Day trading: what the evidence shows — ROR Labs cover showing 97% lost money.
|

Day Trading: What the Evidence Shows

Key takeaways · 12 min read

  • In the classic US study, the most active households earned 11.4% a year against 17.9% for the market.
  • Individual investors across the whole Taiwanese market lost the equivalent of 2.2% of GDP to trading.
  • Fewer than 1% of Taiwanese day traders earned money predictably, and in Brazil 97% of persistent day traders lost money.
  • Commission-free trading lowered costs without removing them, and heavily bought stocks tend to underperform.

Starting to day trade has never been easier. A phone app, no commission and a few minutes of setup are enough to buy a stock at ten in the morning and sell it before lunch. Social media is full of screenshots of winning days, and the market keeps producing stories, from GameStop in 2021 to the meme-stock revivals of 2024 and 2025, that make fast trading look like a skill anyone can learn.

For more than 25 years, researchers have measured what happens to people who trade often, using complete brokerage and exchange records instead of surveys. The data come from the United States, Taiwan and Brazil, from the 1990s to the app era, and they agree to a degree that is unusual in finance: the more individuals trade, the worse they tend to do, and only a tiny fraction of day traders earn money reliably.

There is a real counter-argument. Several careful studies find that retail buying predicts returns, and one found that the Robinhood crowd timed the 2020 crash well. This article looks at both sides, and at why they turn out not to contradict each other.

A pointillist illustration: a desk at night with three monitors showing red and green price bars, a keyboard and a mug, a small amber desk lamp and a window of city lights.
A trading desk at night. The data on frequent trading are unusually consistent.

The study that started it

In 2000, Brad Barber and Terrance Odean published a paper in the Journal of Finance under a blunt title: ‘Trading Is Hazardous to Your Wealth’. They analysed the accounts of 66,465 households at a large US discount broker from 1991 to 1997, sorting them into five groups by how much of their portfolio they turned over each month.

The least active fifth turned over about 0.19% of their portfolio a month; the most active fifth turned over 21.49%, replacing their entire portfolio roughly every five months. After costs, the most active households earned 11.4% a year. The market returned 17.9%, the average household 16.4%, and a portfolio mimicking the least active group about 18.5%.

Before costs, the groups performed about the same. The frequent traders were not picking worse stocks; they were paying more to hold them. In that era a round trip on a trade of more than $1,000 cost about 3% in commissions and 1% in bid-ask spread, and the gap between the groups was almost entirely the cost of trading.

More trading, lower returns

Annual net return, US households at one discount broker, 1991–1997.

Least active fifth (mimicking portfolio)18.5%
Market17.9%
Average household16.4%
Most active fifth11.4%

Barber and Odean, Journal of Finance, 2000. 66,465 households. Costs were far higher then than now.

‘Boys Will Be Boys’ was a follow-up paper by the same authors, using the same broker’s data. Men traded 45% more than women, and trading cut men’s net returns by 2.65 percentage points a year against 1.72 points for women; single men traded 67% more than single women. The authors read this as overconfidence: people who think they know more trade more, and pay for it.

A whole stock market

One broker in the 1990s is a narrow window, but Taiwan’s stock exchange recorded every trade and who made it, so Barber, Yi-Tsung Lee, Yu-Jane Liu and Odean could track the entire market from 1995 to 1999. In 2009 they reported that the aggregate portfolio of individual investors suffered an annual performance penalty of 3.8 percentage points from trading.

The losses added up to 2.2% of Taiwan’s gross domestic product, or 2.8% of total personal income. Institutions gained about 1.5 percentage points a year, and foreign institutions took nearly half of institutional profits. Virtually all individual trading losses traced to aggressive orders, which trade immediately at the market price, the authors found.

What individual investors lost in Taiwan

All trades on the Taiwan Stock Exchange, 1995–1999.

