisted.

Those two answers do not contradict each other. One describes what the market is; the other describes what happened to the people speculating in it.

What separates a futures trade from a bet?

A futures contract is defined by four things at once, and price risk is one of them. Take the CFTC glossary's entry apart and each element is doing work: the contract is a vehicle for taking on price risk or handing it off; both sides are bound to perform at the agreed price; that price is fixed at the moment the contract is entered rather than when it comes due; and a position leaves the book either by being offset or by the commodity changing hands. None of that describes a stake on a roulette wheel, where the risk exists only because the bet was placed.

Congress framed the market the same way. The findings at 7 U.S.C. §5 call transactions under the chapter "affected with a national public interest", on the stated ground that they provide a means of managing and assuming price risks, discovering prices, or disseminating pricing information through liquid, fair and financially secure trading facilities. The CFTC glossary supplies the two roles that make that work: a hedger takes a futures position opposite a cash-market position to limit the damage from an adverse price move, and a speculator does not hedge at all, but trades to profit from correctly anticipating price movements.

That is the structural answer, and it is where the uncomfortable part starts. Open interest, as the CFTC glossary uses the term, is the running count of contracts still outstanding — opened, and neither offset nor delivered. Every one of them is long for somebody and short for somebody else. A price move that credits one side debits the other, and fees come out of the total. Not a casino, then, but between speculators not a machine for producing gains either.

Does US law treat futures trading as gambling?

No — it regulates exchange-traded futures under the Commodity Exchange Act, through a commodities regulator rather than a gaming one. Under 7 U.S.C. §2, the CFTC has exclusive jurisdiction over accounts, agreements and transactions involving swaps or contracts of sale of a commodity for future delivery that are traded or executed on a contract market designated under section 7, a swap execution facility, or any other board of trade, exchange or market. That jurisdiction is subject to the exceptions the statute lists, among them those made by the Wall Street Transparency and Accountability Act of 2010. The uscode.house.gov page carrying that text states it contains laws in effect on 9 September 2026.

The word "gaming" does appear in the Act, and one place it appears is instructive. The listing-review provision at 7 U.S.C. §7a-2 covers event contracts: agreements in excluded commodities, listed by a designated contract market or swap execution facility, that turn on the occurrence, the extent of an occurrence, or a contingency — other than a change in the price, rate, value or levels of a commodity. The Commission may find such a contract contrary to the public interest if it involves activity unlawful under federal or state law, terrorism, assassination, war, gaming, or a similar activity it determines by rule, and anything so determined may not be listed, cleared or traded on a registered entity.

Read the scope carefully, because it is easy to stretch. The provision reaches contracts written on whether something happens; a contract whose payoff tracks a price rather than an occurrence sits outside it by the clause's own wording. That is a statement about how each category is reviewed before listing, and nothing about how trading either one will go for you.

Do most day traders lose money?

Most did, in both of the large trader-level datasets below. The evidence closest to futures is a 2019 University of São Paulo working paper by Chague, De-Losso and Giovannetti — not peer reviewed — built on daily records held by Brazil's securities regulator. Its sample is a complete cohort rather than a survey: all 19,646 individuals whose first day trade in mini-Ibovespa index futures fell between 2013 and 2015. Profit there means what was left after exchange and brokerage fees. Income tax, platform costs and the price of trading courses are not deducted, so the figures below flatter the outcome rather than the reverse.

Group that cohort by how long each person kept at it, and one column does all the talking:

Read it from the bottom up: the more days a trader put in, the smaller the fraction of that group came out ahead. Chague and his co-authors compare the shape to a roulette table, where the share of players still up shrinks as the rounds accumulate, rather than to an activity people get better at. The 3.0% at the top of the table is also a thinner achievement than it sounds. Among the 1,551 who traded on more than 300 days, 47 finished in profit at all — that is the 3.0% row; 17 of them, or 1.1% of the group, cleared the Brazilian minimum wage, a benchmark of about US$16 a day, and eight, or 0.5%, cleared a bank teller's starting salary, about US$54 a day. Nor were those eight collecting anything resembling a salary: their daily profits carried standard deviations ranging from US$632 to US$3,308. The best individual among the 1,551 averaged US$310 a day, on a standard deviation of US$2,560.

