From the Archives: Trend Following Beats Market Timing

May 30, 2013

Mark Hulbert has been tracking advisory newsletters for more than 20 years.  Lots of these newsletters do active market timing, so in a recent column, he asked an obvious question:

The first question: How many stock market timers, of the several hundred monitored by the Hulbert Financial Digest, called the bottom of the bear market a year ago?

And a follow-up: Of those that did, how many also called the top of the bull market in March 2000 — or, for that matter, the major market turning points in October 2002 and October 2007?

If you are relying on some type of market timing to get you out of the way of bear markets and to get you into bull markets, this is exactly what you want to know.  Although there are pundits who claim to have called the bottom to the day, Mr. Hulbert allowed a far more generous window for labeling a market timing call as correct.

… my analysis actually relied on a far more relaxed definition: Instead of moving 100% from cash to stocks in the case of a bottom, or 100% the other way in the case of a top, I allowed exposure changes of just ten percentage points to qualify.

Furthermore, rather than requiring the change in exposure to occur on the exact day of the market’s top or bottom, I looked at a month-long trading window that began before the market’s juncture and extending a couple of weeks thereafter.

That’s a pretty liberal definition: the market timer gets a four-week window and only has to change allocations by 10% to be considered to have “called” the turn.  And here’s the bottom line:

Even with these relaxed criteria, however, none of the market timers that the Hulbert Financial Digest has tracked over the last decade were able to call the market tops and bottoms since March 2000.

Yep, zero.  [The bold and underline is from me.]  It’s not that advisors aren’t trying; it’s just that no one can do it successfully, even with a one-month window and a very modest change in allocations.  Obviously, there is lots of hindsight bias going on where advisors claim to have detected market turning points, but when Mr. Hulbert goes back to look at the actual newsletters, not one got it right!  You can safely assume anyone who claims to be able to time the market is lying.  At the very least, the burden on proof is on them.

We don’t bother trying to figure out what the market will do going forward.  We simply follow trends as they present themselves.  We use relative strength in a systematic way to identify the trends we want to follow: the strongest ones.  We stay with the trend as long as it continues, whether that is for a short time or an extended period.  When a trend weakens, as evidenced by its relative strength ranking, we knock that asset out of the portfolio and replace it with a stronger asset.  The two white papers we have produced (Relative Strength and Asset Class Rotation and Bringing Real World Testing to Relative Strength)  show quite clearly that it is possible to have very favorable investment results over time without any recourse to market timing at all.  Discipline and patience are needed, of course, but you don’t have to have a crystal ball.

—-this article originally appeared 3/17/2010.  It is especially apropos now that many market pundits are busy predicting a top.  It’s certainly possible they are right—but probably equally likely is the proposition that they are just guessing.  Over the long run there is weak evidence that market timing is effective.

Posted by: