The Right Way to Trade SPY's 200-Day Moving Average

A couple of thoughts As The Yuan Turns, and as we hover near the 200-day simple moving average in the SDPR S&P 500 ETF Trust (SPY):

I rarely care much "why" the market moves. I kind of go on this theory that we err way too far on the side of assigning causality to every blip. It's just way more complicated than "News Story X causes Market Reaction Y." If nothing else, timing matters.

Take this week's Official Cause of Market Ugliness: China devaluing the yuan. There's zero doubt that it caused the market drop, in the literal sense. But the mistake will be assigning future causality to the same news story if it happens again.

What I mean is -- right here, right now, it took the world markets by surprise. But the big fear isn't that this is a one-off event, but rather that it's the start of a program of devaluation. So, assuming that’s the case, Simple Causality Theory (as espoused on TV) will expect the markets to get plowed again the next time it happens. And that’s very much not clear. At the very least, it's not coming out of the blue next time. At best, we'll have come to terms with it and discounted it, and we could actually rally on it.

Who knows, though ... and that’s the real point. To me, it's not so much about the actual effect of yuan moves; it's more the surprise and uncertainty about it now. In a week or a month, maybe we'll have a much better handle on how long this will go on and the ultimate impact.

An interesting (amazing) byproduct is that part of the ugliness is now fear that the Fed won't tighten in September. To me, that highlights the absurdity of the whole Fed Worry Cycle to begin with. The best-case all along was that the economy was good and world markets were stable, and the backdrop was good for the rate hike.

A bigger worry is all the technical damage going on. You've probably read a feature or three on the Dow's death cross.  It's like the Hindenburg Omen, only scarier-sounding. Does it lead to market declines? Well, that depends how you define it. The common definition is that it happens when the 50-day moving average crosses below the 200-day. But some stipulate that both moving averages have to tilt downwards ... and right now, the 200-day still has an upwards slope to it.

For what it's worth, I took a look at SPY and its 200-day. The 200-day is about 207.59 right now. We broke through significantly intraday yesterday, before the Big Reversal. We still hover nearby, on a closing basis. Long story short -- if you only held SPY on days it closed over the 200-day, you would have owned it about three-quarters of the time, and captured almost the entirety of the gains since late 1993 (we need 200 days of history to get it started). SPY had a 45 full back then, incidentally, so it's up about 450%, not including dividends.

Here's another way to look at it: What if you walked in cold and literally all you knew about the market was SPY vs. its 200-day?

I show 5,440 trading days from late 1993 through last Friday. On 3,919 of them (72%), SPY closed above its 200-day.  If SPY was >200-day SPY, and you bought and held for one month, you had an average return of 0.91%. If it was less than the 200-day, you had an average return of 0.1%. So it's pretty stark, on that basis. On a median basis, however, it's not so stark -- 1.23% vs. 0.9%.

The reason the average looks worse is pretty clear. There were a couple of huge accidents -- namely, 2002 and 2008. The whole benefit of not owning when SPY is <200-day SPY is that you avoid the occasional meltdown.

It gets even more extreme if we go out further. The average three-month return if SPY is >200-day is 2.87%. If SPY is <200-day, it's 0.09%.

Yada yada yada, most of the time when SPY goes below the 200-day, it reverses course before long and goes back above. Using it as a trading signal won't work especially well in the short term. But for longer-term investing, you'll realize most of the gains -- and sidestep the very occasional persistent ugliness -- if you raise some cash when SPY is under the 200-day.

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