One of the most consistently documented findings in retail investing is the negative correlation between trading frequency and returns. Investors who trade often produce lower returns, on average, than investors who trade rarely, holding portfolio characteristics roughly constant. The mechanism is not simple friction from transaction costs (though those matter); the deeper mechanism is overconfidence — the specific behavioural pattern that produces high-frequency trading in the first place.
The empirical foundation
Terrance Odean and Brad Barber's 2000 paper "Trading Is Hazardous to Your Wealth" analysed roughly 66,000 US discount-brokerage accounts over six years. Their central finding: the accounts that traded most actively produced returns approximately 6.5 percentage points per year below the accounts that traded least actively. The gap was not primarily explained by transaction costs (though those contributed); the underlying stock selection of frequent traders was also worse than the selection of infrequent traders.
Subsequent studies in Taiwanese, Chinese, Finnish, and many other markets have replicated the pattern. The relationship between trading frequency and returns is one of the most consistent findings in behavioural finance.
The overconfidence mechanism
Why do some investors trade frequently? Not primarily because they need to. Retail investors have no institutional pressure to be active. They trade because they believe they can identify profitable trades — that is, they are confident in their ability to make short-term directional calls.
The problem is that this confidence is usually not warranted. The evidence is unambiguous: most retail traders cannot consistently produce alpha through short-term trading. The confidence that produces the trading activity is genuine but is not calibrated to actual skill.
This is the essence of overconfidence as a behavioural pattern. It is not simply "thinking too highly of yourself." It is specifically the disconnect between subjective confidence and objective calibration. Studies of investor self-assessment consistently find that individuals rate their own investment skills well above the empirical distribution — 80% of retail investors, in various surveys, believe they perform "above average," which is mathematically impossible.
The demographic patterns
The overconfidence-trading pattern is not uniform across demographics. Barber and Odean's follow-up work found that men trade more frequently than women, produce lower returns than women, and rate their own investment skills higher than women. The gap between subjective confidence and objective performance is systematically larger for men than for women in retail investing datasets.
Similarly, younger investors trade more frequently than older investors and produce lower returns. This is not purely a "learning curve" phenomenon — the underperformance persists even when controlling for time in market. It appears to be genuinely correlated with the demographic patterns of overconfidence.
Why the pattern persists
Overconfidence in investing does not extinguish easily because the feedback loop is weak. Successful trades produce vivid, memorable positive feedback ("I called that one right"). Unsuccessful trades produce less-memorable negative feedback that is easily reframed ("that was just bad luck" or "I would have been right but for the news"). The asymmetric memory of trading outcomes systematically reinforces overconfidence over time.
This is one of the reasons why aggregate retail trading frequency does not appear to be declining over decades despite the accumulated evidence that it is costly. Each generation of retail investors seems to learn the lesson somewhat but does not learn it deeply enough to change behaviour materially.
The zero-commission complication
The 2019 elimination of trading commissions at most major US retail brokerages removed one of the friction sources that had partly constrained trading frequency. Since 2019, aggregate retail trading has risen substantially. Options trading in particular has grown to unprecedented levels among retail participants.
Studies of the post-2019 retail environment have generally found that the aggregate retail underperformance has widened, not narrowed. The removal of the visible cost has encouraged more trading, and the more trading has produced worse aggregate outcomes — exactly the pattern Barber-Odean would predict.
What actually helps
Three practices measurably reduce the costs of overconfidence.
Impose trading friction. Even if commissions are zero, imposing your own friction — a required waiting period before executing a trade decision, a formal write-up requirement for any new position, a mandatory review with a partner or advisor — reduces the impulsive component of trading. The friction does not need to be large; even small delays produce measurable reductions in trading frequency and in the losses associated with impulsive decisions.
Track outcomes with brutal honesty. Maintain a decision journal that records both the decision and the outcome, with regular reviews. The friction of writing down decisions in advance and evaluating them dispassionately later provides feedback that partially counteracts the asymmetric memory pattern.
Reduce exposure to trading-encouraging content. Financial media, social media investment communities, and platform-embedded trading nudges (real-time alerts, streak counters, gamified interfaces) all systematically encourage more trading. Reducing exposure to these environments is one of the most effective single interventions for retail investors trying to improve their returns.
The paradox of expertise
An interesting complication is that some traders genuinely do have skill — the aggregate pattern of retail trading being costly does not mean no individual can produce alpha. The problem is that the individuals who genuinely have skill and the individuals who merely believe they have skill both trade frequently. Distinguishing between them in real time is nearly impossible from the inside.
The empirical way to distinguish is by rigorous long-term performance tracking against appropriate benchmarks. Someone who has genuinely outperformed their benchmark over multiple years across different market conditions has some case for having identifiable skill. Someone who cannot make that case is more likely to be exhibiting overconfidence than skill.
The base rate of "genuinely skilled" retail traders is very low. Statistical analyses generally find that a small single-digit percentage of active retail traders produce risk-adjusted returns above market benchmarks over 5+ year windows. This is not a criticism of the majority; it is a description of the empirical distribution.
The rule to internalise
The correlation between trading frequency and returns for retail investors is strongly negative, and the mechanism is overconfidence rather than any specific analytical failing. The most powerful single change most retail investors can make to their long-run returns is to trade less. This is not a message about skill or ability; it is a message about the empirical patterns of retail trading behaviour and about the specific behavioural mechanism that produces frequent trading in the first place. Building practices that impose friction on trading decisions is one of the highest-return interventions available to retail investors.
Educational content only. Not investment advice.