Technicians often talk about moving averages. Ten-day, 20-day, 50-day, 200-day, etc. I mean, we could just sit here and list a bunch of numbers, whether Fibonacci-related or round numbers and talk about their significance. Bob Byrne and I often mention to each other about how bad things happen under the 50-day simple moving average (SMA), so I thought I'd take a look at a more basic approach.
How would a trader do if they simply put $1,000 into every stock in the S&P 500 over or under the 50-day SMA and held for either one month or six months? There are no stop losses, no profit taking, no scaling, no adding, nothing. This is just buy and hold on the same day and sell exactly one month or six months later, but only for the stocks in the S&P 500 index. I did not factor commissions into this exercise either, although it is something one would have to consider.
Heading into this exercise, I assumed a trader would perform better buying stocks above the 50-day SMA rather than below the SMA. While one shouldn't have a conclusion before looking at the data, I do think it is important to note your expectations and compare them with the facts.
The split between stocks above and below their 50-day SMA one month ago was fairly even, with 260 above and 241 below. Yes, that makes for more than 500, but remember indexes do have changes over time. From those stocks trading above the 50-day SMA, 172 (66.15%) found their share price higher one month later while 88 (33.85%) were lower.
Comparing this to buying below the 50-day yielded a surprising result. Some 159 stocks that were trading under their 50-day SMA were actually higher one month later, while only 82 were lower. That translated to 65.98% higher and 34.02% lower. From a binary standpoint of either winner or loser, the percentages were almost exactly the same. In simpler terms, you basically had the same odds to make a profit or loss from a random purchase in either category. Of course, this speaks nothing to how big or small of a loss you might see, but it is an interesting note.
Ironically, in terms of average gains, while both returns were greater than either the S&P 500 ETF (SPY) or the Guggenheim S&P 500 Equal Weight ETF (RSP), the stocks below their 50-day SMA significantly outperformed stocks that were trading above it. The biggest seven risers and three largest losers were all found in the group trading under the 50-day SMA, so the next area one might check is the volatility in the group under the 50-day SMA. This is only one sample, but that fact is worth investigating further.
When we scale out to six months, we find a much larger sample size of 356 names below their 50-day SMA. Consistent with the one-month group, 239, or 67.13%, of these names found themselves trading higher after six months. The smaller sample size of 142 names trading above their 50-day SMA found only 60.56% trading higher after six months. While the group trading above their 50-day SMA six months ago did manage to outgain the SPY, a purchase of RSP would have returned almost 1% more. The biggest gains again were found in buying the names that had been trading under their 50-day SMA six months ago. A possible reversion to the mean or catch-up trade thesis, perhaps? The sub-50-day SMA again contained the seven biggest losers and 11 out of 12 of the biggest movers. The increased-volatility idea is starting to pick up speed in my view and will be worth a closer look moving forward.
While a small sample size, the results are enough to warrant further investigation. My ongoing thesis would conclude buying above or below the 50-day SMA doesn't change my chances of a pure profit or loss (binary measurement) much, but if I'm willing to tolerate greater volatility, my chances for bigger gains may come from the stocks trading under their 50-day SMA. Again, this would just be a thesis to pursue, not an actual conclusion, as I would need to include many additional data points. Just a little food for thought as we head into the summer doldrums.
