Friday, May 24, 2013

Survivorship Bias - Lessons From A Brilliant Statistician Named Abraham Wald Who Saved Countless Lives

What is Survivorship Bias ?

The Misconception: You should study the successful if you wish to become successful.

The Truth:
When failure becomes invisible, the difference between failure and success may also become invisible.

Simply put, survivorship bias is your tendency to focus on survivors instead of whatever you would call a non-survivor depending on the situation. Sometimes that means you tend to focus on the living instead of the dead, or on winners instead of losers, or on successes instead of failures. In Wald’s problem, the military focused on the planes that made it home and almost made a terrible decision because they ignored the ones that got shot down.

It is easy to do. After any process that leaves behind survivors, the non-survivors are often destroyed or rendered mute or removed from your view. If failures becomes invisible, then naturally you will pay more attention to successes. Not only do you fail to recognize that what is missing might have held important information, you fail to recognize that there is missing information at all.


If you are thinking about opening a restaurant because there are so many successful restaurants in your hometown, you are ignoring the fact the only successful restaurants survive to become examples. Maybe on average 90 percent of restaurants in your city fail in the first year. You can’t see all those failures because when they fail they also disappear from view. As Nassim Taleb writes in his book The Black Swan, “The cemetery of failed restaurants is very silent.” Of course the few that don’t fail in that deadly of an environment are wildly successful because only the very best and the very lucky can survive. All you are left with are super successes, and looking at them day after day you might think it’s a great business to get into when you are actually seeing evidence that you should avoid it.

As the psychologist Daniel Kahneman writes in his book Thinking Fast and Slow, “A stupid decision that works out well becomes a brilliant decision in hindsight.” The things a great company like Microsoft or Google or Apple did right are like the planes with bullet holes in the wings. The companies that burned all the way to the ground after taking massive damage fade from memory. Before you emulate the history of a famous company, Kahneman says, you should imagine going back in time when that company was just getting by and ask yourself if the outcome of its decisions were in any way predictable. If not, you are probably seeing patterns in hindsight where there was only chaos in the moment. He sums it up like so, “If you group successes together and look for what makes them similar, the only real answer will be luck.”


- One the most fascinating posts of the year; read the whole piece here by David McRanely author of You Are Not So Smart: Why You Have Too Many Friends on Facebook, Why Your Memory Is Mostly Fiction, and 46 Other Ways You're Deluding Yourself

A Final Lesson:

As best I can tell, here is the trick: When looking for advice, you should look for what not to do, for what is missing as Phil Plait suggested, but don’t expect to find it among the quotes and biographical records of people whose signals rose above the noise. They may have no idea how or if they lucked up. What you can’t see, and what they can’t see, is that the successful tend to make it more probable that unlikely events will happen to them while trying to steer themselves into the positive side of randomness. They stick with it, remaining open to better opportunities that may require abandoning their current paths, and that’s something you can start doing right now without reading a single self-help proverb, maxim, or aphorism. Also, keep in mind that those who fail rarely get paid for advice on how not to fail, which is too bad because despite how it may seem, success boils down to serially avoiding catastrophic failure while routinely absorbing manageable damage.



No comments: