Monday, November 12, 2018

is this post a black swan event?

A friend recently asked me to explain some of the things he might learn from reading The Black Swan. My answer turned into a mini-retrospective on one of the books I thought influenced me the most in my early adult years.

Here are some of the highlights:

1) It’s a pragmatic textbook for “if, then…” thinking

Many readers used this book’s ideas to cite certain unlikely events as possible ‘black swans’. No! A black swan event is unpredictable – the point of the book was to know how to respond given such an event has occurred.

A major illness is a good example. Most people become seriously sick without much forewarning. Though there is logic for trying to control every known risk factor out there, it remains very close to impossible to predict onset of a specific illness. Most people would get more out of planning what they would do if they became ill rather than trying to predict what illness they might get.

2) Statistics and common sense make for a powerful duo

This book shows that statistical theory is no substitute for some basic common sense. The example I like to give involves a gambling scenario.

Suppose you are betting on a coin toss. You are told the coin is evenly balanced. The first ten tosses are heads. What should you bet next?

The statistics textbook would say it doesn’t matter because the coin is fair. The past outcomes don’t dictate future outcomes so the odds are fifty-fifty.

However, common sense would say to bet heads. This considers one relevant factor – is the coin really evenly balanced? If it isn’t, the evidence is clear to bet heads. And if it is, well, since it doesn’t matter, what’s the difference if you arbitrarily prefer one or the other?

3) Variation means asymmetry

If point #1 summarizes how this book thinks, this point summarizes what this book thinks. The most common mental model we have is linear – one unit of input equals one unit of output, every gallon of gas is thirty miles of driving, and so on. This model is convenient because it almost eliminates the effects of variation on the final outcome.

The Black Swan challenges the reader to look for situations when one unit of input leads to different units of output. These events are governed not by their linearity but by their variation. Learning to identify these is vital to success in many real-world pursuits because most real-world pursuits are not represented well by linearity.

A good example is something I witness everyday in Boston – jaywalking. The risk of getting hit by a car is so low that most people decide to take the chance by crossing a street full of active traffic. These pedestrians arrive at their destinations seconds (or maybe even a minute) earlier than they would have. But is it worth the possibility of ruining (or losing) the next ten, twenty, or thirty years of life if there is a fatal accident? The answer is obvious, at least for those who think clearly in terms of asymmetry within variation.