Start here
Base rates
Before any of the vocabulary, the outcomes. Five peer-reviewed studies that followed entire populations of retail traders — not volunteers, not survivors, not the ones still posting — and counted how they finished.
What the studies measured
5Five findings, each from a whole population rather than a sample of volunteers. Populations matter here: the people who post their results are not a random draw from the people who trade. Each one takes a screen of its own — keep scrolling and the next will cover it.
97%
of persistent day traders lost money
Every individual who started day trading Brazilian equity futures between 2013 and 2015 was tracked. Of those who kept at it for more than 300 trading days — the ones who treated it as an occupation rather than a experiment — 97% lost money. Only 1.1% earned more than the Brazilian minimum wage from it, and only 0.5% earned more than a bank teller's starting salary.
Read it carefullyOne market, one instrument, one country. The survivorship direction runs against the traders: persistence made the odds worse, not better.
Chague, De-Losso & Giovannetti, Day Trading for a Living? (2020)
<1%
were reliably profitable after costs
Fifteen years of the entire Taiwan Stock Exchange. In a typical year about a fifth of active day traders finished ahead of costs — roughly what chance would produce. But when the researchers asked which traders were predictably profitable, meaning their past record forecast their future record, the answer was under 1% of the population: about 4,000 people out of 450,000.
Read it carefullySkill exists and is measurable. It is also rare enough that the prior odds of any given person having it are roughly one in a hundred.
Barber, Lee, Liu & Odean, The Cross-Section of Speculator Skill, Journal of Financial Markets (2014)
11.4%
vs 17.9% — the cost of trading often
66,465 households at a US discount broker, 1991 to 1996. The average household earned 16.4% a year against a market that returned 17.9%. The fifth that traded most earned 11.4% — six and a half percentage points a year behind an index they could have bought and left alone. Their gross stock picks were fine. Trading costs ate the difference.
Read it carefullyCommissions have since fallen to zero on most retail platforms, which removes part — not all — of this penalty. Spreads, slippage and taxes remain.
Barber & Odean, Trading Is Hazardous to Your Wealth, Journal of Finance (2000)
3.8pp
a year, transferred from individuals to institutions
Across all trading on the Taiwan Stock Exchange from 1995 to 1999, the aggregate portfolio of individual investors ran 3.8 percentage points a year behind, while each of the four institutional groups gained. The transfer came to 2.2% of the country's GDP. Almost all of the individuals' losses traced to their aggressive orders — the ones that cross the spread to get filled now.
Read it carefullyThis is a whole-market accounting identity made visible: for every trade there is another side, and the other side is usually better resourced.
−4.7%
twenty days after the crowd piles in
When Robinhood users bought a stock in unusual concentration — the top 0.5% of daily jumps in the number of accounts holding it — the average abnormal return over the following twenty days was −4.7%. For the most extreme episodes, where the holder count rose by more than 750% in a day, it was −19.6% over the following month. This is the platform whose indicator list set the scope of this site.
Read it carefullyAn average across thousands of episodes, not a rule about any one stock. It says crowding is a cost, not that every popular stock falls.
A thousand people who did not quit
The Brazilian study is the one worth sitting with, because it selects for exactly the behaviour every trading course recommends: showing up every day for more than a year.
Persistence did not filter for skill. Among traders with more than 300 days of experience, the ones who had been losing were the most likely to still be trading twelve months later — a 95.3% chance of coming back, in the Taiwanese data. Continuing is not evidence of learning.
Two ways the numbers usually get hidden
Both of these are reasons a losing activity can look like a winning one from inside it.
What this page does not say
- It does not say markets are unbeatable. The Taiwanese study found real, persistent skill — it also found it in under 1% of participants, which is a statement about prior odds, not about possibility.
- It does not say investing is the same as trading. Every figure here is about frequent, short-horizon, self-directed trading. Buying a diversified fund and holding it for thirty years is a different activity with a different and far better-documented outcome.
- It does not say the indicators on this site are worthless. Most of them are descriptions of what already happened, and they describe it accurately. The claim being tested is the predictive one.
- It is not current to this month. These samples run from 1991 to 2020 across three countries. Zero commissions, fractional shares and options access have changed the costs since — some for the better, some for the worse.
- It is not a reason to trade on feel instead. Nothing here suggests that abandoning method improves the odds. It suggests the odds are the thing to plan around.
This page sits first because of what it costs to put it last. Everything that follows is a vocabulary for talking about price, and a vocabulary is much easier to acquire than an edge. If a reference like this one leaves you more fluent and equally likely to lose money, it has taught you to sound like a trader rather than to think like one — and the number above is the difference between those two outcomes.