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

CHAPTER 40

The End of Statistics as You Were Taught It

Francis Galton, introduced in the last chapter as the father of eugenics, should not be blamed for the horrors committed in its name any more than Einstein was responsible for the bombing of Hiroshima. Galton was also central in the development of the field of statistics. He recently received some well deserved good press in the opening scene of James Surowiecki’s Wisdom of Crowds, in which he appears at an English county fair in 1906 and marvels that the average guess in a contest to estimate the weight of an ox was extremely close to the true value.¹ “The result seems more creditable to the trustworthiness of a democratic judgment than might have been expected,” he wrote upon observing this early prediction market.

Stephen Stigler, a prominent University of Chicago statistician and son of the economist George Stigler, has written extensively on the history of statistics,² including early attempts by Galton and others to generate random numbers through mechanical means. Stigler’s paper on “Stochastic Simulation in the Nineteenth Century”³ shows that even as these pioneers were at work on the foundations of classical (Steam Era) statistics, they were testing their theories with simulation methods.

Galton’s Dice

My favorite part of Stigler’s paper quotes directly from a Galton paper of 1890, in which he discusses his latest random number generator based on dice:

Every statistician wants now and then to test the practical ...