There are many distortions embedded in how investors think about and evaluate “performance.”  One error comes from the presumption that analysts should change their recommendations (or portfolio managers should buy or sell) at precise moments, such as the optimal ones as related to the price performance of a stock.

This simple drawing can illustrate the danger in such an assumption.  In a different forum, I wrote about this once before:  “Implicit in most forms of performance analysis is the notion that the firm providing the rating is going to have you buy a stock while it is going up, sell it at the top, and remain out as it goes down.  That is, walk right up to the precipice, regardless of the risks, and back away at just the right moment.”

It would be nice if investment decision makers exhibited such clairvoyance, but, sadly, we know it’s more a game of trying to be slightly better than average (or random) in evaluating opportunities.  The fact that we aren’t very good at it does not necessarily mean that we should forgo an assessment that considers the ideal — there may be important things to learn from doing so.  As I have indicated in other postings,the research puzzle | Including in last year’s “the research performance derby.” taking a one-dimensional approach to performance analysis yields simplistic and often erroneous conclusions about who is “good” and “bad.”

One reason why is illustrated by the diagram above.  The goal is to buy the stock at point A or point B or really anywhere on the left side of the chart, and to sell it at point C.  Yet, if the rush from B to C is from a flowering of unreasonable assumptions and momentum for its own sake, the very best call at point B might be to “sell.”  (Since you may have to think back a couple of years to situations like this, it is helpful to remember it works in reverse too.  Hold a mirror underneath the price line and it may look more like recent charts.)

It is common for research firms to state the assumptions behind their ratings.  An example may be something like, “A BUY recommendation indicates that the analyst expects the stock to underperform the market over the next twelve months.”  Isn’t it odd, then, that most firms employing those analysts or using their ideas evaluate them without consideration for that supposed time horizon, instead calculating day-to-day metrics fitted to today’s price charts and aggregated for longer periods of analysis?

Of course, doing that for analysts (or for portfolio managers) is one reason that risk is not weighed properly in many investment decisions.  Get off of the roller coaster too early, even if all of the warning systems are flashing that danger lies ahead, and you’ll look worse than everyone else for some period of time.  If the peak is higher and farther away than you surmise, you will be thought out of touch, an old fuddy-duddy, and could lose your job or see assets under your care rush away, even if your assumptions are correct and you have chosen the proper course of action.

There is a conceit in the business that the best among us can pick the highs and the lows, and we evaluate much of our work on that basis.  The real truth is that good investors see the shifting odds of risks and returns and are willing to back away from a position as it becomes more likely to one day disappoint, even if the day of reckoning may not be arriving tomorrow.  Such an approach at least acknowledges that among the imponderables in the great list of unknowns is exactly when things will change.

Our performance evaluation techniques should do the same, rather than ignoring time as an important element in gauging our exposures and beliefs.