A review by nocto
Superforecasting: The Art and Science of Prediction by Philip E. Tetlock, Dan Gardner

4.0

This was a really interesting and engaging read and full of stuff that seems kind of obvious after you read it, which isn't in any way a criticism. It's a good insight into why some people are much better at predicting the future than others and what they do well in order to do that. I love the phrase "master the error-balancing bicycle" and the thread running through the book is largely about making sure you give the right amount of credence to the right bits of evidence, though there's no foolproof method to acheive that balance. My takeaways are that, yes, pundits are (pretty much) always wrong and/or so vague they can frame anything that happens as what they predict. And that the future is really hard to predict in a big picture sense; some people are really good at predicting the answers to specific questions but ask what the big events and stories of the next decade will be and almost no one gets the emphasis in the right places except in hindsight.