Go to the Globe and Mail homepage

Jump to main navigationJump to main content

AdChoices
At the most basic level, your options for available cash include spending it, investing it, paying down debt, or giving it away. (Andrey Popov)
At the most basic level, your options for available cash include spending it, investing it, paying down debt, or giving it away. (Andrey Popov)

How to identify 'super forecasters' when it comes to market predictions Add to ...

It is a discussion in a Woody Allen movie, but it is also a line from the Greek poet Archilochus, who described foxes as self-critical thinkers who can adjust their beliefs and hedgehogs as people with one big idea who try to persuade others to follow along.

It’s the same idea Philip Tetlock, a psychology and management professor at the University of Pennsylvania, studied over decades as he tried to determine what gives one so-called expert more credibility than any other, well, so-called experts.

Most of them are wrong. In his 2005 book, Expert Political Judgment: How Good Is It? How Can We Know?, Prof. Tetlock posed hundreds of questions about political and economic events to an array of experts and non-experts, generating some 80,000 predictions. The so-called experts could explain only 20 per cent of the variability in outcomes in their predictions, regardless of their educational backgrounds, experience and access to information. The more famous the expert, the less accurate he or she tended to be.

That is an important point in these days. Experts predicted Britain would vote to stay in the European Union, and that didn’t happen. Donald Trump won the U.S. presidential election despite the overwhelming predictions and polling forecasts to the contrary. The U.S. stock markets have marched to record highs since then, also to the complete astonishment of many market prognosticators and directly contradicting predictions.

Investing is equally fraught with unpredictable outcomes despite seemingly reasonable expectations. The financial media tend to give hedgehogs most of the headlines because they are the ones with the bold ideas. Charisma and trying to predict what will happen in the short run wins over substantive fundamental analysis and long-term thinking.

Hedgehogs sometimes score big, such as professional short-sellers who accurately predict a company’s demise or a pro who makes a correct short-term market timing call. But they also get a lot of false positives, explaining their failures as things that might have gone right if certain other things happened differently or not at all. A hedgehog may be the one calling the next bear market or predicting a recession, but as former investment columnist Morgan Housel points out, those sounding off alarm bells on the market usually sound smarter, making cogent arguments, and yet all the evidence superior results by taking a long term, disciplined view and not trying to predict short term market trends.

Experts will do better than complete novices. Prof. Tetlock’s research found that ordinary forecasters working in teams would do about 10 per cent better than a crowd at predicting an outcome. That’s slightly less well than what a prediction market would do. But super forecasters would do 15 per cent to 30 per cent better than that.

In a 2010 paper, Prof. Tetlock wrote, “Experts do not know nearly as much as they think they do, and they work hard to cover up mistakes, but they do still at least perform better than the general public.”

What Prof. Tetlock found in his research, which spanned the 1980s and 90s, was that foxes, who apply a little skepticism and change their minds when confronted with data that don’t match their hypotheses, tend to do better than hedgehogs. Their success rate is greater in short-term forecasts as well as the long-term, and they were more accurate in the likelihood they gave to their predictions.

Why are super forecasters better? They are willing to test their hypotheses and change if proven wrong rather than defend them at all costs.

Computer models tend to be better predictors than humans. The simple algorithms Prof. Tetlock analyzed in his research explained as much as 30 per cent of outcomes while more sophisticated models could explain nearly half.

That’s why we designed portfolios that track the quantitative strategies of several successful investors who believe in companies that have solid fundamentals. Models take the emotion out of investing, the human frailty that makes us prone to buying or selling stocks at the worst possible times.

One of those gurus, James O’Shaughnessy, wrote in his book What Works on Wall Street that human forecasters fall short of statistical models because human forecasters are affected by emotion and make inconsistent judgments, act in short-sighted ways or suffer from overconfidence.

Other investors we track similarly understand the importance of data triumphing over emotion. Warren Buffett, Peter Lynch and Benjamin Graham dodged the negative effects of emotional investing by sticking to the numbers. Mr. Graham, whose book The Intelligent Investor has inspired millions, including Mr. Buffett, said: “Investing isn’t about beating others at their game. It’s about controlling yourself at your own game.”

So next time you hear a expert making a prediction, keep Prof. Tetlock’s research in mind and ask yourself, does this person sound like a fox or a hedgehog?

Report Typo/Error

Follow us on Twitter: @GlobeInvestor

Also on The Globe and Mail

How do robo-advisers benefit millennials? (The Globe and Mail)

Next story

loading

In the know

The Globe Recommends

loading

Most popular videos »

Highlights

More from The Globe and Mail

Most popular