Chapter 9 Further Reading: Survivorship Bias — The Most Dangerous Lie Statistics Tell

Foundational Texts

1. Taleb, Nassim Nicholas. Fooled by Randomness: The Hidden Role of Chance in Life and in the Markets (2001). Random House.

Taleb's first book is the most passionate and arguably the most important treatment of survivorship bias for a general audience. His central concept of the "silent cemetery" — the traders and entrepreneurs who tried the same strategies as the visible winners and failed — is the intellectual ancestor of the invisible graveyard metaphor. Taleb's specific focus is financial markets, but his argument ranges widely. Uncompromising and sometimes abrasive, but essential. Read alongside Kahneman for a more empirically grounded version of similar ideas.

2. Kahneman, Daniel. Thinking, Fast and Slow (2011). Farrar, Straus and Giroux.

Chapter 24 ("The Engine of Capitalism") addresses survivorship bias in business directly, with reference to Jim Collins' Good to Great as an example of how business literature generates survivorship-biased lessons. Kahneman is diplomatically but unmistakably critical of the genre. The broader book provides the cognitive framework for understanding why survivorship bias is so hard to detect and correct: our brains are built to process vivid available stories, not statistical base rates.

3. Mlodinow, Leonard. The Drunkard's Walk: How Randomness Rules Our Lives (2008). Pantheon Books.

Several chapters address survivorship bias in Hollywood, publishing, and financial markets. Mlodinow's treatment of how hit movies and bestselling books are produced is especially illuminating — success in entertainment is heavily influenced by random factors, and the advice ecosystem (what makes a hit?) is entirely survivorship-biased. His discussion of how investment fund track records are distorted by survivorship is one of the clearest available.

Key Research

4. Ghosh, Shikhar (2012). "The Venture Capital Secret: 3 Out of 4 Start-Ups Fail." Wall Street Journal, September 20, 2012.

Ghosh's analysis of over 2,000 VC-backed startups found failure rates far higher than commonly acknowledged in the startup advice ecosystem. The article is relatively short but points to the magnitude of the invisible graveyard in venture-backed entrepreneurship. Reading it alongside any popular startup success book is illuminating.

5. Dale, S.B., & Krueger, A.B. (2002). "Estimating the Payoff to Attending a More Selective College: An Application of Selection on Observables and Unobservables." Quarterly Journal of Economics, 117(4), 1491–1527.

The landmark research comparing students admitted to elite universities who attended them vs. students admitted who chose to attend elsewhere. This paper provides the empirical basis for the chapter's claim that elite university outcomes are largely selection effects. Technical but readable in the introduction and conclusion sections. A model of how to correctly account for selection effects in educational research.

6. S&P Dow Jones Indices. SPIVA U.S. Scorecard (published annually).

The most comprehensive data source on active vs. passive fund performance, with explicit survivorship bias correction methodology. Available free at spglobal.com. The scorecard shows, year after year, that the vast majority of actively managed funds underperform passive indexes over long periods, even after accounting for survivorship bias in the dataset. Essential reading for anyone interested in investing.

7. Ioannidis, J.P.A. (2005). "Why Most Published Research Findings Are False." PLOS Medicine, 2(8), e124.

Relevant both to Chapter 7 (small samples) and Chapter 9 (survivorship/publication bias). Ioannidis demonstrates that the combination of small samples, multiple comparisons, and publication bias (itself a form of survivorship — only the significant results get published) produces a scientific literature full of false findings. Free online. The most-cited paper in the methodology of science.

Applied Reading

8. Nisen, Max. "What People Get Wrong About the Mythology of Failure in Silicon Valley." Quartz, 2014.

A journalistic examination of how the startup failure narrative is itself survivorship-biased. Even the celebrated culture of "failing fast" and celebrating failure in tech is mostly visible among people who eventually succeeded — the truly failed founders are less likely to be the ones giving talks about the value of failure. A sharp piece about second-order survivorship bias.

9. Collins, James C. Good to Great: Why Some Companies Make the Leap — and Others Don't (2001). HarperBusiness.

This is on the list specifically as a well-intentioned example of survivorship-biased analysis. Collins identified companies that had exceptional long-term stock performance and compared them to less successful peers to identify what distinguished them. The methodology seems rigorous — there is a comparison group! But the "great" companies were selected after their performance was observable, creating hindsight bias. Many of the traits identified as causes of greatness were present in companies that later declined. Phil Rosenzweig's The Halo Effect (next entry) is the definitive critique.

10. Rosenzweig, Phil. The Halo Effect: ...and the Eight Other Business Delusions That Deceive Managers (2007). Free Press.

Rosenzweig's critique of the business success literature, including Good to Great and Tom Peters' In Search of Excellence, is the most rigorous available. He demonstrates that the research methods used in these celebrated books are contaminated by survivorship bias, hindsight bias, and the halo effect (judging everything about a successful company positively because their overall performance was positive). Essential reading for anyone who consumes business success literature, which is nearly everyone in professional contexts.

Online Resources

• CB Insights Startup Failure Post-Mortems: A continuously updated database of post-mortems from failed companies, often written by the founders themselves. An invaluable window into the invisible graveyard. Available at cbinsights.com.
• First Round Capital's "State of Startups" surveys: Annual reports that include data on the range of startup outcomes, not just successes. More representative than individual success stories.
• The Abraham Wald story primary sources: Wallis, W.A. (1980). "The Statistical Research Group, 1942-1945." Journal of the American Statistical Association, 75(370), 320-330. The account of Wald's wartime statistical work, including the aircraft armor analysis, in its historical context.