Survivorship Bias: The Tale of Forgotten Failures
Survivorship bias is a common logical error that distorts our understanding of the world. It happens when we assume that success tells the whole story and when we don’t adequately consider past failures.
There are thousands, even tens of thousands of failures for every big success in the world. But stories of failure are not as sexy as stories of triumph, so they rarely get covered and shared. As we consume one story of success after another, we forget the base rates and overestimate the odds of real success.
“See,” says he, “you who deny a providence, how many have been saved by their prayers to the Gods.”
“Ay,” says Diagoras, “I see those who were saved, but where are those painted who were shipwrecked?”
— Cicero
The Basics
A college dropout becomes a billionaire. Batuli Lamichhane, a chain-smoker, lives to the age of 118. Four young men are rejected by record labels and told “guitar groups are on the way out,” then go on to become the most successful band in history.
Bill Gates, Batuli Lamichhane, and the Beatles are oft-cited examples of people who broke the rules without the expected consequences. We like to focus on people like them—the result of a cognitive shortcut known as survivorship bias.
When we only pay attention to those who survive, we fail to account for base rates and end up misunderstanding how selection processes actually work. The base rate is the probability of a given result we can expect from a sample, expressed as a percentage. If you play roulette, for example, you can be expected to win one out of 38 games, or 2.63%, which is the base rate. The problem arises when we mistake the winners for the rule and not the exception. People like Gates, Lamichhane, and the Beatles are anomalies at one end of a distribution curve. While there is much to learn from them, it would be a mistake to expect the same results from doing the same things.
A stupid decision that works out well becomes a brilliant decision in hindsight.
— Daniel Kahneman
Cause and Effect
Can we achieve anything if we try hard enough? Not necessarily. Survivorship bias leads to an erroneous understanding of cause and effect. People see correlation in mere coincidence. We all love to hear stories of those who beat the odds and became successful, holding them up as proof that the impossible is possible. We ignore failures in pursuit of a coherent narrative about success.
Few would think to write the biography of a business person who goes bankrupt and spends their entire life in debt. Or a musician who tried again and again to get signed and was ignored by record labels. Or of someone who dreams of becoming an actor, moves to LA, and ends up returning a year later, defeated and broke. After all, who wants to hear that? We want the encouragement survivorship bias provides, and the subsequent belief in our own capabilities. The result is an inflated idea of how many people become successful.
The discouraging fact is that success is never guaranteed. Most businesses fail. Most people do not become rich or famous. Most leaps of faith go wrong. It does not mean we should not try, just that we should be realistic with our understanding of reality.
Beware of advice from the successful.
— Barnaby James
Survivorship Bias in Business
Survivorship bias is particularly common in the world of business. Companies which fail early on are ignored, while the rare successes are lauded for decades. Studies of market performance often exclude companies which collapse. This can distort statistics and make success seem more probable than it truly is. Just as history is written by the winners, so is much of our knowledge about business. Those who end up broke and chastened lack a real voice. They may be blamed for their failures by those who ignore the role coincidence plays in the upward trajectories of the successful.
Nassim Taleb writes of our tendency to ignore the failures: “We favor the visible, the embedded, the personal, the narrated, and the tangible; we scorn the abstract.” Business books laud the rule-breakers who ignore conventional advice and still create profitable enterprises. For most entrepreneurs, taking excessive risks and eschewing all norms is an ill-advised gamble. Many of the misfit billionaires who are widely celebrated succeeded in spite of their unusual choices, not because of them. We also ignore the role of timing, luck, connections and socio-economic background. A person from a prosperous family, with valuable connections, who founds a business at a lucrative time has a greater chance of survival, even if they drop out of college or do something unconventional. Someone with a different background, acting at an inopportune time, will have less of a chance.
In No Startup Hipsters: Build Scalable Technology Companies, Samir Rath and Teodora Georgieva write:
Almost every single generic presentation for startups starts with “Ninety Five percent of all startups fail”, but very rarely do we pause for a moment and think “what does this really mean?” We nod our heads in somber acknowledgement and with great enthusiasm turn to the heroes who “made it” — Zuckerberg, Gates, etc. to absorb pearls of wisdom and find the Holy Grail of building successful companies. Learning from the successful is a much deeper problem and can reduce the probability of success more than we might imagine.
Examining the lives of successful entrepreneurs teaches us very little. We would do far better to analyze the causes of failure, then act accordingly. Even better would be learning from both failures and successes.
Focusing on successful outliers does not account for base rates. As Rath and Georgieva go on to write:
After any process that picks winners, the non-survivors are often destroyed or hidden or removed from public view. The huge failure rate for start-ups is a classic example; if failures become invisible, not only do we fail to recognise that missing instances hold important information, but we may also fail to acknowledge that there is any missing information at all.
They describe how this leads us to base our choices on inaccurate assumptions:
Often, as we revel in stories of start-up founders who struggled their way through on cups of ramen before the tide finally turned on viral product launches, high team performance or strategic partnerships, we forget how many other founders did the same thing, in the same industry and perished…The problem we mention is compounded by biographical or autobiographical narratives. The human brain is obsessed with building a cause and effect narrative. The problem arises when this cognitive machinery misfires and finds patterns where there are none.
These success narratives are created both by those within successful companies and those outside. Looking back on their ramen days, founders may believe they had a plan all along. They always knew everything would work out. In truth, they may lack an idea of the cause and effect relationships underlying their progress. When external observers hear their stories, they may, in a quasi-superstitious manner, spot “signs” of the success to come. As Daniel Kahneman has written, the only true similarity is luck.
Consider What You Don’t See
When we read about survivorship bias, we usually come across the archetypical story of Abraham Wald, a statistician studying World War II airplanes. His research group at Columbia University was asked to figure out how to better protect airplanes from damage. The initial approach to the problem was to look at the planes coming back, seeing where they were hit the worst, then reinforcing that area.
However, Wald realized there was a missing, yet valuable, source of evidence: Planes that were hit that did not make it back. Planes that went down, that weren’t surviving, had much better information to provide on areas that were most important to reinforce. Wald’s approach is an example of how to overcome survivorship bias. Don’t look just at what you can see. Consider all the things that started on the same path but didn’t make it. Try to figure out their story, as there is as much, if not more, to be learned from failure.
Considering survivorship bias when presented with examples of success is difficult. It is not instinctive to pause, reflect, and think through what the base rate odds of success are and whether you’re looking at an outlier or the expected outcome. And yet if you don’t know the real odds, if you don’t know if what you’re looking at is an example of survivorship bias, then you’ve got a blind spot.
Whenever you read about a success story in the media, think of all the people who tried to do what that person did and failed. Of course, understanding survivorship bias isn’t an excuse for not taking action, but rather an essential tool to help you cut through the noise and understand the world. If you’re going to do something, do it fully informed.
To learn more, consider reading Fooled By Randomness, or The Art of Thinking Clearly.
from Hacker News https://ift.tt/2DB6f7B