I followed Tyler Cowen’s recommendation to read the Tetlock interview (it’s gated, but in the past Financial Times has just wanted you to register for free).
Tetlock thinks experts make bad forecasters. That somehow they (and presumably me too) get biased. He advocates for teams of non-experts with data and quantitative skills.
… [Tetlock’s] work has shown that the big ideas people … should expect to be proven wrong by history. “Yes, but could they do as good a job if they knew that?” he said with a laugh. Later he put the same point more darkly: “There is a price to be paid for feeling good about your beliefs.”
And what should ideas people do (Iraq edition):
… Tetlock uses the example of the New York Times columnist, Tom Friedman, a classic big ideas man who has faced pointed criticism for being hugely wrong about the Iraq war. Yet Friedman posed a great question early on: whether Iraq was the way it was because of Saddam, or whether Saddam was the way he was because of how Iraq is. “That’s a pretty darn good question”, Tetlock notes; it’s just that Friedman, the grand theorist, was the wrong person to try and answer it. The big ideas people have a crucial role: they give real forecasters interesting questions to work on.
Lastly, here’s one I haven’t heard in quite a long time:
He repeats the joke of the US conservative writer, William Buckley, about members of that denomination, that “at most they believe in one God”.