Is There a Track Record Over the Long Term?

Experts can get lucky, but over the long haul people who get lucky regress to the mean. If you take a thousand financial analysts and ask them to predict what's the market going to be like in four months, half of them will get it wrong. So if you start with a large enough number, there's going to be some who slip through. I really like Tetlock's work very much, but in a naturalistic setting, it's often very difficult to apply.