Why most day traders lose money — what 3.7 billion trades actually show
The short answer
The best evidence comes from Barber, Lee, Liu and Odean's analysis of the entire Taiwan Stock Exchange — 3.7 billion transactions from 1992 to 2006. More than 8 in 10 day traders lost money. Only about 1% were predictably profitable after fees. Day traders as a group lost roughly 23.9 basis points per day net of costs, with aggregate performance negative in 14 of the 15 years studied. Meanwhile 40% quit within their first month and 93% were gone within five years. The losses aren't bad luck — they're the arithmetic of costs applied at high frequency.
Why this study settles the question
Most claims about trader failure rates trace back to broker disclosures or surveys with obvious selection problems. This one doesn't. The researchers had the complete order-level record of an entire national exchange over 15 years — every participant, every trade, winners and quitters alike. Nobody could drop out of the sample by going quiet.
That's why the specific numbers survive:
- ~5% of active day traders were consistently profitable in any given year
- ~1% were predictably profitable after fees — the population where skill, not luck, is the plausible explanation
- 72% of aggregate day trading volume came from traders with a history of losses
- −23.9 basis points per day net of fees for the group
That third number is the one people miss. The market's day-trading volume is dominated by people who are losing — which means the counterparty to a typical trade is not an amateur making a mistake, it's usually a professional or a market maker.
The mechanism: costs applied at frequency
There's no mystery here, and it isn't about intelligence.
Every trade costs something — commissions where they exist, the bid-ask spread always, and slippage on anything of size. Call it a small fraction of a percent per round trip. Trade twice a day and you've applied that toll several hundred times a year.
Now the trader must be right often enough to clear that toll before earning a cent. Not right on average — right by a margin that exceeds accumulated costs. The paper's −23.9 bps per day is what that arithmetic produces at scale.
And short-horizon price movement is close to random, so the underlying edge you're paying that toll to express is very small for almost everyone.
"Day traders with strong past performance also achieve strong future returns. However, after fees are deducted, only one percent of them are predictably profitable." — Barber, Lee, Liu & Odean
Note carefully what that quote concedes: skill exists. Past performance does predict future performance among day traders. The problem is that the fee hurdle consumes nearly all of it, leaving roughly 1% clearing the bar.
The survivorship trap that keeps this going
If 93% quit within five years, then almost everyone visibly talking about day trading at any moment is either in their first months — before the arithmetic has caught up — or is in the rare surviving fraction.
The losers don't post. They quietly stop, and they're invisible. So the observable population of day traders looks vastly more successful than the actual population, and every new entrant calibrates their expectations against a sample that has had its failures deleted.
This is the same distortion behind "everyone I know made money on that trade." You're not hearing from everyone you know.
What separates this from investing
Day trading and long-term investing are frequently discussed as points on one spectrum. They aren't — they're structurally opposite bets.
| Day trading | Long-term investing | |
|---|---|---|
| Source of return | Short-horizon price moves | Business earnings over years |
| Cost drag | Hundreds of round trips a year | A handful |
| Time horizon | Minutes to hours | Years to decades |
| Base rate | ~1% predictably profitable after fees | Broad indexes historically positive over long horizons |
The long-term investor is being paid for holding productive assets through time. The day trader is trying to win a zero-sum contest against better-resourced counterparties while paying a toll on every attempt.
If you still want to try
This isn't a lecture — some people genuinely want to, and a small number are in that 1%. But there's an obvious sequencing question: you don't yet know whether you're in the 1%, and finding out with real money costs an average of a few years and a lot of tuition.
The cheap version:
- Simulate it first, with real prices and honest accounting. Include spread and costs. A strategy that's profitable before costs and unprofitable after is the single most common self-deception in this area.
- Grade decisions, not outcomes. Over a short window, results are mostly noise — a profitable month tells you almost nothing. Sydnical's coach marks each trade Brilliant, Good, Inaccuracy, Mistake, or Blunder based on reasoning and risk at the moment you acted, so a lucky win still registers as a Blunder.
- Look for your pattern, not your P&L. Do you size up after wins? Hold losers and cut winners? Trade more after a loss? These show up within weeks and they're what actually determines the outcome.
- Set a decision rule in advance for what evidence would tell you to stop — written down before you start, because the 93% didn't plan their exit either.
FAQ
What percentage of day traders lose money?
In the most comprehensive study available — every transaction on the Taiwan Stock Exchange from 1992 to 2006, 3.7 billion trades — more than 80% of day traders lost money, around 5% were consistently profitable in a given year, and only about 1% were predictably profitable after fees.
Why do day traders lose money?
Primarily costs applied at high frequency. Every round trip pays a spread and often a commission, and trading frequently applies that toll hundreds of times a year. Since short-horizon price movement is close to random, the edge most traders are paying to express is smaller than the toll. The Taiwan data showed day traders losing about 23.9 basis points per day net of fees.
Can anyone actually make money day trading?
Yes, but rarely. The research found genuine performance persistence — strong past performers did tend to perform well again — yet after fees only around 1% were predictably profitable. Skill exists; the fee hurdle consumes almost all of it.
How long do most day traders last?
About 40% quit within the first month, and 93% had stopped within five years, according to Barber and Odean's follow-up work. This creates a survivorship illusion, because the people who quit stop talking about it while survivors remain visible.
Is day trading the same as investing?
No. They're structurally different bets: day trading tries to extract short-horizon price moves in a near zero-sum contest while paying costs on every attempt, whereas long-term investing is compensated for holding productive assets through time and incurs a fraction of the trading costs.
The data is unusually clear here, which is rare in investing. If you want to find out which population you're in, do it where being wrong is free.