Survivor Bias & the Teleology Trap

I have been reading an annoying book called Tribes: We Need You to Lead US.  It is typical of many such books about change, new technologies etc. in that it has that insufferable air of superiority and fawning homage to coolness.   There are some good ideas; enough to fill an article or maybe a blog post, but not a book.   But it wasn’t a waste.   I thought about some of the reasons these books sell so well and how the faculties they employ sound so convincing … and why I probably cannot use much of the advice.

This book and others like them (some very good such as the Tipping Point) suffer from survivor bias.   They talk about how little movements turn into big things.  You get examples, like Google or Facebook, that seemed to come from nowhere to dominance.  The examples are real, but the game is rigged.   It is “survivor bias” & the teleology trap.    It is when you assume that conditions you observe today were a natural or even an inevitable outcome. You infer a pattern or purpose where none is implied.  It comes from a faulty perspective.   A lottery winner asks, “What are the chances that I would win?”   But the correct question is “What are the chances somebody will win?”  Of course, the chances of that are 100%.   If you bet on that, you are always right.

A simple illustration of survival bias is a game of Russian roulette.   Survival is a random event, but at the end of the game, somebody will have survived.    If a large enough group starts out, somebody will survive even repeated plays.   You just don’t know who it will be.   No doubt the “winner” will come to believe that he knows some secret of success.   He may write books about it.   But there is nothing to it.

Of course there is more involved in technology companies, celebrities or investing, but sometimes less than you would think.    Let’s take the example of celebrities.   All celebrities have talent.  They are all attractive in some way.   They are generally smart (or at least smarter than they seem).  A celebrity needs these things.  They are threshold requirements, the minimum you need them just to get through the gate, but they are not enough to provide success. Beyond that, the winning combination will not be known in advance.   It looks very different depending on when you look.    

If I know that a person is a celebrity, it is fairly easy to go back and explain why her ascent was inevitable.   But if I substituted a non-celebrity, I might find much the same biography.   When you listen to the winning Super Bowl team, they always say that they knew they were going to win and confidence is important, but the losers also knew they were going to win.   They were just wrong.

Books like Tribes play on this bias.  It is like naming the winner of the lottery AFTER the drawing.  It is easy to be prescient about the past.   The other thing I don’t like about this book is the appeal to coolness.    They talk about the latest styles and sing paeans to change.    People today demand the latest, they say.   Once again, they are right, but so what.   A lot of change is just froth.  I have been reading these sorts of books for more than twenty years.   They are always talking about the changes and the changes are always happening.   But most of them don’t stick.   In fact, the authors are usually self-contradictory.   They seem to think radical change will happen and then it will be the change they want, but things keep on moving.

Another thing that annoys me about the cool change folks is that they don’t seem to understand cycles.  The author of Tribes triumphantly states that smaller organizations are often growing faster than bigger ones.   What a surprise!   A young tree grows faster than an old one, but trends often do not continue.  Beyond that , when praising the growth of the small, it is useful to do some simple math.  Which would you rather have,  50% of $1 or 5% of $1000? 

I have a special perspective on cool fads.  As an FSO, I am away from America sometimes for years.   Sometimes I miss entire fads.   They come and go while I am away and there was no point in even thinking about them.   Ephemera.   Meanwhile, my bookshelf still has lots of classics that are never wildly popular but endure.   The book that has been longest in my possession is a copy of Gibbon’s Decline & Fall of the Roman Empire that my mother gave to my father the Christmas of 1954, the year before I was born.   It was first published in 1776.   I first read it in 1966 and it is still a joy today.  Who could have predicted that the yet unborn baby would still be reading that book fifty-five years after my mother bought it?   

Change is good & and so is continuity.   A balanced person treasures both but is beguiled by neither.

BTW – Survivor bias is why we think stuff was better in the past.  All the junk has long since gone in the garbage.  What is left is better or at least by definition the longer lasting.