Across every serious dataset, between roughly 80% and 97% of day traders lose money, and the share who earn a real living from it sits somewhere between 1% and 4%. Those aren’t forum guesses. They come from studies that tracked every trader in an entire market for years, plus payout records from hundreds of thousands of funded-trader accounts, all linked at the bottom of this page.
This page exists because most day trading statistics you’ll see online are recycled without sources. We pulled the numbers below from the original academic papers, regulator publications, and disclosed industry data, and we tell you which famous figures have no study behind them at all.
The headline numbers
| Statistic | Figure | Source |
|---|---|---|
| Persistent Brazilian day traders (300+ days) who lost money | 97% | Chague, De-Losso & Giovannetti (2020) |
| Brazilian day traders earning more than minimum wage | 1.1% | Chague, De-Losso & Giovannetti (2020) |
| Taiwanese day traders who predictably profit net of fees | Under 1% | Barber, Lee, Liu & Odean (2014) |
| Day traders who quit within two years (Taiwan, full market) | Over 75% | Barber, Lee, Liu, Odean & Zhang (2017) |
| US day traders who lose vs. make money | Roughly 2 to 1 | Jordan & Diltz (2003) |
| Prop firm accounts that ever receive a payout | 7% | FPFX Tech data, 300,000+ accounts (2024) |
| Years in which Taiwanese day traders were net profitable in aggregate, 1992–2006 | 0 of 15 | Barber, Lee, Liu, Odean & Zhang (2017) |
Every row traces to a primary source in the citations section. Now the detail, because the detail is where the lessons are.
The two studies that watched entire markets
Most trading research samples one broker’s clients. Two studies did something better: they observed every individual day trader in an entire country’s market. That kills the sampling-bias objection dead.
Taiwan, 1992–2006: less than 1% reliably profit
Barber, Lee, Liu and Odean analyzed complete Taiwan Stock Exchange transaction data over 15 years. In the average year, about 450,000 individuals day traded, and their round trips accounted for 17% of all exchange volume. Among the heavier traders (roughly 277,000 people moving more than about US$20,000 a year), about 20% finished a given year with positive returns net of fees.
One profitable year is noise. The paper’s real finding is what happens next: sort traders by performance in year one, then watch year two. Only about 4,000 traders, less than 1% of the day trading population, went on to earn reliably positive returns after costs. The top 500 were genuinely skilled, earning 37.9 basis points per day net of fees. Traders with a losing history went on to lose 28.9 basis points per day. Skill exists. It’s just brutally rare, and last year’s losers keep losing.
The companion paper on trader learning adds the attrition data: more than 75% of all day traders quit within two years, and in every single one of the 15 years studied, day traders as a group lost money net of fees. Not most years. All of them. The strangest finding is who stays: experienced traders with a history of losses kept trading at a 95.3% rate over the following year, almost identical to the 96.4% rate for experienced winners. Losing doesn’t make people stop. That’s the behavioral hole this entire page is trying to help you avoid.
Brazil, 2013–2015: 97% of the persistent lost money
The Getulio Vargas Foundation study tracked every individual who started day trading Brazilian equity futures between 2013 and 2015, one of the world’s largest futures markets by volume. The researchers deliberately stacked the deck in traders’ favor by looking only at the committed ones, those who persisted for more than 300 trading days.
Among that committed group: 97% lost money. Just 1.1% earned more than the Brazilian minimum wage. Only 0.5% out-earned the starting salary of a bank teller, and the authors note even those few carried large risk to do it. Their abstract opens with the conclusion: it is virtually impossible for individuals to day trade for a living, contrary to what course providers claim. Academic papers rarely talk like that. This one had the data to.
Persistence, in other words, is not the edge the gurus say it is. The 300-day filter selected for dedication, and dedication alone produced a 3% survival rate.
The US data
The cleanest peer-reviewed US study is Jordan and Diltz in the Financial Analysts Journal (2003). Using two separate methodologies on a sample of American day traders, they found about twice as many traders lose money as make it, with roughly 20% more than marginally profitable. They also found day trader profitability moved with the Nasdaq Composite, which is a polite way of saying a lot of “skill” in that era was a rising index.
