Overconfidence and overtrading: the research on trading too much

Overconfidence is believing that your knowledge, skill or forecasts are better than they really are. It is one of the most robust findings in psychology, and in investing it has a clear, measurable symptom: trading too much. The more confident investors are in their own judgment, the more often they buy and sell, and the research on millions of real trades says that, on average, the extra trading costs them money.

This article explains the forms overconfidence takes, walks through the main studies of individual investors and day traders with their actual numbers, and shows how to measure your own trading against a benchmark. It goes beyond the short overview in Trading psychology: loss aversion, overconfidence and other biases and the practical list in Common trading mistakes and the math behind them.

Three faces of overconfidence

Miscalibration is being more certain than the evidence allows. Ask people for a range they are 90% sure contains an unknown quantity, such as the length of a river or next year's market return, and their ranges miss far more often than 10% of the time. Reviews of calibration research since the 1970s, such as Sarah Lichtenstein, Baruch Fischhoff and Lawrence Phillips's 1982 survey, found this pattern in many settings. Professionals are not immune. In a study of thousands of forecasts collected from US chief financial officers, Itzhak Ben-David, John Graham and Campbell Harvey reported in 2013 that the realized S&P 500 return fell inside the executives' 80% confidence ranges only about 36% of the time.

The better-than-average effect is rating your own skill above the median. In Ola Svenson's 1981 survey, 93% of the American students asked put themselves in the top half of drivers for skill. Investors behave the same way about stock picking: almost everyone who trades actively believes their picks are better than average, which cannot be true for the group as a whole.

Illusion of control is believing you can influence outcomes that are mostly luck. Ellen Langer's experiments in the 1970s found that people valued lottery tickets they had chosen themselves more than tickets handed to them. Screens full of live data, charts and order buttons create a strong sense of control over prices that no individual trader moves.

Two other biases keep overconfidence alive. Self-attribution credits wins to skill and blames losses on bad luck, and hindsight bias makes past outcomes look predictable. Simon Gervais and Terrance Odean modeled the result in 2001: traders who attribute early successes to their own ability become more overconfident precisely after a run of luck.

The evidence from brokerage accounts

Trading is hazardous to your wealth

The landmark study is Brad Barber and Terrance Odean's 2000 paper in the Journal of Finance. They analyzed the accounts of 66,465 households at a large US discount broker from 1991 to 1996 and sorted them into five groups by turnover, the share of the portfolio bought and sold each year. The average household turned over about 75% of its portfolio a year. The most active fifth turned over more than 250%.

The average household earned 16.4% a year after costs against 17.9% for a value-weighted market index. The most active fifth earned 11.4%. Before costs, the active traders' picks were roughly as good as everyone else's. The gap came from the cost of trading so often: commissions, which were much higher in the 1990s, and the bid-ask spread paid on every round trip. The authors' explanation was overconfidence: investors traded as if their information were more valuable than it was.

Buying the wrong stocks, too

Trading costs are not the whole story. In a 1999 paper Odean followed the trades of 10,000 accounts at a discount broker from 1987 to 1993 and compared what investors bought with what they sold to raise the money. Over the following year, the stocks they bought underperformed the stocks they sold by roughly 3 percentage points, even before counting costs. Switching was not just expensive; on average it moved money into worse positions.

Who trades most

In 2001 Barber and Odean used gender as a natural experiment, since psychology research finds men more overconfident than women in areas such as finance. In more than 35,000 households, men traded about 45% more than women, and trading reduced men's net returns by about 2.65 percentage points a year against 1.72 for women. The difference was larger between single men and single women.

Other studies point the same way. In 2002 Barber and Odean followed about 1,600 investors who switched from phone-based to online trading in the 1990s. Before switching, they had beaten the market, which likely made them confident. After going online they traded more, more speculatively and less profitably. In Finland, where detailed records exist, Mark Grinblatt and Matti Keloharju found in 2009 that men who scored as more overconfident on military psychological tests, and men with more speeding tickets, a proxy for sensation seeking, traded more often.

Day traders

Day trading is the purest test of trading skill, and the results are sobering. Using complete trading records from Taiwan, Barber, Yi-Tsung Lee, Yu-Jane Liu and Odean found in 2014 that fewer than 1% of day traders were able to earn predictable profits after fees, even though a small group of skilled traders did exist. A 2019 working paper by Fernando Chague, Rodrigo De-Losso and Bruno Giovannetti studied Brazilians who day traded index futures for more than 300 days: about 97% lost money, and only around 1% earned more than the Brazilian minimum wage. Both studies cover specific markets and periods, but no comparable study has found the opposite. See Day trading vs swing trading vs position trading compared for how trading styles differ.

