That’s the question that we all face every day. Life sends us umpteen demands to make predictions. Some are pretty easy: how long will it take me to drive to work; will my spouse appreciate a special gift on Valentine’s Day? Others are hard: how much concrete will be needed for a building; what will Facebook’s earnings be next year?
When we make predictions, we draw on a combination of calculation and experience. Some calculations are precise, using engineering tables and detailed formulas. Others are more intuitive: chess masters can read a chessboard quickly because games often fall into familiar patterns and they remember optimal responses from prior games.
Where we get into trouble is when we make snap judgments based on partial knowledge where we substitute an easy question for a harder one and rely on irrelevant data. In estimating earnings, analysts often simply extrapolate current trends, which can result in spectacular over or underestimation. Apple comes to mind here: ¾ of their revenues now come from products that didn’t exist five years ago, so just extending the trend didn’t work.
Adjusting our intuition is hard work. First impressions are powerful. Sometimes there isn’t time to critically assess all the relevant facts, like when we’re driving in heavy traffic. But when we can think twice about things, we should. Snap judgments are usually too extreme, and we’re too confident in them. Instead, we need to discipline ourselves to rely only on relevant information, and the more information, the better. Thus, estimates of 3M’s earnings are likely to be more reliable than Google’s, as the company has been around longer.
When we make predictions, it’s best to sit back and evaluate: am I predicting an extreme event from a small amount of data? By their very nature, extreme events are rare. Building your life—or your investment portfolio—around them isn’t likely to work out very well.
Douglas R. Tengdin, CFA
Chief Investment Officer
Hit reply if you have any questions—I read them all!
Follow me on Twitter @GlobalMarketUpd