Weekly Insider (Foxes, Hedgehogs, Hussein & Hendrix)

Quick note of interest from last week’s column: a reader pointed out that planned economy like the former-USSR had horrible washing machines, but exceptional MiG Jet fighters. The reason: competition. There was no capitalistic competition to produce better washing machines to win the wallets of consumers. But there was immense competition in the cold war to make sure Russia’s pilots had the absolute best. Found that interesting.

Today’s topic: the more I learn, the more I learn that nobody knows nothing. People have horrible forecasting accuracy.

Do you know what you do and don’t know? As Mark Twain said, “it ain’t what you don’t know that gets you into trouble. It’s what you know for sure that just ain’t so.” We’re guided mostly by what we know: the past (and even that we sometimes get wrong). Patrick Henry said, “I have but one lamp by which my feet are guided, and that is the lamp of experience. I know no way of judging the future but by the past.” But that doesn’t stop people from making lots of wrong predictions.

UC Berkeley psychology professor Phil Tetlock has empirically found that how you think is more important than what you think and that there is one thing you can do to make better predictions: be a fox instead of a hedgehog. Over 500 years ago, Archilocus said, (Multa novit vulpes, verum echinus unum): “The fox knows many things, but the hedgehog knows one big thing".

Foxes are skeptics and less confident in making predictions and build a latticework of mental models. Hedgehogs are more enthusiastic (especially about what they know) and more confident in making predictions and then pushing those predictions into all domains. As you’ll see, the quick brown renaissance Fox jumps over the staunchly opinionated Hedgehog.

Good decision making is about judgment under uncertainty. If you’re judgment is really well calibrated it means your assignment of the odds that something happens is really close to its actual frequency. So if you think something will occur 100% or 75% or 0% of the time—how often does that thing in fact occur? If you’re intellectually honest, you keep track over many predictions.

The experimenter equivalent of Tetlock on TV is The Daily Show’s John Stewart—whose video clips capture pundits (and presidents) proselytizing and predicting on the outcome of the primaries and then backpedaling to rationalize and justify why their prediction failed to materialize; only to go on with yet more soon-to-be-proven-inaccurate predictions. Tetlock has found the same held true with the prediction of the rise of Gorbachev; the fall of the Soviet Union; the Iraq war, Long-Term Capital Management; and current election predictions. As Tetlock says, ““Partisans across the opinion spectrum are vulnerable to occasional bouts of ideologically induced insanity.” Stewart Brand wrote of this, “The political expert who bores you with a cloud of “howevers” is probably right about what’s going to happen. The charismatic expert who exudes confidence and has a great story to tell is probably wrong.”

Consider the recent assassination of Benazir Bhutto. George Friedman brilliantly wrote in a recent Stratfor piece, “The murder of a major political leader is always hard to unravel. Confusion reigns from the first bullet fired in a crowd. The first account of events always turns out to be wrong, as do the second through fifth accounts, too. That is how conspiracy theories are spawned. Getting the facts straight in any murder is tough. Getting them straight in a political assassination is even harder. Paradoxically, more people witnessing such incidents translates into greater confusion, since everyone has a different perspective and a different tale. Conspiracy theorists can have a field day picking and choosing among confused reports by shocked and untrained observers.”

Sound like anything else you know? The Market. Even Larry Summers, in a 1989 research study, “What Moves Stock Prices?” showed that nobody knows nothing. All the post facto accounts of what caused a major stock market move failed to convincingly explain the moves. Robert Folsom recapped the following headlines and the time they occurred.

Hussein's Capture Seen Boosting U.S. Stock Market
(Sunday, Dec. 14, 6:42pm ET)
Stocks Rise in Morning Trading on Hussein News
(Monday, Dec. 15, 10:18am ET)
Stock Indexes Continue Rally at Midday on Hussein Capture
(Monday, Dec. 15, 12:21pm ET)
Stocks Close Sharply Lower Despite Hussein's Capture
(Monday, Dec. 15, 4:59pm ET)

Every day financial pundits explain yesterday’s events with great certainty. Try to explain today’s events with great uncertainty and even greater confidence. Rinse, repeat.

Some hedgehogs are often seen to predict big extreme changes. Not because they are more prescient, but they are tend to be in a minority of opinion holders for an outlandish outcome. But those outlandish outcomes are important to have out there. Hedgehogs cling to very extreme assignment of odds to something: i.e. it absolutely will never happen: 0% or it is certain to happen: 100%. As the saying goes, even a broken clock is right twice a day. The cost of being a hedgehog is a lot of false positives. They constantly predict some certain outcomes, but they are more often wrong as most do not ever occur: (remember Dow 36,000?). Hedgehogs are also more likely to be on TV as talking heads because they are more confident, more assertive and assign higher probabilities to low frequency events—which also make them more interesting to watch than someone who is more reserved.

It’s almost like Hendrix was a Hedghog singing, “Foxy/ You make me wanna get up and scream/ Foxy/ Ah, baby listen now / I’ve made up my mind / I’m tired of wasting all my precious time”

Foxes are the intellectual equivalent of a Web 2.0-mashup—(sorry, my own eyes just rolled as I wrote that)—taking pieces from all the hedgehogs who staunchly advocate one big idea and turning it into a latticework of mental models. Remember to the man with a hammer, every problem looks like a nail. And study after study has shown that diverse teams and diverse ideas have better predictive accuracy. Tetlock found that not only do foxes have more accuracy in their predictions; they are also more accurate in the odds they give to their predictions being correct.

I’ll leave Tetlock with the last word: “We live in an inherently probabilistic world. Nobody is expecting omniscience. It turns out that a somewhat contrarian, self-critical style does translate into more realistic subjective probabilities being assigned to possible futures. And if you think that good policy—or making money--depends on assigning more realistic probabilities to possible futures—that’s probably a result worth taking into consideration.”