Definition
Analysis samples only visible winners while ignoring the failures that disappeared.
Example
A trader analyses only the hedge funds still operating after a turbulent decade and concludes that "consistent outperformance is achievable," overlooking the large number of funds that closed due to poor returns or excessive risk-taking. The visible sample does not represent the full reality.
Cognitive Driver
The mind anchors to what is observable. Failures disappear from the dataset, creating an illusion of higher success rates. Without deliberate inclusion of missing cases, probability estimates become biased toward optimism and understate downside risk.
Market Expression
Strategies are assessed based on visible winners instead of the entire distribution. Historical analogues are chosen selectively because the unsuccessful cases are forgotten or unavailable. Performance expectations drift upward as weak or failed examples are excluded.
Trigger Conditions
- Markets with high churn or strategy turnover
- Thematic trades with widely publicised winners
- Environments where failed products or funds quietly disappear
- Periods following large rallies where survivors dominate narratives
- Limited transparency into datasets of closed or underperforming entities
Diagnostic Markers
- Analysis relies on top decile performers without acknowledging attrition
- Backtests exclude delisted assets or failed strategies
- Position rationales reference visible winners but ignore failed peers
- Underestimation of tail risks due to missing negative examples
- Assumptions built on incomplete historical distributions
Cost Profile
- Systematic underpricing of risk
- Overexposure to strategies with high failure rates
- Misjudgment of expected returns and drawdowns
- Fragile portfolio construction built on selectively positive histories
- Poor scenario analysis due to missing left-tail cases
Differentiation From Adjacent Biases
- Not availability bias: availability is about salience; survivorship is about missing data.
- Not confirmation bias: survivorship hides failures; confirmation selects supportive evidence.
- Not outcome bias: survivorship concerns dataset completeness, not decision evaluation.
Corrective Lens
Incorporate delisted assets, failed strategies, and abandoned trade ideas into datasets and reviews. Assess performance using the full distribution, not just visible survivors. Use tooling that highlights null or negative cases to restore realistic expectations of risk, return, and durability.