The equity evidence points the same way. Barber, Lee, Liu, Odean and Zhang worked from every transaction on the Taiwan Stock Exchange between 1992 and 2006, counting as a day trader anyone who bought and sold the same stock inside one session. Year by year, the group's returns come out reliably negative — before fees and after them — in all fifteen years bar 1992, the first in the sample. At the level of the individual the profitable share is small; the paper's own characterisation is that the great bulk of the people doing this did not make money.

Does day trading get easier with experience?

In the Brazilian futures cohort it did not. The table above already cuts against what learning would predict, and against self-selection too: Chague and his co-authors point out that persistence normally sorts for ability, since whoever sticks with something tends to be among those doing it well, and yet the profitable share here falls at every step. To test learning directly rather than infer it from that shape, they moved the question inside the individual. Taking only the 1,551 traders who lasted beyond 300 days, they modelled a single day's profit against a counter of how many days that same trader had already traded, with a fixed effect for each trader so that the comparison is a person measured against their own earlier self. Experience buys nothing in that model: the profit line does not tilt upward as the day count rises, and it stays below zero across the range. The obvious objection is that the opposition was getting harder — high-frequency traders accounted for 11.6% of all deals closed in 2012 and 42.0% by 2016 — which could be masking genuine improvement. Adding a control for that presence leaves the answer where it was.

Do losing day traders quit?

Most quit: over three quarters of the day traders in the Taiwan Stock Exchange records had stopped inside two years. Whether losing was what drove them out is a separate question, and the same data answers it in reverse, not whether losers improve but whether they stop. Take traders with at least 50 prior day-trading days and split them by whether they had been making money. The losing group traded again within the following twelve months 95.3% of the time; the winning group, 96.4%. A record of losses moved the odds of coming back by about a percentage point. The volume figures say the same thing from another angle: roughly three quarters of day-trading volume came not from newcomers but from people who had ten days or more behind them and were losing. Attrition is real, then, but among the experienced it tracks results only faintly.

None of this proves improvement is impossible; no dataset can prove that. What these two studies say is narrower and still uncomfortable. The direct test of learning in the Brazilian panel found none, and in the Taiwanese data a losing record barely changed whether people came back for more.

Is there any evidence that day trading is a skill?

Yes, in a narrow and demanding sense. Barber, Lee, Liu and Odean ran the standard test for skill in the Journal of Financial Markets in 2014: rank Taiwanese day traders on one year's returns, then set that year aside and look only at what the same people did in the next one. If day-trading results are noise, last year's ranking should say nothing about next year's. It said a good deal.

A gap that size, carrying over from one year into the next, is not the behaviour of a game of chance. The same authors are just as firm about how few people sit on the right side of it: Barber and his co-authors write that "less than 1% of the day trader population is able to predictably and reliably earn positive abnormal returns net of fees." The lower half of the table makes a second point worth pausing on. The bottom group was already down 11.5 basis points a day before a single fee was deducted, so costs are not the only thing standing between them and a profit.

The honest answer therefore has two halves, and both matter. Skill of a measurable, repeating kind does exist in this population. It is also confined to a sliver of it — under 1%, by that paper's own count — and a person with no track record yet has no claim on that sliver. What the studies above describe is what happened to the populations who started, which is why our risk disclosure says most people who attempt this will fail.

When does trading become gambling for the person doing it?

