Survivorship Bias: The Trap Of Only Seeing Successful Individuals

Survivorship bias is a selective cognitive bias that occurs when we only see the successful cases in a dataset. We ignore the failures, resulting in distorted conclusions and faulty decision-making. It is especially prevalent in business, investing, history, science, and self-improvement. People overestimate the probability of success because they only study the winners. A successful subgroup is mistaken as the entire group. This is because the failure group is either invisible or ignored.

Examples come from various fields. Particularly the corporate world. Even though more than 90 percent of startups fail, entire degrees are dedicated to entrepreneurship. Dozens of students claim they will one day start a business and become successful. Looking at Steve Jobs, Bill Gates, and Mark Zuckerberg, its easy to get blinded by the flashing lights. We might conclude that to reach their level, we just need an idea, to drop out of school, and to start working on it full-time. But that’s not the case of course. We aren’t taking into account all the college dropouts who failed.

Simply put, many of us forget that unicorn startups are exactly that: unicorns. One-offs. Unique cases. Thousands of people follow analogical paths to success as these tycoons and fail. Their stories just aren’t as popular and widely shared. Thus we get the impression that our capabilities and potential achievements are far beyond reality. This isn’t to discredit hard work and talent, but that we, as a species, tend to ignore common failures and anchor onto successes to tell the tale. We kind of forget the whole luck, timing, networks, connections, and socioeconomic background side of things.

The problem with the survivorship bias is how often it occurs. When making decisions it is crucial to evaluate both the successes and failures. Survivorship bias is everywhere we look. It’s pervasive since it’s in our very nature to filter data by survival. Only the entities that make it through a selection process are considered. Those that died—or failed—are ignored. As they say, the victors write the pages of history. Visible successes are studied and publicized. Failures disappear and are not recorded or analyzed. The result? A misattribution between actions and outcomes and no account for missing data. In other words, terrible decisions. Correlation is not causation.

Where you might notice survivorship bias and how to handle it:

• Business and entrepreneurship: the illusion of guaranteed success. People believe that starting a business is a great way to become rich. This is because they only see lucrative stories like those of Elon Musk, Steve Jobs, Bill Gates, or Jeff Bezos. What they don’t see are the millions of failed startups that never made the news. Mistake: assuming that copying the habits of the successful will guarantee success. Reality: the failure rate of startups exceeds 90 percent. What to do: study both failed and successful businesses to get a comprehensive view of risks and pitfalls.

• Investing: chasing the winners. People assume that certain investing strategies always work because they look at surviving companies such as Apple, Amazon, or Tesla. What they don’t take into account are the countless failed companies that employed the exact same strategies but collapsed. Mistake: believing that past winners—stocks, funds, industries—will always perform. Reality: most businesses fail, and past asset performance does not say anything about future success. How to apply it: look at investment strategies across all companies, not just the winners.

• Self-help: the myth of the ultra-successful personal growth guru. People often read self-help books, interviews, and biographies of the most prosperous individuals and believe that walking in their footsteps will ensure the same success. Mistake: believing that if, for instance, successful people wake up at 5 am, you should too to succeed. Reality: many unsuccessful people also wake up at 5, and survivorship bias makes you tie the habit to prosperity. What you can do about it: look for common factors across success and failure to identify the real, standout causes of achievement.

• Military and engineering: learning from failure, not just success. During World War II, allied forces wanted to reinforce their planes by studying where surviving aircrafts had bullet holes. But there was a problem. Statistician Abraham Wald outlined it. Mistake: reinforcing the bullet holes of surviving planes. Reality: the planes that got hit in precisely the other areas never made it back—they didn’t land in the data sheet. How to apply it: look for what’s missing in your data, not just what is present; always study failures as well as successes.

The idea: recognize hidden variables that you might be overlooking. Don’t assume success. Look at failure. See it as a likely outcome. Not just a probability. Failure is, more often than not, the reality. Develop contingency plans. Actively seek out how you might not make it. Study what didn’t work, not just what did—those bullet holes. Don’t base your decisions solely on the visible winners. This will stop you from being overly optimistic, resulting in conclusions that skew your thinking: like the effect of only studying successful businesses and entrepreneurs. Many of those dropouts became successful despite dropping out, not because of it. Clean up your data. Take into account those errors.