Why SPY Action Could Signal Short-Term Pain, Long-Term Gains

As we noted yesterday, a general strategy of going long when the SPDR S&P 500 ETF Trust (SPY) is below its 200-day moving average doesn't tend to work out so well on a relative basis. That's mainly because it puts you on the sideline for long-term bull moves, while it also keeps you exposed to short-term implosions.

But, those implosions are few and far between. We identified three of them in 21-plus years of data. So, the real key to investing, no matter how you slice it, is to avoid being too exposed when those accidents occur.

And who says I don't have a grasp of the obvious!

Of course, avoiding them isn't so simple. There is clearly intense fear that we're about to hit an air pocket now. The current proximal causes are Ebola, ISIS, Greece, the weakening economy, the end of QE and, well, I'm sure there's more. I'm a strong believer that markets move independently of actual news, and the causality correlation is all hindsight, but whatever, that's neither here nor there.

One thing I did notice going through the data: We never had a pattern where an extended gain riding long when SPY was above the 200-day was immediately followed by an extended dip after passing through the 200-day, with one sort-of exception (guess what: it involves 2008).

I bring this up because we're fresh off a blast below the 200-day, after realizing a 34.7% return for sitting long while it was above the 200-day for nearly two years.

So, let's say you went long when SPY went below the 200-day the last six times the immediately previous "SPY Over" trade yielded over 10%. Here's how the returns look over one-month, three-month and six-month holding periods:

Generally speaking, the trade doesn't work well at first. After three months it was OK, and then after six months, it panned out. Well, except once. Actually, that August 2007 experience is an anomaly all around. In that instance, going long when SPY busted below the 200-day did OK early, then awful over time.

The bazillion dollar question, of course, is whether we're more 2007 or 2011 right now. If it's 2007, you basically had two-to-three months' more time to get out before we'll have to look out below. If it's 2011, it's the opposite -- we'll have a flat-to-weak few months, and then a terrific buying opportunity. The common denominator in all the cases is that the worst of the ugliness involved the drop to the 200-day to begin with. Remember that we were coming off impressive uptrends in each case.

Disclaimer: Mr. Warner's opinions expressed above do not necessarily represent the views of Schaeffer's Investment Research.