3.8 ptsannual performance penalty for individual investors as a group
2.2%of Taiwan’s GDP, the total cost of those losses

Barber, Lee, Liu and Odean, Review of Financial Studies, 2009. Taiwan also charged a transaction tax on sales.

Day traders in particular

The Taiwanese data also allowed the first large study of day traders, who buy and sell the same stock within one day. Across 1992 to 2006, about 450,000 people a year day traded in Taiwan, accounting for 17% of the market’s volume. About one in five day traders earned positive returns after fees in a typical year, yet fewer than 1% could do so predictably, year after year.

That small group was real: the top 500 day traders earned about 38 basis points a day after fees, which is a lot. A 2020 follow-up paper, ‘Learning, Fast or Slow’, asked why everyone else kept going. Unprofitable traders were more likely to quit, but slowly: 74% of day-trading volume came from people with a history of losses, and the authors estimated that 97% of day traders were likely to lose money if they continued.

A pointillist illustration: a smartphone on a bedside table showing a jagged red price line, beside a red digital alarm clock under an amber lamp.
A phone by the bed. Most day-trading volume comes from people with a history of losses.

Brazil supplies the most quoted modern figure. Fernando Chague, Rodrigo De-Losso and Bruno Giovannetti studied everyone who began day trading index futures between 2013 and 2015, 19,646 people. Only 1,551 kept going for more than 300 trading days, and 97% of those persistent traders lost money. Just 1.1% earned more than the Brazilian minimum wage, and 0.5% more than the starting salary of a bank teller.

That paper remains a working paper, and futures are not shares, so Chague and Giovannetti repeated the exercise for Brazilian stocks in a short 2020 article. Of 98,378 people who started day trading stocks between 2013 and 2016, only 554 lasted more than 300 sessions, with an average daily result, before income tax, of a loss of about 49 reais. Setting aside the first 200 sessions as a learning period made the average worse.

Day trading for a living

Brazilians who started day trading index futures in 2013–2015 and continued for more than 300 days.

Lost money97%
Earned more than the minimum wage1.1%
Earned more than a bank teller’s starting pay0.5%

Chague, De-Losso and Giovannetti, working paper, 2019–2020. 1,551 persistent traders out of 19,646 who started.

Zero commissions changed the cost, not the pattern

Commission-free trading became standard in the United States around 2019 and removed the largest cost in the Barber and Odean data, but costs remain. Most US brokers now route retail orders to wholesale market makers, which pay the broker for the order flow and earn the spread. In 2025, Christopher Schwarz, Barber, Xing Huang, Philippe Jorion and Odean placed about 85,000 simultaneous orders through six accounts at five brokers to see what trades actually cost.

Even with no commission, the average round trip cost between 0.07% and 0.46% depending on the account. Wholesalers pricing the same trade differently for different brokers explained the differences, and payment for order flow did not. For an investor who trades a few times a year, those costs hardly matter, but for someone who turns over their account every week, even 0.07% a trade adds up.

The app era added a second problem, attention. In a 2022 study of Robinhood users, Barber, Huang, Odean and Schwarz found that the stocks users bought most each day went on to underperform, with average abnormal returns of −4.7% over the next 20 days, and −19.6% in extreme herding events when user numbers in a stock jumped suddenly. The authors linked part of the effect to the way the app displayed lists of popular stocks.

What the app era added

Two costs that commission-free trading did not remove.

0.07–0.46%average round-trip cost per account across five US brokers, with no commission
−4.7%20-day abnormal return of the stocks Robinhood users bought most each day

Schwarz and colleagues, Journal of Finance, 2025; Barber, Huang, Odean and Schwarz, Journal of Finance, 2022.

The case for the retail trader

Not all the evidence points one way. Eric Kelley and Paul Tetlock found in 2013 that daily retail buying imbalances positively predicted stock returns up to 20 days ahead, with no reversal later, and that retail market orders seemed to anticipate the tone of news. In 2021, Ekkehart Boehmer and colleagues found that stocks with net retail buying outperformed those with net retail selling by about 10 basis points the following week.

