Superforecasting: The Art and Science of Prediction by Philip E. TetlockMy rating: 4 of 5 stars
How many political forecasters have their past predictions analyzed before they're invited to make more predictions on TV? How many viewers check how accurate Jim Cramer has been over his career?
Some other notes I took:
- Medicine: from Ancient Greece to George Washington, bloodletting wasn't questioned.
- Galen said: "with my medications only the incurable die" --> confirmation bias (if they live it's thanks to him, if they die it's not his fault).
- Modern scientists (and all forecasters) need to ask themselves: what would convince me I'm wrong?
- The more ideological, the worse the predictions, and the less they admitted it.
Much of the material seems to build off of Daniel Kahneman's work, but still very good.
View all my reviews
No comments:
Post a Comment