Mark Sellers: Tend to overanalyse stocks? Bring out the dart-wielding monkey

Listen to this article


Should you spend countless hours researching a stock before buying it, or just a few hours? The common wisdom is that the more time you spend on research, the better your investment results will be. But is this correct?

Not according to studies on the subject. At least two I have seen came to the conclusion that anything more than a cursory review of a company’s prospects is a waste of time. One study looked at whether more information was useful to experienced horse racing gamblers. The researchers first gave the gamblers a few items of relevant information (age of horse, pedigree, jockey, etc) and asked them to predict which horse would win the race.

They then gave out dozens more items of information about the race
and asked the experts to predict again, based on the additional information they had received. The conclusion was that the first few items of information given to them were useful but further information had no effect on the accuracy of their predictions. The gamblers were far more confident about their predictions after receiving more information, but this didn’t translate into accuracy.

Another study showed that sell-side stock analysts, who spend 50-80 hours
a week on research and presumably have loads of information at their fingertips, are no better than a monkey with a dart at predicting a company’s five-year earnings growth rate. In fact, they are actually worse than a monkey with a dart because they tend to project the recent past into the distant future, rather than assuming growth will revert to the mean. If the analysts’ projections were simply random, the results may have been closer to reality.

Furthermore, large mutual funds tend to hug the return of their benchmark more closely than smaller funds do, in spite of the fact that larger funds have more resources and presumably are able to do more research. They can pay an army of analysts to research every nook and cranny of a company’s business and come up with proprietary earnings estimates and valuation models. Yet more research does not equate to better returns, at least among mutual funds (and probably hedge funds).

It would, therefore, be easy to conclude that research is a waste of time. But I have a different take on this issue. In my view, the problem is not with research, per se. The problem is with the type of research people do. They miss the forest for the trees.

The typical fund manager is bombarded with dozens of news items about his stocks and makes many seemingly important buy and sell decisions every day. The noise doesn’t get filtered out. Everything seems important to some managers.

The thought process is something like the this: What will the Federal Reserve do at its next meeting? Will company X beat earnings estimates next quarter or miss them? Which direction will oil go from here, up or down, and how might that affect consumer spending? Is the economy headed into a recession? What will company X’s stock price do over the next three months? Will the analysts reduce earnings estimates after I buy the stock, resulting in short-term losses? Are analyst estimates too low
so that there is a short-term profit opportunity? Which will perform better over the next year, foreign stocks or domestic stocks?

These are the thoughts of a typical fund manager on a typical day. Thoughts are scattered across a range
of subjects, with questions that can’t possibly be answered with any accuracy (like those above) gaining the lion’s share of the attention span. If you waste much of your time worrying about questions that can’t be answered, you are spending less time on the things that can be answered.

Contrast this with the investment process of Warren Buffett and his partner, Charlie Munger. Their record is astounding, yet they have no analysts on staff to crank out spreadsheets or discounted cash flow valuations. I’m not even sure these two guys know how to use Microsoft Excel to model financial projections (probably not). How can
two people out-research a staff of 100 highly-trained CFAs using the latest
and greatest computer technology?

The answer is that Buffett and Munger are able to filter out the noise better than just about anyone else. Based on the shareholder letters Buffett has written, it appears that he does not get distracted by unanswerable questions and daily noise, at least to the same degree as everyone else. He and Munger avoid researching companies they can’t easily understand. They ignore questions that can’t be answered, such as which direction the economy is headed this year. They don’t waste time.

Instead, they focus on just a few things: the integrity and talent of
a company’s management team, the size of its economic moat and whether its valuation seems reasonable in light of the cash flows it is throwing off today. They don’t try to predict the future except in a very general way. They
stay within their circle of competence and ignore all else. They don’t get distracted by shiny objects – or Bloomberg terminals.

The writer is a former equities strategist at Morningstar who manages a hedge fund, Sellers Capital, in Chicago.

Copyright The Financial Times Limited 2017. All rights reserved. You may share using our article tools. Please don't copy articles from and redistribute by email or post to the web.