Ivo Welch made the strongest defence in 2022. Looking at Robinhood holdings from mid-2018 to mid-2020, he found that users increased their holdings during the March 2020 crash, without panic or margin calls, and were rewarded in the rebound. In his words, their aggregate portfolio had both good timing and good alpha over that period.

How can retail buying predict returns while retail traders lose money? In 2024, Barber, Shengle Lin and Odean offered an answer. The prediction studies weight every stock equally, but retail money is concentrated in a small number of attention-grabbing stocks. A strategy following extreme retail buying lost 14.8% a year in stocks with heavy retail trading and earned 6.6% in all other stocks. Retail traders were right on average across many small positions, and wrong where most of their money was.

Resolving the paradox

Annual return of a strategy that follows extreme retail order imbalances, by type of stock.

StocksReturn of following retail buyingWhat it means
Heavily traded by retail investors−14.8% a yearWhere most retail money goes, it loses
All other stocks+6.6% a yearAcross many small positions, retail buying is informative

Barber, Lin and Odean, Journal of Financial and Quantitative Analysis, 2024.

Welch acknowledged limits to his finding: it describes a crowd portfolio built from holdings snapshots over two years of a rising market, and does not give the realised returns of individual accounts. The studies on the other side have weaknesses too. Barber and Odean’s original data came from one broker in an era of high commissions, Taiwan taxed every sale, and the Brazilian figures describe futures. The consistency across such different settings makes the core finding hard to dismiss; no single study does.

What regulators say

The investor guide on day trading from the US Securities and Exchange Commission, first published in 2005 and still online, is unusually direct: it tells readers to ‘be prepared to suffer severe financial losses’, notes that day traders typically depend heavily on borrowed money, and warns against claims of easy profits.

For two decades the main US rule was FINRA’s pattern day trader rule: anyone who made four or more day trades within five business days in a margin account had to keep at least $25,000 in equity. In April 2026 the SEC approved FINRA’s replacement, which removes the pattern day trader concept and requires brokers to monitor intraday margin deficits in any margin account instead. The new rule took effect on 4 June 2026 and is being phased in through October 2027, so some brokers still apply the old limits.

On the most leveraged products, European regulators took a harder line: in 2018 the European Securities and Markets Authority restricted contracts for difference after finding that 74% to 89% of retail accounts typically lost money on them. We covered those products in our article on options and futures trading.

Recent cases

GameStop in January 2021 remains the defining example. According to an SEC staff report, the stock closed at $347.51 on 27 January, more than 1,600% above its close on 11 January, and reached $483 during trading the next day. The number of unique accounts trading it on a given day rose from fewer than 10,000 at the start of the month to nearly 900,000. Short sellers buying to cover, the staff found, made up only a small fraction of the buying.

It has recurred since. In May 2024 GameStop closed up 74% in a single day after the investor known as Roaring Kitty posted online again, and in July 2025 a new wave of meme trading pushed Opendoor up more than 400% within the month. In the first half of 2026, Citadel Securities, the largest wholesaler of US retail orders, reported record retail activity on its platform, with 12 June 2026 the largest single day of retail net buying it had seen. Those are the firm’s own figures, and the firm profits from that order flow.

A pointillist illustration: a laptop on a kitchen table showing a price chart, a mug of coffee and a window at dawn.
A laptop on the kitchen table. Costs are lower than in the 1990s, but the pattern has not changed.

A worked example: a $4,000 margin account and the 25% floor

The regulators section above covers the June 2026 change. This section puts numbers on one rule FINRA’s investor page states. The arithmetic is ours and is only an illustration; it is not a FINRA example.