Frequent trading and problem gambling do show up together in the research, though the study behind that is about stock investors rather than futures. Mosenhauer, Newall and Walasek published a preregistered, retrospective cross-sectional study in the Journal of Behavioral Addictions in 2021, surveying 795 US participants who reported both active gambling and stock market holdings. The more of their portfolio those participants said they turned over, the higher they scored on the Problem Gambling Severity Index, and that pattern survived controls for financial literacy, overconfidence, age and gender and held at every portfolio size they reported.

The limits matter as much as the finding, and the authors state most of them: a convenience sample recruited from a crowdsourcing platform rather than a cross-section of US investors; everyone in it reporting a gambling history by design; portfolio value and trading recollected rather than observed; and other explanatory variables left unmeasured. The design is cross-sectional, so what it reports is an association and not a direction. It also measures stock turnover, not futures day trading.

The association is worth knowing anyway, because behavioural markers are easier to check on yourself than a motive is. Buying another account within a day of losing one, sizing up after a loss instead of after a win, hiding what the habit costs from someone close to you, trading longer sessions than planned most weeks — our responsible trading page lists seven such questions and the free, confidential helplines to call if the answers are uncomfortable. A trading firm is not the right body to help with that, which is why that page points elsewhere.

Does a funded account change any of this?

It changes the downside, not the odds. Every Lumen Futures account, evaluation and funded, is a simulated account trading live market data, so the fee is the amount at stake — no margin call, and no way to owe money on a breach, unlike trading your own capital in a live market, where losses are not necessarily limited to the account balance. No simulated environment makes a losing method profitable.

Two of our rules are worth reading in that light. The maximum drawdown is the single performance rule that ends an account, so a bad run stops at a defined point rather than continuing. Against that, none of the programs we currently sell have a daily loss limit — a deliberate trade-off, explained in why there is no daily loss limit — which means nothing external ends a session that is going badly. On the day that matters, the stop is yours.

The evidence above also says something about evaluations generally, ours included: passing one records what happened once. The Brazilian futures panel found no learning effect, and the Taiwanese cross-section put reliable net-of-fee outperformance at under 1% of the day traders studied. Neither figure changes with where an account is held or how its fee is structured.

FAQ

What percentage of day traders are profitable? Low, and lower the longer people stay. Brazil's securities regulator recorded 19,646 individuals taking their first day trade in mini-Ibovespa index futures between 2013 and 2015; after exchange and brokerage fees, 29.8% of those who traded a single day were ahead, against 3.0% of the 1,551 who traded on more than 300 days. Taiwanese exchange records from 1992 to 2006 give another cut of the same picture: taken as a group, day traders there posted reliably negative returns before and after fees in every year of that sample bar the first.

Is day trading futures regulated as gambling in the United States? No. Under 7 U.S.C. §2 the CFTC has exclusive jurisdiction, subject to the exceptions the statute lists, over transactions in contracts of sale of a commodity for future delivery traded or executed on a designated contract market. The Commodity Exchange Act's "gaming" category sits in the listing-review provision at §7a-2, which reaches event contracts based on an occurrence, extent of an occurrence or contingency — not contracts based on a change in price.

Is trading a prop firm evaluation gambling? That turns less on the fee structure than on the trading and on where the fee came from. Here the fee is one-time and buys a simulated account with published rules, and the account's outcome is decided by trades made against those rules — the drawdown, the contract limits, and the profit target where a program has one. None of the research above stops applying because an account is simulated; it describes how day trading went for the populations studied. And if the fee is money that has a job to do elsewhere, the responsible trading page treats that as the signal to stop.

Is frequent trading linked to problem gambling? In one survey of people who already gamble, yes. Mosenhauer, Newall and Walasek's 2021 study in the Journal of Behavioral Addictions surveyed 795 US participants who reported both active gambling and stock holdings, and found that the more of their portfolio people said they turned over, the higher they scored on the Problem Gambling Severity Index, with controls for financial literacy, overconfidence, age and gender. It is a cross-sectional convenience sample of stock investors, so it shows an association rather than a direction, and it does not measure futures day trading.