The US numbers look kinder than Brazil’s because of timeframe: a 20% profitability snapshot is what Taiwan showed in single years too. The longer the window, the smaller the winners’ circle gets. Single-year profitability rounds to 20%. Multi-year, repeatable profitability rounds to 1%.
Prop firm pass rates: the modern numbers
Funded-trader programs publish the freshest large-scale data, because their business model is built on the failure rate. The biggest dataset comes from FPFX Tech, a technology provider whose platform data covers more than 300,000 accounts from 100,000 traders across 10 prop firms, as disclosed to Finance Magnates in 2024:
- 14% of traders passed the evaluation and got a funded account.
- About 45% of those funded ever achieved a payout, which is 7% of all traders.
- The average payout was 4% of the plan size.
- The average account spent about $800 on challenge fees across its life, typically buying three attempts.
The Funded Trader’s CEO disclosed similar internals in 2025: a 5–10% challenge pass rate, with only about 20% of funded traders reaching a payout. That works out to 1–2% of all clients ever getting paid.
The worked math on a prop challenge
Run the expected value with the FPFX averages. Say you’re chasing a $100,000 funded plan. The average payout, when one happens, is 4% of plan size: $4,000. The probability that an account ever reaches any payout is 7%. Expected payout per account: 0.07 × $4,000 = $280. Average lifetime spend on challenges: $800.
So the average dollar spent on prop challenges returned about 35 cents. These are averages across 300,000 accounts, and a small group of repeat earners does far better than the mean, the same skew the Taiwan data shows. But if you’re average, a prop challenge is a negative-expectancy product with a lottery-shaped payoff. Buy it knowing that, or don’t buy it.
Which famous statistics are real and which are folklore
Three numbers get repeated endlessly. Here’s their provenance status:
“90% of traders lose 90% of their money in 90 days.” Catchy, symmetrical, and unsourced. We could not connect it to any study, and the pages that repeat it never cite one. Treat it as a slogan, not a statistic.
“95% of day traders fail.” Also no single study behind it. The honest version is messier: the verified loss rates run from about 67% (twice as many losers as winners in the 2003 US sample) to 97% (persistent Brazilian traders), depending on market, era, and how long you watch. The folklore number isn’t far off the truth. It’s just a number someone rounded into existence.
“Only 1% make it.” This one actually checks out, if “make it” means repeatable profits over years. Taiwan: under 1% predictably profitable. Brazil: 1.1% above minimum wage. Funded accounts: 1–7% ever paid. Three independent datasets, three methodologies, same neighborhood.
Why broker-published loss rates understate the problem
European regulators force CFD brokers to publish client loss rates, and the disclosures typically say something like 70% of retail accounts lose money. That sounds better than the academic numbers, and the gap is survivorship bias: brokers report on accounts that currently exist. The trader who lost 40% in three months and closed the account isn’t in the denominator anymore. The academic studies above counted everyone who ever started, which is the number that describes your actual odds on day one.
The same bias runs through social media. Profitable traders post; the 97% mostly don’t. The visible sample is the winners’ circle, and the winners’ circle is not the population. The Taiwan persistence data shows where that leads: losing traders kept trading at almost the same rate as winning ones, year after year.
What separates the profitable few
The studies say more than “most lose.” They say something about who doesn’t:
The Taiwan data found past performance was by far the best predictor of future performance, and the next best predictor was concentration: profitable day traders focused on a small number of stocks, consistent with building an informational edge in a few names rather than chasing whatever’s moving. The profitable group also made its money in hard-to-value situations (small, volatile stocks, earnings windows) where an information edge pays, exactly where the unprofitable group lost the most. Same stocks in play, opposite outcomes, because one side had an edge and the other side was the edge.
None of the datasets found a shortcut. They found a small, persistent group with measurable skill, surrounded by a very large group funding them. If you’re going to try this, your only honest project is finding out cheaply which group you’re in. That’s what simulators and journals are for, and it’s why our guide on whether day trading is worth it leans on these numbers rather than income screenshots. For what the rare winners actually earn, see how much day traders make.
The rules changed in 2026. The odds didn’t.
