Moving Averages: How They Should Really Be Used

Market commentators seem to love talking about moving averages. It seems like major market averages are always near an important moving average. One reason for this is that there are a lot of important moving averages, or MAs.

Traders often follow the 200-day MA to assess the direction of the long term trend. They might add the 50-day MA to their charts to track the intermediate term trend and the 20-day MA can be used to determine the short term trend of the market.

The result is a chart that looks like the one below.

Looking closely at the chart, we see that prices frequently move through the shorter MAs while occasionally cutting through the longer term average. We also see that the MAs tend to cross each other occasionally. While this is interesting, it doesn’t really tell us much about what an MA is or how it can be used.

Defining an MA

The MA was one of the first indicators technical analysts developed. They were already widely used when they were written about in the standard textbook of technical analysis, the 1948 edition of Technical Analysis of Stock Trends by Robert Edwards and John Magee.

This was the first edition of their book, which is now in its tenth edition. Edwards and Magee called MAs “automated trend lines” and described how they felt when they first saw this tool in action.

The authors were “still filled with starry eyed ignorance” and were searching for the “sure, unbeatable formula” that would allow them to “apply the magic and telegraph our broker periodically did from Nassau, or Tahiti…”

The automated trendline looked like their ticket to paradise but, they quickly learned that “it was too good to be true.”

Other investors have been fooled by the apparent profits a strategy based solely on an MA can deliver. The next chart provides an example.

This is a weekly chart of the SPDR S&P500 ETF (NYSE: SPY).  We selected a weekly chart to show a large amount of data and used the 40-week MA. This closely matches the performance of a 200-day MA on a daily chart.

Many traders see the long trend on the right side of the chart where prices have been above the MA for more than a year. They also see long trends, both up and down, elsewhere on the chart. Some traders notice the large number of crossovers near the left side of the chart and at other spots.

With testing, we can determine whether or not the long trends overcome the short trade signals.

Putting MAs to the Test

The chart above provides a perfect test for the MA strategy. The chart begins in April 2004. A buy and hold investment would have led to a total return of 192%. The buy and hold strategy suffered a loss of more than 55% in the bear market that ended in 2009.

The trader buying when SPY closes above its 40-week and selling when the ETF is below its 40-week MA did not do as well. This trader made 22 trades instead of 1 like the buy and hold investor did. Just 40.9% of the trades were winners, a typical win rate for an MA system.

The low win rate is caused by whipsaw trades. These are trades that last just a few days or a few weeks while prices are in a trading range. These trades can deliver a small gain or a small loss but there are so many of them that whipsaw trades are a serious drag on performance.

Using a cost of commissions of just 0.25% in our back test, the 40-week MA delivered a total return of 52%, about a quarter of the market gain.

Now, the MA strategy did avoid the bear market of 2008 and 2009 and the worst drawdown was just 11%. That may be the biggest advantage of the MA. It allows an investor to avoid the worst of a bear market.

This test is typical of how an MA strategy performs. Win rates will typically be near 40% and the system will underperform the market in the long run, assuming the long run is sufficiently long.

A Popular Myth

Many long term investors believe the 10-month MA has been proven to beat the market. This was demonstrated in The Ivy Portfolio by Mebane Faber. The results presented showed that the 10-month MA strategy beat the market through 2008.

However, since then, the system has lagged the market. The market, no matter which market average is used with the strategy, made a large up move from the bottom in 2009 before a buy signal was given. This difference erases the outperformance over the history of more than 100 years.

The data in 2008 did create a myth of how an MA could beat the market. But, it simply depended on the time period used. The truth is that none of us are investing with a 100-year time horizon. We are investing for a shorter time horizon, somewhere between a few years up to perhaps 30 years.

In that time frame, the problem is clear. Sometimes the MA will beat the market and at other times it will not. The difference is whether or not we are in a bull market. In a bear market, the MA strategy does better. In a bull market, the buy and hold strategy generally does better.

A Better Way to Use MAs

Rather than using MAs as an all in or all out approach to market timing, they can be used to determine how aggressive a strategy should be. This can be done with MA envelopes which are shown in the next chart.

This chart shows two lines. Each is 5% away from the MA. In this case, the 40-week MA is used and the top line is 5% above the MA while the bottom line is 5% below the MA. This technique is, in some ways, similar to Bollinger Bands which use standard deviations instead of percentages to form the bands.

When percentages are used, the indicator is often called an MA envelope. As the chart shows, there are extended periods of time when the price will be above or below the envelop lines. These are times to be aggressive.

For example, when the price is more than 5% above the MA, as it has been for most of the past year, a leveraged ETF could be added to the portfolio. This would enhance returns when the market is clearly in a strong uptrend.

When the price action is between the envelope lines, an investor might want to avoid leverage or more aggressive investments. That could be the ideal time to add income stocks or large cap names to a portfolio.

When prices are below the lower envelope line, a bearish or inverse ETF could be added to the portfolio. This could deliver a gain even the worst bear market.

Although a 5% envelope is shown in the chart, there is nothing ideal about this value. Very conservative investors could set the bands at 8% or 10% to minimize the use of leverage or inverse funds. An aggressive investor could use 3% envelopes.

Asymmetric envelopes could also be used. For example, a 3% upside envelope could be paired with a 5% downside envelope. This would result in more upside leverage and could be appealing to some investors.

Envelopes are a flexible tool that may not be used by many investors. But, they offer possible strategies for investors who understand the limitations of a moving average.