Psychology & Behavior
Section 1
Core Idea
Publication Bias is the systematic tendency for studies with positive, significant, or novel results to get published while studies with null, negative, or replicative results remain in the file drawer. The result is a distorted evidence base: the published literature overrepresents what works and underrepresents what doesn't. In business, the same dynamic operates through case studies, success stories, press coverage, and conference talks — creating a skewed map of what strategies actually work. Founders read about the companies that succeeded with a particular playbook but never hear about the hundreds that tried the same approach and failed. Publication Bias means that the more you read, the more confidently wrong you can become — because the evidence you're consuming has already been filtered for the answer you want.
Free playbook
Get The Business Model Playbook
58 business models, one visual page each: how the money flows, the metrics that matter, and who runs it. Free when you join the Faster Than Normal email.
Free. No spam. Unsubscribe anytime.
Section 2
How to See It
Strategy Research
You're seeing it when a team builds strategy around published case studies and best practices without asking: "How many companies tried this and failed?" The case studies that get published are the survivors; the failures never made it to print.
A/B Testing
You're seeing it when a company only shares winning experiment results internally, creating a false impression of test success rates. The experiments that showed no effect or negative results are quietly buried, distorting the team's calibration.
Section 3
How to Use It
Actively seek disconfirming evidence and null results. When evaluating a strategy, ask: "What's the base rate of success for this approach?" not just "Who succeeded with it?" Build internal cultures that celebrate negative findings — knowing what doesn't work is as valuable as knowing what does. Require pre-registration of experiments so results can't be selectively reported.
Decision filter
"Am I seeing the full evidence base — or only the results that were interesting enough to publish, share, or discuss?"
As a founder
Create an internal "failure library" — a shared record of experiments, initiatives, and strategies that didn't work. This counterbalances publication bias by ensuring negative results inform future decisions. The companies that learn fastest are the ones that give null results the same visibility as wins.
Section 5
Founders & Leaders
Jim SimonsFounder of Renaissance Technologies
Simons built the most successful quantitative fund in history by systematically correcting for publication bias in financial research. Renaissance didn't trust published trading strategies because Simons understood that the strategies that made it into journals were a biased sample — overfitted to historical data and selected precisely because they showed positive results. Instead, his team generated and tested their own hypotheses against raw data, treating published findings as starting points to be rigorously stress-tested rather than trusted at face value. For founders, Simons demonstrates that in any domain where results are selectively reported, building your own evidence base from primary data is more reliable than consuming the curated, bias-filtered conclusions of others.
Section 7
Connected Models
Reinforces
Survivorship Bias
Survivorship Bias focuses on winners while ignoring losers. Publication Bias is its information-layer equivalent — focusing on positive results while ignoring null ones. Together they create a world where everything you read and every company you study gives you a misleadingly optimistic picture.
Pairs-with
Confirmation Bias
Confirmation Bias makes you seek evidence that supports your existing beliefs. Publication Bias ensures that supportive evidence is disproportionately available. The pairing is insidious: you're biased toward confirming evidence, and the evidence ecosystem is biased toward providing it.
Tension
Replication Crisis
The Replication Crisis is partly a consequence of Publication Bias — when only positive results get published, many turn out to be false positives upon replication. The tension: publication bias creates the literature that the replication crisis reveals to be unreliable.
Section 8
One Key Quote
— John Ioannidis"Most published research findings are false."
Section 11
Summary & Further Reading
Publication Bias filters the evidence base toward positive and novel results, hiding the null findings and failures that would give a complete picture. In business, it distorts strategy by making successful playbooks look more reliable than they are. Counter it by seeking disconfirming evidence, tracking negative results, and building primary data rather than relying on curated success stories.