On Tuesday I told you about how myself, Dan Denning and Ben Traynor found ourselves in a lecture theatre in central London surrounded by people who call themselves “superforecasters”.
If you’ve not read the article yet, you can find it here. I’d suggest you read it before continuing.
Anyway on Tuesday I explained roughly what a superforecaster was (a person with the ability to forecast future events 60% better than average) and how superforecasting came about (a CIA funded academic research project).
What I didn’t say was exactly what the superforecasters did, or how they did it. In other words, the good bit!
Today, I want to talk about the single most important attribute of a superforecaster.
Who they are
Professor Philip Tetlock organised the forecasting tournament, and he also helped to put together the teams. One team was made up of a very large number of ordinary volunteers from around the United States.
After a year, Tetlock spotted that a group of forecasters among the volunteers were consistently outperforming the others. And it wasn’t just beating the other volunteers – it was beating the best of the CIA.
They were the superforecasters. And after the first year, Tetlock separated them out into their own teams in order to study them more carefully.
Charlie Munger, superforecaster
At this point, I’d like to take an abrupt left turn to talk about Charlie Munger, Warren Buffett’s famous business partner. He’s a billionaire investor of course, the man credited with convincing Buffett to evolve his value investing strategy into a “value plus quality” strategy. He’s made Buffett a lot of money. He serves on the board of Facebook. And he’s admired far and wide for the way he thinks.
I was listening to a podcast featuring the author of a new book on Charlie Munger at around the same time I went to the superforecasting conference (The book? Charlie Munger, The Complete Investor by Tren Griffin). And I remember being struck by how much the two subjects had in common.
Basically when Phil Tetlock went through the data to work out the key traits of the superforecasters, he came up with a list of Charlie Munger’s personality traits.
There’s a lot to say about superforecasters traits, and about the genius that is Charlie Munger. I’d like to just focus on one, the single most important trait in a superforecaster. Tetlock calls it “belief updating”.
According to Tetlock, the willingness to update your beliefs improves forecasting accuracy more than any other trait. What it means is that you constantly test your beliefs and update them, based on new evidence. That’s hard though! Nobody wants to actively try to convince themselves that their cleverest ideas are no good. As Tetlock put it, “Beliefs are hypotheses to be tested, not treasures to be guarded.”
Munger says the exact same thing (except his line is better): “Any year that passes in which you don’t destroy one of your best loved ideas is a wasted year.”
As an investor, your portfolio is a full of your “best loved ideas”. You take the time to research and analyse different potential investments. You go with the ones which seem most compelling to you. But Tetlock and Munger are saying is that this is just the beginning of the process. To make good predictions, or to run a portfolio profitably, you need to force yourself to constantly test your ideas, and update them, and assess whether your original hypothesis/forecast/decision to buy a stock still makes sense.
It’s hard to make yourself do it. But it’s an important step to being a better investor.
As Munger said, “The only way to win is to work, work, work, work, and hope to have a few insights.”