Fox vs. hedgehog
Greek warrior-poet Archilochus wrote: “The fox knows many things, but the hedgehog knows one big thing.” For predicting the future, foxes are better than hedgehogs.
Psychologist Philip Tetlock discovered this by conducting a 15-year study with 300 experts in political science and economics. About 60% had been interviewed by a major news outlet. Experts made 80,000 predictions in response to 27,000 forecasting questions.
Overall, experts performed slightly better than chance. About 30% of predictions stated that an outcome was highly unlikely or impossible, but these events actually happened 15% of the time.
Experts who performed well were open-minded, skeptical, and humble. They gathered information from diverse sources. They were willing to change their minds. In contrast, the worst were ideological and stubborn.
But even the best experts only had a prediction accuracy of 20%. In comparison, mathematical models approached 50%.
You can be an intellectual yet still be an idiot. “Educated philistines” have been wrong on everything from Stalinism to Iraq to low-carb diets.
The fastest way to lose trust in experts is to be one.
Emperor of Wall Street
In 1978, Jim Simons left his job as a math professor to start a hedge fund called Renaissance Technologies. Based on his experience at Harvard, MIT, and the National Security Agency (NSA), Simons built computer models for trading stocks. His goal was to replace human emotion with machine logic.
Since 1988, Renaissance has earned $100 billion by betting against human misjudgments in the stock market. This corresponds to an astounding annual return of 66%. For perspective, returns over this period were 19% for Warren Buffett’s Berkshire Hathaway Inc. and 12% for the S&P 500 Index.
If stock market experts were so expert, they would be buying stock, not selling advice.
Tetlock PE. (2005). Expert political judgment. How good is it? How can we know? Princeton University Press.
From 1987 to 2003, 284 professional experts in political science and economics made 82,361 subjective probability estimates in response to 27,450 forecasting questions
Experts worked at universities, think tanks, U.S. government, World Bank, International Monetary Fund (IMF), media, and other leading organizations
61% of experts had been interviewed at least once by a major media outlet, and 21% had been interviewed more than 10 times
Overall, experts performed slightly better than chance
Out of all predictions made, experts declared outcomes to be either highly unlikely or impossible 30.3% of the time, but these outcomes actually occurred about 15% of the time
Dilettantes who read news sources such as the New York Times, Wall Street Journal, and Economist performed as well as domain experts, and both performed better than Berkeley undergraduates
“Foxes” were defined as experts who were flexible, skeptical, and humble
“Hedgehogs” were inflexible, confident, dismissive, and believed in one overarching theory (e.g., environmental doomsters, cornucopian boomsters, socialists, and free-market fundamentalists)
Foxes chose different analytical tools depending on the problem; gathered information from diverse sources; talked about possibilities and probabilities instead of certainties; and more readily admitted mistakes and changed their minds
Foxes’ probability calibration scores deviated from reality by 12% vs. 18% for hedgehogs
Foxes’ discrimination scores captured 18% of variance in outcomes vs. 14% for hedgehogs
Comparing humans to computers, the best foxes approached a prediction accuracy of 20% vs. 25–30% for case-specific extrapolation algorithms, and 47% for sophisticated statistical models
Stevens P. (2019, November 5). The secret behind the greatest modern day moneymaker on Wall Street: Remove all emotion. CNBC. https://www.cnbc.com/2019/11/05/how-jim-simons-founder-of-renaissance-technologies-beats-the-market.html
Jim Simons (mathematician). (2021, March 8). Wikipedia. https://en.wikipedia.org/wiki/Jim_Simons_(mathematician)
Buffett WE. (2021). Berkshire Hathaway Inc.: 2020 annual report. https://berkshirehathaway.com/2020ar/2020ar.pdf
Did you enjoy this chapter? Read the full book. Download your free copy: EPUB MOBI PDF
Copyright © by Raven Dojo Inc.
Limit of Liability/Disclaimer of Warranty