The cost of trading in 2026

US online brokers cut stock commissions to zero in 2019, and many readers assume that removed the problem Barber and Odean measured. It removed only part of it.

• Spreads and price impact. Every round trip still crosses the bid-ask spread, which is small for large ETFs but can be large for small stocks and options. See Liquidity and slippage: volume, depth, spread and market impact.
• Taxes. In a US taxable account, gains on positions held one year or less are taxed at ordinary income rates, higher than long-term rates for most people (IRS Topic 409; see Capital gains tax on stocks: 2026 rates, holding periods and basis). Frequent trading turns potential long-term gains into short-term ones.
• Timing. Odean's finding that bought stocks lag sold ones is a cost unrelated to commissions.
• Leverage and options. Active traders are more likely to use margin and short-dated options, which multiply the effect of mistakes.

The figure shows what a steady drag does over time. It is an illustration, not a measurement: the right drag for any trader depends on turnover, spreads and taxes.

At 7% a year before costs, $10,000 grows to about $76,000 in 30 years. A 1.5% annual drag, which a high-turnover trader can reach through spreads and taxes alone, cuts that to about $50,000. A 3% drag leaves about $32,000, less than half.

Measuring your own overconfidence

Overconfidence is hard to see from the inside, so the defenses rely on numbers.

1. Compute your turnover. Add up the value of everything you sold over the past year and divide by your average portfolio value. Above 100% means you replaced your whole portfolio at least once.
2. Compare with a benchmark. Track your portfolio's return against a simple alternative, such as an S&P 500 fund, over the same dates and after all costs. If the active portfolio trails over several years, the trading is not adding value.
3. Log forecasts with probabilities. Before a trade, write what you expect and how sure you are, for example "70% chance this is higher in six months." After enough entries, check whether your 70% calls come true about 70% of the time. Most people find they are less accurate than they felt. A Trading plan and trading journal: what to write down is the natural place for this.
4. Run a pre-mortem. Before acting, imagine the trade has failed badly and write the most likely reasons. The exercise, popularized by the psychologist Gary Klein, surfaces risks that confidence hides.
5. Separate skill from luck. A handful of wins proves little. A rule tested on years of prices, as in a backtest, is better evidence than recent results, though a backtest has its own traps.

Limiting trading in practice

• Set a trade budget, such as a maximum number of trades a month, and treat it like a spending limit.
• Split the portfolio. A core of diversified funds that is rarely touched and a small "active" sleeve with a fixed size contain the damage from overconfidence while leaving room to learn. See How to build your first portfolio: a step-by-step beginner outline.
• Raise the bar for a switch. Before selling A to buy B, require a written reason why B is better by more than the cost and tax of switching.
• Make trading slower. Removing trading apps from your phone, or placing orders only at set times, adds friction where impulse lives.

How Finabase helps

Finabase is a research app and does not place trades, which is itself a useful separation between research and execution. The backtester runs an indicator or candlestick rule over real prices and lists every trade, a check on whether a strategy works beyond your last few results. The comparison tool puts a stock or your favorite fund next to a broad index over the same period, and a portfolio records cost basis and gains so you can compute turnover and compare with a benchmark honestly.

Frequently asked questions

What is overconfidence bias in investing?

Overconfidence bias is overestimating your own knowledge, skill or forecasting accuracy. For investors it typically means trading more often, holding fewer stocks and taking larger positions than the quality of their information justifies.

Does trading more frequently lower returns?

On average, yes. In Barber and Odean's study of 66,465 US households from 1991 to 1996, the most active fifth earned 11.4% a year after costs against 17.9% for the market, while the average household earned 16.4%. Lower costs today shrink the gap but do not close it.

Can anyone make money day trading?

A few people do, but studies of complete records find very few. In Taiwan, fewer than 1% of day traders earned predictable profits after fees; in Brazil, about 97% of those who persisted for more than 300 days lost money.

How do I know if I am overtrading?

Compute your annual turnover and compare your after-cost returns with a simple index fund over the same period. High turnover combined with returns that trail the benchmark is the clearest sign.

Summary

Overconfidence makes investors believe their information and skill are better than they are, and its most measurable effect is trading too much. The research is consistent: the households that traded most earned several percentage points a year less than the market, the stocks investors bought did worse than those they sold, and very few day traders profit after costs. Zero commissions removed only one of the costs. The defenses are measurement and structure: know your turnover, compare with a benchmark, record forecasts and their outcomes, and make trading slower and rarer than impulse would like.