Why Investors Shouldn’t Trust Short-Term Predictions

It’s common for stock market participants to see predictions on the direction that a certain share or the market is heading over the short-term. And truth be told, most market participants lap these predictions up, thinking that these ‘experts’’ view on short-term movements will give them an edge in the market.

But, as it turns out, predictions can be a sucker’s game.

The charts below will show you why.

investors shouldnt trust predictions chart 1

Source: The Little Book of Behavioural of Behavioural Investing by James Montier

investors shouldnt trust predictions chart 2

Source: Value Investing – Tools and Techniques for Intelligent Investment by James Montier

The first chart shows the discrepancy between analysts’ forecasts for earnings of the S&P 500 Index (a broad index for the American stock market) and what actually happened. And, it shows how consistently these professionals’ estimates of the future actually lag reality.

The second chart shows economists’ forecasts for US GDP and the actual result. Turns out, economists are just as bad at predicting future GDP as analysts are at predicting future earnings.

Our pattern-seeking brains

Looking at studies of forecasts such as those shown above would likely make you wonder just why we can’t seem to get it right. Well, the reason actually lies with our brain.

According to studies done by psychologists Amos Tversky and Daniel Kahneman (who’s a Nobel Laureate) that’s recounted in Jason Zweig’s book, Your Money and Your Brain, “our minds are wired in such a way as to base long-term trends on surprisingly short-term samples of data – on factors that are not even relevant.”

Take the following study as an example.

Two bowls contain both red and white balls of which two-thirds must be of one colour and one-third must be another. The bowls are kept unseen and balls are taken out of both; four white and one red ball was taken from Bowl A; 12 red balls and 8 white balls were taken from Bowl B.

Now, you’re blindfolded and made to take out only one ball. You’ll win $5 if you can make the correct guess of the colour of the ball you pick. So, the million dollar question becomes: Should you bet on a white ball from Bowl A or should you bet on a red ball from Bowl B?

Turns out, most chose to bet on picking a white ball from Bowl A. But, that’s irrational decision-making. Balls drawn from Bowl B constitute a much larger sample and so makes it more likely that Bowl B is the one that’s mostly red. In the end, however, most people get distracted by the small sample of balls drawn from Bowl A.

We’re only good at extrapolating from recent events

The above experiment illustrates how easy it is for us to fall prey to making predictions based on short-term samples of data. And the worst part of it? It is an automatic process. We cannot make it go away, even after knowing it.

That’s why we can see that lagging-pattern for analysts’ forecasts of S&P 500 earnings compared to actual reality. They see earnings going up by 5% last quarter, and use that as a ball-park figure to estimate next quarter’s earnings. If last quarter had a 10% fall in earnings, then a negative-10% is the likely starting point for the next quarter’s estimate.

In short, that’s what’s known as extrapolation – we take what has happened over a short span of time and project it over longer periods of time. There’s nothing wrong with such a predictive technique, except that the financial markets rarely moves in such predictable linear fashion.

How to fix this in investing?

Naturally, a discussion on the fallibility of predictions would lead to a question of, ‘What should we do then, if we still want to invest?’ And fortunately, all hope’s not lost.

A good way to overcome the problems of predictions would be to focus on base rates, which is a fancy way of saying, ‘always look at how things have historically panned out’.

If some pundit says the Straits Times Index (SGX: ^STI) will go on a bull run over the next three years, gaining 20% per year, check back with history.

Turns out, the STI has compounded at a rate of around 5.6%, since the start of 1988 to current levels of around 3,300 points, for a period of close to 25 years. And, there has only ever been two rolling-three-year periods out of 23 between the years 1988 and 2012 where the index compounded at more than 20% per year for three years (the periods are measured from the start of January of the starting-year to the end of December of the ending-year).

Investors can then use such statistics to make informed judgements on the likelihood of that pundit’s claims.

This is just one example of how we can focus on base rates to aid us in our investing decision-making.

Foolish Bottom Line

Our brains are already naturally handicapped in-terms of us being able to make predictions, so why should we carry on with the folly?

Ultimately, investors have to look at what has happened in the past – over the long-term – with the greatest odds, instead of conveniently extrapolating what has happened over a short span of time far out into the future. As Pericles of ancient Greece once said, “We should wait for the wisest of all counselors, time.”

Click here now  for your  FREE subscription to Take Stock Singapore, The Motley Fool’s free investing newsletter. Written by  David Kuo ,   Take Stock Singapore  tells you exactly what’s happening in today’s markets, and shows how you can GROW your wealth in the years ahead.  

The Motley Fool’s purpose is to help the world invest, better. Like us on Facebook  to keep up-to-date with our latest news and articles.

The information provided is for general information purposes only and is not intended to be personalised investment or financial advice. Motley Fool Singapore contributor Chong Ser Jing doesn’t own shares in any companies mentioned.