How Good Are Expert Predictions of Geopolitical Events?

by | Aug 29, 2018


As humans we crave certainty; uncertainty makes us uncomfortable. (Related IFOD: Decision-Making Under Uncertainty: The Ellsberg Paradox.)Thus, we love to listen to experts explain what is happening in the unpredictable world we live in and tell us what is likely going to happen in the future, particularly with respect to the economy/investing and geopolitical events.

How good are experts at predicting geopolitical events? Should we listen to them at all?

Philip Tetlock, a sociologist at University of California-Berkeley, has made it his life’s work to collect and examine expert political judgment. Over the past few decades he has collected over 80,000 political predictions, including ones relating to whether Soviet Union would collapse, whether there would be a non-violent end to Apartheid, whether the U.S. go to war in the Gulf, whether Canada split apart, and most major events over the last 30 years. What he has found is quite interesting:

  • Expert political predictions provide almost no value. They barely outperform control groups of college students and under perform a simple formula equal weighting various outcomes. Almost nobody predicted the collapse of the Soviet Union. Not even the Soviet experts at the CIA and top Universities.
  • Knowing a bit about a topic increased prediction ability over someone who knew nothing, but specialists with great amount of knowledge didn’t do any better than non-specialists with little knowledge. The point of diminishing returns for the amount of knowledge useful in predictions is reached quite quickly.
  • The more confident an expert was, the more they are interviewed by the press, the less accurate the predictions. Thus, those experts we hear from most are the ones we should least be listening to!!!!
  • Like all of us, Tetlock found that experts were subject to confirmation bias and thus sought out facts that supported their preconceived views and tended to dismiss conflicting information.
  • Over the course of Dr. Tetlock’s studies, events classified as experts as having “no chance of happening” actually occurred about 15% of the time and those things experts declared were absolutely going to occur failed to occur 25% of the time!
  • Tetlock found that experts, like all of us, fail generally to appreciate the complexity of the present and unpredictable nature of the future. There are too many factors and too many unknowns to appropriately predict most situations and most experts were unaware of this fact. They confuse knowledge with the ability to properly identify, weight and analyze the many relevant factors involved determining geopolitical events. Simply put: knowing a lot of information and the ability to synthesize that information into useful predictions are two completely different things. Thus, knowing a lot about something and being able to predict what’s going to happen have very little relationship.

Related IFOD: Ignorance and the Dunning-Kruger Effect

Have a nice hump day


  1. So, flipping a coin is maybe better than placing confidence in what experts predict, from the sound of it.

  2. Just trust the settled science! Don’t question the experts! Do as you are told!
    …hmmm…maybe not so much…


Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.


Subscribe To The IFOD

Get the Interesting Fact of the Day delivered twice a week. Plus, sign up today and get Chapter 2 of John's book The Uncertainty Solution to not only Think Better, but Live Better. Don't miss a single post!

You have Successfully Subscribed!

Share This
%d bloggers like this: