Authors: Philip E. Tetlock, Dan Gardner

Type: Non-fiction, single subject

Full title: Superforecasting: The Art and Science of Prediction

Published: 2015

I read it: January 2016


I received a copy of this book as part of a Read It Forward giveaway.

Who is going to be elected president this fall? Clinton? Trump? Sanders? Cruz? Rubio? While it’s impossible to predict the future, we’ll hear plenty of pundits give their best guesses. They’ll probably speak them with confidence, couched in vague words such as “probably” or “unlikely” or “clearly.” What you probably won’t hear is a phrase like, “I put Clinton winning in November at a 72% chance, leaving a 28% chance she will lose.”

Yet this type of specificity is exactly what superforecasters shoot for. These aren’t the people, or predictions, you’ll see on news shows. Why?

Mostly it’s a demand-side problem: The consumers of forecasting–governments, business, the public–don’t demand evidence of accuracy. So there is no measurement. Which means no revision. And without revision, there can be no improvement.

Unlike pundits, superforecasters do not have a reputation or brand to protect. They are guided by their own initiative and spent countless hours weighing all possibilities. Many of them display the trait of active open-mindedness (AOM), an actual thing studied by psychologists. The superforecasters have an entirely different approach to prediction than the talking heads. But most importantly, they quantify their predictions, which are then scored based on the outcome of the world events being tracked.

This book explains how these are real people with real skills, and how the value of their forecasts could have huge implications in our complex world. Aside from the inherently interesting subject matter, the quality of this book makes it a stellar contribution to the popular science literature. It’s never overly technical, and the analogies and metaphors are well-crafted and helpful. The authors lead the reader to the book’s points with specificity and guidance. If asked a rhetorical question, the reader might then be told “Don’t answer yet” so that the authors can further develop a point. After the point is made, the reader might be asked to “Note three things,” a strategy that seems on the nose but is the exact type of writing needed in popular science books. Clarity over everything. Persuasion through exactness. It’s no accident that these are concepts that the superforecasters themselves are asked to stick to in their work.

The effectiveness of the book is probably due in large part to the dual authors. Tetlock narrates things in first person because it’s mainly his research, while Gardner is a journalist who has published work on similar topics and presumably helped to tighten things up. It comes together so well. Seriously, now: this is how you tackle a large topic with a lot of research, and actually get your ideas across to a wide audience. This type of work amazes me more than most rave-review novels I come across. (It’s also a great example of structure, for those who need help with chapter creation.)

I could probably go on. My final plug would be praise for the book’s intense skepticism. To take a slice:

A corporation or executive is on a roll, going from success to success, piling up money and fawning profiles in magazines. What comes next? Inevitably, it’s a book recounting the successes and assuring readers that they can reap similar successes simply by doing whatever the corporation or executive did. These stories may be true—or fairy tales. It’s impossible to know. These books seldom provide solid evidence that the highlighted qualities or actions caused happy outcomes. And they rarely acknowledge that factors beyond the hero’s control—luck—may have played a role in the happy outcomes.

You’ve seen these books. They perpetuate the Horatio Alger nonsense that puts personal anecdotes above all else. In reality, we live in a world of probabilities. The rational way to shape a society (or to simply survive one) is to study what happens in large numbers. Find the patterns, if they exist. Accept the gray areas. Allow ourselves to change our minds. There’s a whole lot of chaos out there, so we should be honest about what we do and do not know. And keep tabs on whether people who claim to know something are actually worth listening to.

One thought on “Superforecasting

Please leave your thoughts:

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s