Self-Attribution Bias

Investors exhibit self-attribution bias when they ascribe profitable trades to their own skill and acumen, while attributing losses to bad luck, market shocks, or other factors beyond their control. This asymmetry distorts self-perception and drives costly repeat behaviour.

The asymmetry of credit and blame

Imagine two traders. Trader A buys a small biotech stock ahead of positive clinical trial results and sells for a 40% gain. Trader B buys the same stock at a similar valuation and sells for a 10% loss when the company misses a secondary endpoint. Both face uncertainty; both made reasonable bets. Yet Trader A is likely to attribute their win to superior research, earlier access to information, or better intuition. Trader B is likely to blame the company’s management or an unforeseen competitive development.

If Trader A had lost 10%, they might blame bad luck or note that the timing was just off—bad timing that could have happened to anyone. If Trader B had gained 40%, they might credit their contrarian thesis or risk tolerance. Most people don’t consciously flip their narrative this way; rather, the mind naturally gravitates toward explanations that preserve self-image. Wins are skill; losses are circumstance.

This is not identical to simple overconfidence, though it often produces overconfident behaviour. Self-attribution bias is specifically about asymmetric causal reasoning: the same outcome gets a different explanation depending on whether it favours or hurts the ego. The bias operates largely outside conscious awareness.

How the bias perpetuates underperformance

Self-attribution bias creates a vicious cycle. After a profitable quarter, traders become more confident in their strategy and increase position sizes or trading frequency. The logic seems sound: they’ve proven they can pick winners, so doubling down is justified. But if their outperformance was partly luck—and statistical evidence strongly suggests it usually is—increased risk-taking on the back of a lucky run is irrational.

Many studies of mutual fund managers show that expense ratios and trading costs consistently drag actively managed returns below index funds. Yet most fund managers and their clients attribute their underperformance to bad luck or unsuitable market conditions, not to structural inability to beat benchmarks after fees. Some managers close underperforming funds, then start new ones with clean track records, attributing the reset to a better strategy rather than regression to the mean. The cycle repeats.

For individual traders, self-attribution bias is especially destructive because it combines with the house edge. Markets are zero-sum or negative-sum (after costs and taxes) for the retail participant. A trader outperforming in a given year is more likely lucky than skilled. But the winner attributes the win to skill, ramps up risk, and then suffers a regression. Losses that follow are blamed on bad luck or bad market timing, not on the fact that taking outsized risks doesn’t pay over time.

The illusory control component

Self-attribution bias is often bundled with a related bias: illusory control. Investors overestimate their influence on outcomes they can partially influence. A stock rises after you buy it, and you feel you read the market correctly—even though thousands of other buyers were acting on the same information. A position you hold closely monitored outperforms the market; you feel your vigilance made the difference—even though you can’t actually control the company’s decisions or broader market sentiment.

This illusion of control is reinforced by modern trading technology. Real-time pricing, instant execution, and the ability to transact with one click create an impression of agency and skill. The simplicity of placing a trade feels like mastery, when in fact execution has become trivial; the hard part—knowing what to trade at what price—remains genuinely difficult and success-dependent on information the individual trader likely does not have.

Evidence from professional and retail data

Studies of professional investors reveal the bias vividly. Successful fund managers who had strong performance in one period often underperform in the next, contradicting their self-attribution narrative. Yet they, and their clients, rarely update the belief that they possess special insight. Fees continue as though past outperformance proves skill. When a skilled manager retires, their successor often underperforms—suggesting the value was tied to luck or temporary conditions, not to systematic skill that transferred to the firm.

Retail trading data shows similar patterns. Accounts that experience large gains in a year tend to increase turnover and risk-taking in the next year, then underperform. The winners think their luck reflects skill and bet accordingly. Data on day-traders shows the majority lose money after costs, yet surveys reveal most believe they will eventually beat the market. Self-attribution bias keeps them in the game despite mounting evidence.

Distinguishing skill from luck in practice

Identifying true skill in investing is genuinely hard. A manager could outperform for a decade and still be lucky, especially in a noisy market. However, several patterns improve the odds of spotting real skill:

• Consistency of edge: Did the manager outperform in up and down markets, or primarily when their style favoured? Styles rotate, and a manager who performs well only in narrow conditions may lack broad skill.
• Out-of-sample performance: Did past success predict future success, or did outperformance fade? Reversion to the mean is the default hypothesis.
• Explicit hypothesis: Can the manager articulate a specific, testable reason for expected outperformance, or is the narrative vague? Vagueness often masks luck.
• Risk-adjusted returns: Did the outperformance come from accepting tail-risk without being compensated for it, or from genuine superior risk-adjusted returns?

Most investors lack the discipline or data access to run such analyses on themselves. Self-attribution bias steps in to fill the void: the mind attributes success to skill, regardless of evidence.

Combating the bias through structural change

Unlike some biases that fade with awareness, self-attribution bias is stubborn because it is ego-driven. Telling a trader they’re lucky is met with defensive scepticism. Structural defences work better.

One approach is to establish clear, measurable benchmarks and track performance versus them ruthlessly. If your strategy is to beat the S&P 500 after costs, and you consistently underperform by 1% annually, the data forces a reckoning that gut attribution does not. Similarly, separating the role of decision-maker from the role of outcome evaluator helps: have an independent advisor review your trades before knowing the result, to assess whether the idea was sound, separate from whether it profited.

Limiting portfolio turnover by force—say, a commitment to hold every position for at least two years—reduces the feedback frequency that feeds overconfidence. With longer holding periods and fewer trades, the role of luck becomes more obvious; short-term trading can feel skillful because noise is mistaken for signal.

The professional poker analogy

Professional poker players, despite also playing a game involving luck, develop better intuitions about skill and variance than most investors. Many use precise statistical language: “I made the mathematically correct bet; whether I won this hand doesn’t validate my decision.” Good poker players update beliefs after many hands, not one. They separate decision quality from outcome quality.

Investors would benefit from similar discipline. The correct framework is: “Was the decision rational given available information?” not “Did it make money?” Good decisions sometimes lose; bad decisions sometimes win. Conflating the two is the core error of self-attribution bias.

See also