Very few. In the largest study with complete data, 97% of individuals who day traded Brazilian equity index futures for at least 300 days lost money, and only 0.4% earned more than a bank teller. A separate study of Taiwanese day traders found under 1% could reliably earn positive returns net of fees. Those are the real numbers.
They are worse than the figure most people repeat, and they come from actual regulator and exchange records rather than surveys or broker marketing. Before looking at what they mean, it is worth retiring the number they replace.
The "90% of traders lose money" claim has no source
You will see it everywhere, usually as 90/90/90 — 90% of traders lose 90% of their capital within 90 days. It is quoted by brokers, course sellers and journalists alike, and nobody can point to the study. There isn't one. Trace any citation of it far enough and it terminates in another article citing a third article, or in a broker's blog post with no reference at all.
This matters more than it looks. A statistic with no provenance is unfalsifiable, which makes it equally useful for scaring people off and, in the hands of a course seller, for positioning their product as the way to be in the other 10%. Published figures can be checked, argued with and updated. Use those.
What the research actually measured
| Chague, De-Losso & Giovannetti (2019) | Barber, Lee, Liu & Odean (2014) | |
|---|---|---|
| Market | Brazilian equity index futures | Taiwan Stock Exchange |
| Period | Entrants 2013–2015, tracked onward | 1992–2006 |
| Population | All individuals who persisted 300+ days | All day traders in the market |
| Headline finding | 97% lost money | Under 1% reliably profitable after fees |
| Second finding | 0.4% earned more than a bank teller | Top-ranked traders persistently outperformed |
| On learning | No evidence of learning by day trading | Skill is identifiable and persistent |
The Brazilian futures study
Chague, De-Losso and Giovannetti observed every individual who began day trading in the Brazilian equity futures market — third in the world by volume in that instrument — between 2013 and 2015, and followed those who kept going for at least 300 trading days. Their abstract states it plainly: "97% of them lost money, only 0.4% earned more than a bank teller (US$54 per day), and the top individual earned only US$310 per day with great risk (a standard deviation of US$2,560)" (Chague, De-Losso & Giovannetti, University of São Paulo working paper, 2019).
The design is what makes it strong. It is not a survey and not one broker's clients — it is the complete population, including everyone who quit, which removes the survivorship bias that inflates almost every other figure you will read. And the sample is deliberately filtered to persistent traders: these are not people who tried it for a fortnight. 97% is the failure rate among the committed.
The Taiwanese equities study
Barber, Lee, Liu and Odean analysed the entire Taiwan Stock Exchange from 1992 to 2006 and reached a more nuanced conclusion: "less than 1% of the day trader population is able to predictably and reliably earn positive abnormal returns net of fees" (Barber, Lee, Liu & Odean, Journal of Financial Markets, 2014).
Crucially, that paper also found the top-ranked day traders went on to earn positive after-fee returns while the bottom-ranked continued to lose. Performance persisted, which means it was not purely luck. Skill in day trading is real and measurable — it is simply very rare, and the paper's own framing is that identifying it requires ranking traders on past performance across large samples, not on a good quarter.
How to read these numbers honestly
Both directions of misuse are common, and both are wrong.
- They do not prove day trading is impossible. A sub-1% success rate is a base rate for the population, not a physical law about you. The same Taiwanese data shows persistent skill at the top of the distribution.
- They do not entitle anyone to assume they are the exception. Everybody in the 97% believed they would not be. The base rate is the correct starting estimate for a new entrant, and it should only be revised by evidence you have actually collected — a few hundred logged trades — not by confidence.
- They are market-specific. Brazilian index futures and Taiwanese equities are two markets in two eras. The underlying mechanisms — costs, leverage, dispersion — exist everywhere, but treating "97%" as a universal constant is not what the papers support.
- "Profitable" is doing a lot of work. Profitable before or after costs? Over what sample? The Brazilian study's second figure is the sharper one: clearing a bank teller's wage, after everything, was achieved by 0.4%. Gross profit and a living are very different bars.
Why the failure rate is this high
The papers point at causes rather than blaming the participants, and the causes are structural.
- Costs compound against you. Spread and commission are paid on every round trip whether the trade wins or loses. High turnover multiplies a small per-trade cost into a large annual drag — which is why the same research group found, in an earlier study of 66,465 US households, that the most active traders earned 11.4% a year against a market return of 17.9% (Barber & Odean, Journal of Finance, 2000).
- Leverage converts a bad week into a terminal one. Futures and forex allow position sizes that make a normal losing streak unsurvivable. The trader is not wrong more often; they are simply not there any more when they would have been right.
- Feedback is too noisy to learn from. A good decision often loses and a bad one often wins, so outcome-based learning misleads. That is the most plausible explanation for the Brazilian study's finding of no learning even after 300 days — the topic we take apart in how long it takes to become a profitable trader.
- Most people never build a testable process. Without a written setup, a fixed risk percentage and a logged invalidation, there is nothing to evaluate, so nothing improves.
What actually separates the small group at the top
The research describes them rather than prescribing a method, and the description is consistent across both papers: persistence of performance, high activity concentrated in a few participants, and results that survive fees. Nothing in either study suggests a secret pattern or an indicator. What it does suggest is that the distinguishing variables are the unglamorous ones — cost per trade, size relative to account, and whether a method is followed consistently enough to be measured at all.
That is also why claims deserve scrutiny. Any room, course or signal service quoting a win rate without a verifiable statement history is asking you to take the most important number on trust. The checks worth running are in how to tell if a trading room's results are real.
Frequently Asked Questions
What percentage of day traders are profitable?
The best available evidence comes from two studies using complete market data. Among individuals who began day trading Brazilian equity index futures between 2013 and 2015 and persisted at least 300 days, 97% lost money and only 0.4% earned more than a bank teller. Across Taiwanese day traders from 1992 to 2006, less than 1% could predictably and reliably earn positive returns net of fees.
Where does the 90% of day traders lose money statistic come from?
Nowhere verifiable. The 90/90/90 formulation — 90% of traders lose 90% of their capital in 90 days — is repeated constantly and has no identifiable original study behind it. The real published figures, from Brazilian and Taiwanese market data, happen to be less flattering than the folklore, which is why the folklore is worth replacing rather than repeating.
Does anyone make money day trading?
Yes, but the share is small and concentrated. The Taiwanese data shows the top-ranked day traders going on to earn positive after-fee returns while the bottom ranks lose, so skill is measurable rather than mythical. The honest reading is that a persistent minority does earn, and that the base rate for any given new entrant is poor.
Do these studies apply to stocks and forex too?
The two headline studies cover equity index futures in Brazil and stocks in Taiwan, so applying them directly to other markets is an extrapolation. The mechanisms that drive the results — trading costs, leverage and dispersion of outcomes — are present in every retail market, which is why regulators in several jurisdictions require brokers to disclose the share of retail accounts that lose money on leveraged products.
Bottom line
Use 97% and under 1% instead of the sourceless 90% figure, and read them as base rates rather than verdicts. They tell you that most persistent day traders lose, that the small group who do not show genuinely persistent skill, and that the gap between the two is driven far more by costs, leverage and process than by pattern recognition. The rational response is not to quit or to assume you are exceptional — it is to keep risk small enough that you survive long enough to gather your own evidence, and to check that evidence honestly. Whether the activity is really a bet at all is the subject of is day trading gambling, and the process we hold ourselves to is the Method.