Confirmation bias in investing: how seeking agreement undermines portfolio decisions

Confirmation bias is the tendency to seek, interpret, and recall information in ways that reinforce existing beliefs. In investing, it distorts the research process—leading investors to build conviction in positions they already hold while systematically discounting evidence that challenges them.

What confirmation bias is

The tendency was documented in cognitive psychology by Peter Wason (1960) in On the Failure to Eliminate Hypotheses in a Conceptual Task, Quarterly Journal of Experimental Psychology, in his selection task experiments, which showed that people systematically seek confirming evidence rather than evidence that could falsify their hypotheses. In investment contexts, it operates through two channels. Selective exposure refers to the tendency to consume information that is likely to confirm an existing view: following commentators who share one's thesis, seeking analyst reports that reinforce a buy case, gravitating toward news that supports a position already held. Selective interpretation refers to weighting confirming information more heavily than disconfirming information even when both are presented simultaneously.

Confirmation bias is reinforced by social environments. Investors who discuss their ideas primarily within communities of like-minded participants—online forums, investment groups, social networks oriented around specific strategies or sectors—receive disproportionate confirming feedback. The portfolio consequences of a strongly held and socially reinforced view can be severe, particularly when the view supports a concentrated position.

Unlike some other behavioural biases, confirmation bias is difficult to self-diagnose. An investor who is in the grip of confirmation bias typically believes they are engaging in genuine research; the selectivity is largely unconscious. This makes structural interventions more effective than awareness alone.

How it manifests in investing

The most direct expression is biased research. An investor who has formed a preliminary view—that a particular sector is attractively valued, that a specific company has been overlooked by the market—will tend to seek evidence that supports that view. Evidence pointing to regulatory risk, deteriorating fundamentals, or a weakening competitive position is acknowledged and then rationalised rather than integrated. The research process becomes post-hoc justification rather than genuine inquiry.

Confirmation bias also explains why investors hold losing positions for longer than evidence warrants. Having invested in a thesis, they continue to seek information suggesting it remains intact—interpreting short-term setbacks as temporary, treating positive indicators as confirming the original view and negative indicators as noise. The position is maintained not because the balance of evidence supports it, but because the investor's information environment has been filtered toward confirming evidence.

A third manifestation is resistance to systematic signals that contradict a discretionary conviction. An investor running a rules-based strategy who strongly believes their current market view is correct may override a sell signal as an artefact of the model rather than as evidence about the underlying asset. The model is held to a higher evidential standard than the investor's own conviction.

The cost

The cost of confirmation bias is difficult to isolate from related biases, but it contributes to the broader pattern of active investor underperformance documented in the literature. Daniel, Hirshleifer, and Subrahmanyam (1998), Investor Psychology and Security Market Under- and Overreactions, Journal of Finance, proposed a model in which investor overreaction to private information and underreaction to public information—both consistent with confirmation bias—produce systematic momentum and reversal patterns in asset prices. The implication is that confirmation bias is not only costly to the individual investor holding a biased position, but is also a source of exploitable mispricing in aggregate.

What helps

Systematic strategies that derive buy and sell signals from rules applied consistently to available data are structurally resistant to confirmation bias. The signal is generated by a methodology, not by the investor's interpretation of information through the lens of an existing position. The investor's conviction has no channel through which to filter the inputs or override the output.

For the research and portfolio review process, pre-commitment to exit criteria provides a specific structural check: if the conditions for exiting a position are defined before entry, the investor cannot later rationalise away evidence that those conditions have been met. The threshold is fixed; confirmation bias cannot raise it after the fact.

Confirmation bias in pfolio

pfolio's systematic strategies remove the research and conviction formation process from the portfolio decision loop. Signals are generated by applying rules to price and volatility data—there is no step at which the investor evaluates information and decides whether it confirms or challenges an existing view. The methodology specifies what constitutes a buy or sell signal; confirmation bias has no input it can selectively filter. Investors who wish to verify that signals apply consistently regardless of their views on specific markets can review the methodology at how we build portfolios.

Related articles

• Overconfidence in investing: why most active investors underperform the market
• Action bias in investing: why doing nothing is often the right move
• The disposition effect: why investors sell winners too early and hold losers too long
• Momentum investing explained: how price trends generate systematic returns