A regulatory note, because most statistics pages online predate it: the FINRA pattern day trader rule and its $25,000 minimum no longer exist. The SEC approved FINRA’s replacement framework on April 14, 2026, effective June 4, 2026, with brokers permitted a transition window through October 20, 2027. There’s no trade counting and no PDT designation anymore. Instead, brokers monitor intraday margin in real time; the $2,000 minimum equity for leveraged trading still applies, you must hold maintenance margin throughout the day rather than just at the close, and repeatedly failing to cover an intraday margin deficit can get your account restricted for up to 90 days. Full breakdown in our PDT rule explainer and the intraday margin requirements guide.
What the change did: removed the $25,000 gate, so small accounts can day trade on margin. What it didn’t do: change a single number on this page. The Brazilian traders who went 97% red faced no PDT rule. Lower barriers mean more entrants, and every dataset above says most entrants pay for the few. Access got easier. The math didn’t.
What to do with these numbers
If the data hasn’t talked you out of it, at least let it set your process. Start with the field manual at how to start day trading, then prove an edge where losing is free, using one of the paper trading apps before real money sees a fill. Size every live trade like a member of the 97% until you have a year of evidence otherwise; the position size calculator does the arithmetic. And log everything in the trading journal template, because the Taiwan data’s darkest finding is traders who lost for years without acting on it. A journal is how you avoid becoming that row in someone’s dataset.
Sources
- Chague, De-Losso & Giovannetti, “Day Trading for a Living?” (2020): SSRN working paper 3423101
- Barber, Lee, Liu & Odean, “The Cross-Section of Speculator Skill: Evidence from Day Trading,” Journal of Financial Markets (2014): full paper (Berkeley Haas)
- Barber, Lee, Liu, Odean & Zhang, “Do Day Traders Rationally Learn About Their Ability?” (2017): full paper (Berkeley Haas)
- Jordan & Diltz, “The Profitability of Day Traders,” Financial Analysts Journal 59(6), 2003: CFA Institute abstract
- FPFX Tech prop account data (2024): Finance Magnates exclusive
- The Funded Trader pass-rate disclosure (2025): Finance Magnates report
- FINRA, “Understanding the New Intraday Margin Requirements” (April 2026): FINRA investor insight
- FINRA Regulatory Notice 26-10, adopting intraday margin standards (April 20, 2026): FINRA.org
Frequently asked questions
What percentage of day traders are actually profitable?
In any single year, roughly 20% of active day traders finish profitable after fees, based on both the Taiwan market data and the 2003 US study. Over multiple years the number collapses: under 1% of Taiwanese day traders earned reliable profits net of costs, and 1.1% of persistent Brazilian day traders out-earned minimum wage. Annual profitability and durable profitability are very different claims.
Do any day traders make a living?
Yes, a small and measurable group. The Taiwan study identified about 4,000 traders out of roughly 450,000 (the top 500 earned 37.9 basis points per day net of fees, year after year), and the Brazilian study found 0.5% out-earning a bank teller’s starting salary. Skill is real and persistent in the data. It’s also rarer than almost any other career outcome people plan around.
How do prop firm traders compare?
Worse than the headline pass rates suggest. Across 300,000+ accounts in FPFX Tech’s dataset, 14% passed an evaluation but only 7% ever received a payout, averaging 4% of the plan size, while the average account spent about $800 on challenge fees. The Funded Trader disclosed that only 1–2% of its clients ever got paid. Passing a challenge and getting paid are separate hurdles, and most traders clear neither.
Did eliminating the PDT rule improve the odds?
No. The 2026 rule change removed the $25,000 minimum and trade counting, replacing them with intraday margin requirements, which lowers the barrier to entry but changes nothing about the loss-rate data. The Brazilian futures traders in the 97%-loss study never faced a PDT rule at all. Easier access historically means more participants, not more winners.
Why do most day traders lose money?
Costs and competition, in that order. Commissions, spreads, and slippage create a negative-sum game for the group: in Taiwan, day traders in aggregate lost money net of fees in all 15 years studied. On the other side of those trades sit the small skilled minority and professional firms. Behavior compounds it; the data shows losing traders persist at nearly the same rate as winning ones instead of stopping or sizing down.
Are these statistics relevant to US traders if the biggest studies are foreign?
The Taiwan and Brazil studies are used worldwide because they had something no US dataset offers: every trader in an entire market, not one broker’s clients. The US-specific evidence that exists (the Financial Analysts Journal study finding two losers per winner, and prop firm data with heavy US participation) points the same direction. Three markets, three decades, one conclusion.