The moving average is one of the oldest and most popular tools used by technicians. Its strength is as a trend-following device which offers the technician the ability to catch major moves. Thus, it is utilized most effectively in trending markets. However, since moving averages are lagging indicators they can catch a trend only after it has turned.


The most basic of the moving averages is, as the name implies, the simple moving average. This is the average of all the price points used. For example, let us say that the last five gold closing prices were $380, $383, $394, $390, and $382. The five-day moving average of these closes would be

The term “moving” in moving average is applicable because, as the newest data is added to the moving average, the oldest data is dropped. Consequently, the average is always moving as the new data is added.

As seen in the simple moving average example above, each day’s gold price contributed 1, to the total moving average (since this was a five-day moving average). A nine-day moving average means that each day will only be % of the total moving average. Consequently, the longer the moving average, the less impact an individual price will have on it.

The shorter the term of the moving average, the closer it will “hug” prices. This is a plus insofar as it is more sensitive to recent price action. The negative aspect is that it has a greater potential for whipsaws. Longer-term moving averages provide a greater smoothing effect, but are less responsive to recent prices.

The more popular moving averages include the four- , nine- , and 18-day averages for shorter-term traders and the 13- , 26- , and 40-week moving averages for position players. The 13- and 40-week moving averages are popular in Japan. The spectrum of moving average users runs from the intra-day trader, who uses moving averages of real-time trades, to the hedger who may focus on monthly, or even yearly, moving averages.

Other than the length of the average, another avenue of analysis is based on what price is used to compute the average. Most moving average systems use, as we did in our gold example, closing prices. However, moving averages of highs, lows, and the mid-point of the highs and lows have all been used. Sometimes, moving averages of moving averages are even used.


A weighted moving average assigns a different weight to each price used to compute the average. Almost all weighted moving averages are front loaded. That is, the most recent prices are weighted more heavily than older prices. How the data is weighted is a matter of preference.


The exponential moving average is a special type of weighted moving average. Like the basic weighted moving average, the exponential moving average is front weighted. Unlike other moving averages, though, the exponential moving average incorporates all prior prices used in the data. This type of moving average assigns progressively smaller weights to each of the past prices. Each weight is exponentially smaller than the previous weight, hence, the name exponential moving average.

One of the most popular uses of the exponential moving average is for use in the MACD (Moving Average Convergence-Divergence). The MACD is composed of two lines. The first line is the difference between two exponential moving averages (usually the 26- and 12-period exponential moving averages). The second line of the MACD is made by taking an exponential moving average (usually a 9 period) of the difference between the two exponential moving averages used to make the first line. This second line is called the signal line. More about the MACD in Exhibits 13.7 and 13.8.


Moving averages can provide objective strategies with clearly defined trading rules. Many of the computerized technical trading systems are underpinned on moving averages. How can moving averages be used? The answer to this is as varied as there are different trading styles and philosophies. Some of the more prevalent uses of the moving average include:

  1. Comparing the price versus the moving averages as a trend indicator. For instance, a good gauge to see if a market is in an intermediate-term uptrend could be that prices have to be above the 65-day moving average. For a longer-term uptrend prices would have to be higher than the 40-week moving average.
  1. Using the moving average as support or resistance levels. A close above the specified moving average would be bullish. A close below the moving average would be bearish.
  1. Monitoring the moving average band (also known as envelopes). These bands are a certain percentage above or below the moving average and can serve as support or resistance.
  1. Watching the slope of the moving average. For instance, if the moving average levels off or declines after a period of a sustained rise, it may be a bearish signal. Drawing trendlines on the moving averages is a simple method of monitoring their slope.

5. Trading with a dual moving average system.

The examples that follow use various moving averages. They are not based on optimum moving averages. An optimum moving average today might not be the optimum one tomorrow. The moving averages used in this text are widely monitored along with some which are not as widely used but which are based on such tools as Fibonacci numbers. The moving averages used here are not the important point. What is meaningful is how moving averages can be melded with candlesticks.

I like using the 65-day moving average as a broad spectrum moving average. From my experience, it seems to work well in many of the futures markets. Exhibit 13.1 illustrates a 65-day moving average that offered support to the market at areas 1, 2, and 3. Beside the moving average shoring up the market at these points, we see a bullish engulfing pattern at area 1, a hammer and harami at 2, and another hammer like line at 3.

Exhibit 13.2 reveals that a confluence of technical factors joined on April 2 and 3 to warn alert eyes of trouble ahead. Let us take a look at the specifics:

  1. In early March, prices broke under the 65-day moving average. From that point, the moving average became resistance.
  1. The two candlesticks on April 2 and 3 formed a dark-cloud cover. This dark-cloud cover was also a failure at the moving average’s resistance area.
  1. April 3 was not only a dark-cloud cover and a failure at a moving average, but it was also within 7 ticks of a 50% retracement of price decline A-B.

