The Flaw of Averages: Why We Underestimate Risk in the Face of Uncertainty

CHAPTER 2

The Fall of the Algebraic Curtain and Rise of the Flaw of Averages

In 1973 I received a PhD in the application of computers to operations research. OR, as it is also known, grew out of the first widespread use of mathematics in decision making to tackle the unprecedented operational problems of World War II. (It should not be confused with the modern field of opposition research, in which people running political campaigns scour the backgrounds of their adversaries for dirt to discredit them.) After my graduate work I spent a year and a half at General Motors Research Laboratory, working on issues ranging from optimal production to marketing. I was a lot more interested in real-world problems than I had been in school, and had GM acted on my suggestion to relocate the lab to the Italian Riviera, I would probably still be there today.

Becoming a Management Scientist

In 1974, I joined the University of Chicago Graduate School of Business and started teaching management science, which is what they call OR in business schools. This appeared to be the opportunity of a lifetime, but I was soon disappointed to discover that only 10 percent of my students understood what I was talking about and that, of those who did, only 10 percent would go on to apply the stuff.

One of the primary offenders was a resource allocation technique called linear programming (LP). Although in theory it had the potential to optimize profits, in practice it was too cumbersome for all but the largest ...