Survivorship Bias

· 5 min read
Survivorship Bias


Survivorship bias or survival bias is the logical error of concentrating on entities that passed a selection process while overlooking those that did not. This can lead to incorrect conclusions because of incomplete data.

Survivorship bias is something I think a lot about. It's prevalent in so many of the non-fiction books I really enjoy reading, including, I suspect, in Quiet by Susan Cain that I've been reading for my book club.

It seems to me that authors have a hypothesis that they're trying to convince their readers is true, so they tell a lot of stories about successful people (the 'survivors') whose stories fit their hypotheses. But, as anybody with experience in life knows, there are so many stories of failure that look almost the same as stories of success, and what's truly interesting is not the elements common to stories of success (for example, if all the successful people drank coffee, we would hardly credit coffee for their success) but what is at least somewhat consistently present in the stories of success that is missing from the stories of failure.

The Wikipedia page on this topic is a fascinating read:

If sufficiently many scientists study a phenomenon, some will find statistically significant results by chance, and these are the experiments submitted for publication.

It seems like a pervasive and depressing reality – and one that's hard to test against. We need people to be encouraged to publish their results that find nothing, that don't support their hypotheses, and then as readers we need to take into account all the failed as well as successful studies.

One example of survivorship bias in the Wikipedia page (that challenges my own deeply held view 😂) is around the beauty of buildings in days gone by. They say that ugly buildings have been torn down and that largely the best examples of architecture remain – perhaps that is true, but I would argue that the best examples of architecture that we are building today in the cities that I visit are not nearly as beautiful as the examples of architecture that have survived from ages past. So yes, perhaps the idea that we used to regularly build beautiful buildings is incorrect (and not a position I've ever held) but the idea that we used to build some buildings that were more beautiful than the most beautiful buildings we build today could be sustained without being accused of survivorship bias.

Whether it be movie stars, athletes, musicians, or CEOs of multibillion-dollar corporations who dropped out of school, popular media often tells the story of the determined individual who pursues their dreams and beats the odds. There is much less focus on the many people that may be similarly skilled and determined, but fail to ever find success because of factors beyond their control or other (seemingly) random events. There is also a tendency to overlook resources and events that helped enable such success, that those who failed didn't have.

This is probably the classic picture of survivorship bias I have in my mind. Elton John may be an amazing musician (and I personally absolutely love his music) but in truth, he is no better than millions of others who were not able to make it to the big time.

I came across an unlikely reference to survivorship bias in The Singer's Audition & Career Handbook by Claudia Friedlander, which I think has a wonderful take:

The Singer's Audition & Career Handbook (Friedlander, 2019)

Survivorship bias is a logical fallacy that correlates one person’s triumphs with the specific strategies one took to achieve them while overlooking the fact that those strategies may not have panned out for others with any consistency. If you only examine the careers of operatic super-stars, your statistical sampling fails to include the many singers who followed the same procedures without achieving the same results. As a result, you may conclude that you have discovered the recipe for career success, when in fact the methods and outcomes you are observing may have little practical correlation.… In reality, these superstars comprise a very small percentage of our community. There are infinite ways to create a satisfying, successful career as a singer; there are also infinite ways to apply the communicative skills, emotional vulnerability, creativity, and poise that singers cultivate to other pursuits and professions.The idea of survivorship bias has weighed heavily upon me … it became clear that this book should include the stories of singers whose journeys exemplify the delightfully unique, sometimes eccentric, but nevertheless satisfying career paths that may await you.

I am not a particularly successful person. I have a good job, I have a sense that I'm in about the middle of my field. I have never published a book, and if I did, nobody would want to read it. There are very few people who know my name and I have not made any unique or notable contribution to any field. But I enjoy my life and my career path has been personally very rewarding at times (though the majority of it has been a hard slog). Though I think I've learnt a few things along the way that have helped me and I think might help others, I wouldn't suggest anyone try to copy me to emulate my success because it has largely been about right place right time right decision. Undoubtedly I could've made better decisions, done more to put myself in the right places, taken more risks, and so on, but there's no way of telling what those decisions were and there's an equal chance that I could've made worse decisions, put myself in the wrong places, and taken the wrong risks.

I wouldn't recommend anyone copy Steve Jobs – that only works if you get to meet Steve Wozniak, if you happen to stumble into the industry that's about to change the world, if your creativity is supported by an amazing team, etc. Steve Jobs was an amazing creative genius and he was hugely successful. There are countless others who were just as creative, just as intelligent, just as uncontrolled by conventional wisdom, who never made it big through, frankly, no fault of their own. I don't have this hero-worship that seems to pervade so much of our culture for people who have been successful because I've seen so many amazing people fail – and not through lack of trying.

Of course, as someone who was so successful, Steve Jobs is incredibly interesting to learn about, and there are absolutely things that his life and choices can teach us about how to succeed – the point of survivorship bias is that those who have failed have also got a lot to teach us.

There's a book called You're about to Make a Terrible Mistake by Olivier Sibony and in it there's a section entitled 'What Steve Jobs Did Was Brilliant, So I Should Imitate Him: Survivorship Bias':

You're about to Make a Terrible Mistake (Sibony, 2020)

The halo effect causes us to get an overall impression based on a few salient characteristics.– Steve Jobs's remarkable success makes us think that all his practices are worthy of emulation.– We try to imitate the practices of successful companies even when they are unrelated to their performance.– We don't challenge these practices enough to learn whether they are really applicable to us.

Sibony talks about the 'halo effect', which is related to survivorship bias, and it's where we think that everything a successful person does is worth emulating. That might be true of Jesus (picking up on one possible origin of the term's name), but it's certainly not true of Steve Jobs.

Okay, that's all for today. Thanks for reading.