Exhibit 13.3 shows that late February’s test of the 65-day moving average support line was confirmed with a hammer. The market retested these lows a few days later and, in the process, formed a tweezers bottom.


There are many ways two moving averages can be used. One way is as an overbought/oversold indicator or oscillator. This indicator is obtained by subtracting the shorter-term moving average from the longer-term moving average. This indicator has plus or minus values. Thus a value above 0 means the shorter-term moving average is above the longer-term moving average. Anything under 0 means the shorter-term moving average is less than the longer-term moving average. In doing this, we are comparing the short-term momentum to a longer-term momentum. This is because, as discussed earlier, the short-term moving average is more responsive to recent price activity. If the short-term moving average is relatively far above (or below), the longer-term moving average, the market is said to be overbought (or oversold).

Another use of two moving averages is to monitor crossovers between the short-term and longer-term moving averages. If the shorter term moving average crosses the longer-term moving average, it could be an early warning of a trend change. An example would be if a shorter-term moving average crosses above a longer-term moving average. This is a bullish signal. In Japan, such a moving average crossover is called a golden cross. Thus, if the three-day moving average crosses above the nine-day moving average it is a golden cross. A dead cross in Japan is the opposite. It is a bearish indication which occurs when the shorter-term moving average crosses under the longer-term moving average.

For a short-term overbought/oversold indicator, some technicians monitor the current close in relation to the five-day moving average. (See Exhibit 13.4.) For instance, if copper’s five-day moving average is $1.10 and today’s close is $1.14, copper would be $.04 overbought. In this example, the lower graph’s line is made up of the five-day moving average subtracted from the current close. As can be seen from this chart, when this dual moving average line gets about 400 points (that is, $.04) overbought the market become vulnerable-especially with bearish candlestick confirmation. At time period 1, an overbought reading coincided with a harami; at period 2, it hit another harami; at 3, it hit a doji; and at 4, it hit another harami. A market can relieve its overbought condition by selling off or by trading sideways. In this example, time periods 1 and 3 relieved the overbought situation by easing into a lateral band. Periods 2 and 4 saw selloffs. Overbought markets usually should not be shorted.

Instead, they should be used by longs to take defensive measures. The reverse is true in oversold markets.

Two moving averages can be plotted as two lines overlaid on a price chart. As previously mentioned, when the shorter-term moving average crosses above a longer-term moving average it is called a golden cross by the Japanese and is a bullish indication. Exhibit 13.5 has a bullish golden cross and a fry pan bottom. The fry pan bottom was confirmed by the window on July 2. Note how the window became support in the first half of July and how the shorter-term moving average became support as the market rallied.

Dual moving average differences are also used as a divergence vehicle. As prices increase, the technician wants to see the short-term moving average increase relative to the longer-term moving average. This would mean increasing positive values for the moving average difference line. If prices advance and the difference between the short- and long term moving averages narrows, the market is indicating that the shorter term momentum is running out of steam. This suggests an end to the price advance.

In Exhibit 13.6, we have a histogram between two moving averages. During time periods 1 and 2, advancing prices were echoed by a widening differential between the short- and long-term moving averages. This means the shorter-term moving average is increasing more quickly than the longer-term moving average. This bodes well for a continuation of the uptrend. Time period 3 is where the market experienced problems. The $.50 rally, which began February 23, was mirrored by a narrowing of the moving average differential. This reflects a weakening of the short-term momentum. Add to this the dark-cloud cover and you have a market vulnerable to a price pullback.

The histogram also displays when the short-term moving average crosses above or below the longer-term moving average. When the histogram is below 0, the short-term moving average is under the long-term average. When it is above 0, the short-term average is above the long term moving average. Thus, an oscillator reading under 0 represents a bearish dead cross; above 0 would be a bullish golden cross.

There was a golden cross at time frame A. A few days before this golden cross there was a bullish inverted hammer. At B, there was a dead cross. At time frame C, prices had rallied but the short-term moving average could not get back above the longer-term moving average (that is, the oscillator remained under 0). In addition, a bearish signal was sent when the dark-cloud cover formed on April 2 and 3.

The MACD has two lines. They are shown on the lower chart in Exhibit 13.7. The more volatile solid line is the signal line. A sell signal occurs when this signal line crosses below the dashed, less volatile line. In this example, the bearish implications of the two bearish engulfing patterns were corroborated by the bearish crossovers of the MACD indicators
(see arrows).

In Exhibit 13.8, the signal line of the MACD pushed above the slower moving line in early July (see arrow). This was a notable clue that the market might be bottoming. Shifting to the candlesticks shows that the first morning star’s bullish implications were voided by the dark-cloud cover. The price decline from this dark-cloud cover ended with another morning star. After a temporary set back with the hanging man, the market’s upward force gained steam.