The rules, as FINRA states them (investor page dated 20 April 2026). You need at least $2,000 in equity to trade on margin. You must keep equity of at least 25% of the market value of long margin-eligible stock throughout the trading day, and your broker may set a higher figure. The old $25,000 minimum for day trading no longer applies under the new standard, although during the transition (which runs to 20 October 2027) some firms still apply the old rules.

The example. You put in $4,000 and borrow $6,000 to buy $10,000 of stock. At the start the 25% requirement is $2,500, so you have a cushion of $1,500. Your equity is always the stock’s value minus the $6,000 loan.

What a falling price does to the $4,000 (illustrative)
Price moveStock worthYour equity25% requirementResult
0%$10,000$4,000$2,500Cushion of $1,500
−10%$9,000$3,000$2,250Still above the floor, but $1,000 (25%) of your own money is gone
−20%$8,000$2,000$2,000Exactly at the floor; half your own money is gone
−25%$7,500$1,500$1,875Short by $375

A fall of 20% in the stock costs you 50% of your own money, because the loan does not shrink. A fall of 25% leaves you short of the requirement, which is when a broker asks for more money or sells. This ignores interest, trading costs and any higher requirement your broker sets.

What happens next. FINRA says repeated unmet deficits can lead to an account restriction of up to 90 days, and that you could lose more than you invested. The page itself says frequent margin trading remains high-risk despite the lower entry bar, and advises using only money you can afford to lose.

What we could not confirm. The FINRA and SEC pages we read give no statistic for how many day traders lose money (the studies earlier in this post cover that). We could not confirm which brokers have already switched to the new standard; FINRA says it varies by firm, so ask yours.

What the evidence suggests

The research does not say that nobody can trade successfully. It says the people who can are very rare, that they are hard to identify in advance, including by themselves, and that most of the people who keep trying have already been losing. Costs and attention do most of the damage: frequent traders tend to pay more, buy what everyone else is buying, and concentrate their money where the odds are worst.

The same researchers’ work on long-term investing points to the opposite habits: low turnover, low costs and broad diversification. We looked at why a small number of stocks drive most market returns in our article on stock investing losses.

What to do

These are checks, not trading instructions. Steps sorted by when they matter. Each one says how strong the evidence is: Measured Official advice Untested

Before

  • Read the SEC’s investor guide on day trading first. It tells readers to be prepared to suffer severe financial losses. Official advice see Sources
  • Count what you pay in costs and how often you trade. Costs and attention do most of the damage. Measured see Sources
  • Check whether the money is borrowed. Day traders typically depend heavily on borrowed money. Official advice see Sources

What not to do

  • Do not believe claims of easy profits. The rare traders who succeed are hard to identify in advance, including by themselves. Measured see Sources

Not financial advice.

Questions people ask

What percentage of day traders make money?

In a typical year in Taiwan, about one in five day traders made money after fees, but fewer than 1% did so predictably. In Brazil, 97% of those who persisted for more than 300 days lost money.

Is day trading profitable with zero commissions?

Commissions were the largest cost in older studies, but trades still carry spread costs, and studies of app users find that heavily bought stocks tend to underperform afterwards.

Do men trade more than women?

In one US broker’s data, men traded 45% more than women, and trading cost them more return.

Is the $25,000 pattern day trader rule still in force?

FINRA replaced it with an intraday margin standard that took effect on 4 June 2026, with a phase-in until October 2027. Some brokers may still apply the old rule during that period.

Can retail traders beat professionals?

Some studies find retail buying predicts returns on average, but the money is concentrated in attention-grabbing stocks where following retail buying has lost money.

The short version

  • In the classic US study, the most active households earned 11.4% a year against 17.9% for the market.
  • Individual investors across the whole Taiwanese market lost the equivalent of 2.2% of GDP to trading.
  • Fewer than 1% of Taiwanese day traders earned money predictably, and in Brazil 97% of persistent day traders lost money.
  • Commission-free trading lowered costs without removing them, and heavily bought stocks tend to underperform.
  • Retail buying can predict returns on average, but not where most retail money goes.

