Making Moving Averages Work

All traders at some time in their career have used moving averages as a basis for a trading system. These trading systems come in a variety of disguises, from plain vanilla “buy when the price crosses of the average” to dual and triple moving average crossings. What we are going to confine ourselves are simple methods of making moving averages work more effectively.

There is no question that moving averages can and do produce impressive results. Using some variation of MA to buying or shorting during bull or bear markets is child’s play. Unfortunately, their biggest problem occurs during trending markets. The reality is that most stocks spend 70% or more of their time in a trading range. It is during these trading ranges that the MA systems show their weaknesses by whipsaws and commission expenses.

Diversification is not necessarily a solution. No doubts it will help, but many stocks are strongly correlated. Approximately, 70% of all stocks will rise and fall at the same time. If we are going to diversify, then we need to use different stocks in different fields. Money management would have to deal with adjusting the stock mix to maintain a constant volatility for each the stocks. I discuss this in the eBook, The Three Day Hammer.

The best recommendation is trade only in trending markets. Easy to say, difficult to sometimes do. I would recommend Welles Wilders concept of ADX as a reliable method of trend trading. His book, New Concepts In Technical Trading Systems goes into depth concerning the mathematics of ADX. Most charting programs have this as a built in function. Basically, ADX is a method of determining if a market is trending. An increase of the ADX signifies that the market is in a trend, up or down. A decreasing ADX can indicate that the trend is decreasing.  As a general rule, I like to see the ADX greater than 20 for MA systems.

Another approach is incorporating a channel breakout system into the MA system. These were made famous the Richard Donchian and his followers. They incorporated it into the “Turtle Trading” system. Basically, a buy confirmation occurs when a stock crosses the high of so many days and a short confirmation occurs when the stock goes below the low of so many days. Commonly used numbers include 10, 20 and 50 days. 

There are problems using long term moving averages to close a trade. Longer term moving averages react too slowly to use as exits. If you use slow moving averages you may be too late to exit and give back most of the profits.

A solution is to use a different set of moving average for both entry and exit. Use a long term moving average to determine that a trend is in place and get you into the trade. Use a fast moving average to take profits. With fast moving average, you will have better exits but using them for entries will often get you whipped sawed. You can use a longer MA for entry and a shorter MA for exit. The best of all worlds. If the shorter MA is not effective, the solution sometimes can be found in using a dollar trailing stop. I have found that it is better than allowing a moving average exit to give back profit.

Another solution that I employ is a variation of the channel breakout system. I will use a longer breakout to confirm the long position and I will use a shorter breakdown to close that position. For example, if the price of the stock exceeds the moving average that I am using and that price is confirmed by being the highest price of the last 20 days, I will buy. I will close that position if it becomes the lowest position of the last 10 days. You reverse this when going short.

There are no end to ideas and combinations that you can employ. You can combine a channel breakout close with a dollar stop. Read the article on Creating a Trading System. Just use you imagination or someone else’s. And don’t forget, back test your system, make sure it is profitable and has a positive expectancy.