Check Your Priors: Superforcasting by Philip Tetlock with Dan Gardner

 I had never read the previous book that made Tetlock’s name, but that 2004 book gets a lot of play in the other social science books that I read that it is as if I feel that I had read it – especially the framing of the Hedgehog and the Fox (though I like it, I can never remember which one is better).

Therefore, when this one came around, I thought I would have a look. Tetlock and his coauthor tell a fun story about the art and science of prediction, but I would say it leans a bit towards the popular side. Not sure if it hurts the story that is being told. There are a couple of things that you need to do to be good at predictions, and theory in here. You know, don’t be too wedded to your priors, and have a good sense of probability and what it really means. That was the key thing for me, looking at how people predict normally versus someone that is trying to really figure out what the difference between a 60% and a 63% probability of something happening. I’d say we have to teach people statistics and maybe before that basic numeracy.

What really got me was that this book is really about (to me) Bayesian prediction, but the book doesn’t really mention that until page 170. I would have brought that closer to the front. Overall, it was a fun read, but I’m not sure if it really made me better at predictions. I guess I have a good resource if I’m ever in a prediction contest like the one they talk about in the book.