This article summarises published research on trading by individual investors. It is not financial, investment or tax advice, and it does not recommend any security, broker, product or strategy. Past results in these studies do not predict future results. For decisions about your own money, consider speaking to a regulated, independent financial professional.

Further reading. Barber and Odean, ‘Trading Is Hazardous to Your Wealth’, Journal of Finance, 2000, is the starting point. Barber, Lin and Odean, Journal of Financial and Quantitative Analysis, 2024, is the best reconciliation of the evidence for and against retail traders.

Sources

  1. Barber BM, Odean T. Trading is hazardous to your wealth: the common stock investment performance of individual investors. Journal of Finance, 2000;55(2):773–806. doi:10.1111/0022-1082.00226.
  2. Barber BM, Odean T. Boys will be boys: gender, overconfidence, and common stock investment. Quarterly Journal of Economics, 2001;116(1):261–292. doi:10.1162/003355301556400.
  3. Barber BM, Lee YT, Liu YJ, Odean T. Just how much do individual investors lose by trading? Review of Financial Studies, 2009;22(2):609–632. doi:10.1093/rfs/hhn046.
  4. Barber BM, Lee YT, Liu YJ, Odean T. The cross-section of speculator skill: evidence from day trading. Journal of Financial Markets, 2014;18:1–24.
  5. Barber BM, Lee YT, Liu YJ, Odean T, Zhang K. Learning, fast or slow. Review of Asset Pricing Studies, 2020;10(1):61–93. doi:10.1093/rapstu/raz006.
  6. Chague F, De-Losso R, Giovannetti B. Day trading for a living? Working paper, SSRN 3423101, 2019, revised 2020.
  7. Chague F, Giovannetti B. É possível viver de day-trade em ações? Brazilian Review of Finance, 2020;18(3):1–4.
  8. Schwarz C, Barber BM, Huang X, Jorion P, Odean T. The ‘actual retail price’ of equity trades. Journal of Finance, 2025;80(5):2507–2541. doi:10.1111/jofi.13467.
  9. Barber BM, Huang X, Odean T, Schwarz C. Attention-induced trading and returns: evidence from Robinhood users. Journal of Finance, 2022;77(6):3141–3190. doi:10.1111/jofi.13183.
  10. Kelley EK, Tetlock PC. How wise are crowds? Insights from retail orders and stock returns. Journal of Finance, 2013;68(3):1229–1265. doi:10.1111/jofi.12028.
  11. Boehmer E, Jones CM, Zhang X, Zhang X. Tracking retail investor activity. Journal of Finance, 2021;76(5):2249–2305. doi:10.1111/jofi.13033.
  12. Welch I. The wisdom of the Robinhood crowd. Journal of Finance, 2022;77(3):1489–1527. doi:10.1111/jofi.13128.
  13. Barber BM, Lin S, Odean T. Resolving a paradox: retail trades positively predict returns but are not profitable. Journal of Financial and Quantitative Analysis, 2024;59(6):2547–2581. doi:10.1017/S0022109023000601.
  14. US Securities and Exchange Commission. Day trading: your dollars at risk. Investor publication, 2005.
  15. FINRA. Regulatory Notice 26-10: FINRA adopts new intraday margin standards to replace the day trading margin requirements. April 2026.
  16. European Securities and Markets Authority. ESMA agrees to prohibit binary options and restrict CFDs to protect retail investors. Press release ESMA71-98-128, March 2018.
  17. US Securities and Exchange Commission. Staff report on equity and options market structure conditions in early 2021. October 2021.
  18. Citadel Securities. 1H 2026 market structure and flows. 2026.
  19. FINRA. Regulatory Notice 26-10, “FINRA Adopts New Intraday Margin Standards to Replace the Day Trading Margin Requirements” (20 April 2026).
  20. FINRA. “Understanding the New Intraday Margin Requirements” (20 April 2026).

Similar Posts