MACD DOTS By Phil McGrew

MACD DOTS By Phil McGrew pmcgrew@myacc.net



Hello dear reader. I want to take a few minutes of your time to explain a few things about the dots that may or may not be intuitive. They are:

• What is MACD and how is it developed?
• Why I think MACD is the best indicator around
• How the MACD dots were developed
• What other indictors should be considered
• How to trade the MACD dots as a system


I think very few of us can actually explain how MACD is developed. It is of utmost importance to understand your indicator. All have limitations and if you don’t know and appreciate them you’ll soon be looking for another indicator.



Figure 1

In figure 1 we begin to develop the MACD. It consists of a 12 period EMA (olive line) and a 26 EMA (purple line). The difference between those 2 lines is the “Fast Line” (yellow line). Now we have an oscillator. The only difference between this and other oscillators such as Stochastics, RSI, and others, is that they are normalized in a way that they are usually bounded by 0 on the bottom and 100 on the top. They “oscillate” between those two extremes. There are limitations to that type of oscillator which we will later. But, MACD isn’t done yet.


Figure 2

The “Fast Line” (yellow line) is smoothed again by a 9 period EMA to create a “Slow Line” (blue EMA). The difference between the Fast Line and the Slow Line is often plotted as a histogram (Green Plot). Smoothing the Fast Line and creating an additional moving average is why MACD is often referred to as an “oscillator of an oscillator”.

Personally, I think MACD is the best and most versatile indicator around. By observing MACD you can tell 4 things about price action:

• The trend of price action. By observing the relationship of the Fast Line and the Slow Line we can tell the direction of the market. If the Fast Line crosses above the Slow Line the trend is up. This is the premise of a moving average crossover system.
• Divergent situations. By comparing neighboring peaks and valleys of the histogram we can identify areas of regular and hidden divergence. If you do not understand divergence read this.
• Momentum. When the market makes a move the Fast Line and the Slow Line separate. The difference can be seen on the histogram. When this movement subsides the lines come back together and the histogram approaches zero. We can observe the histogram rolling over, or rolling up, towards zero. This is an indication that momentum is drying up.
• Market noise. If the market is going sideways there will be no separation between the Fast Line and the Slow Line. The histogram will necessarily be very close to 0. This is a good time to stay out of the market or look for opportunity when price breaks out of the existing range.

I started this system by trying to develop an indicator that would alert me to MACD crossovers. This is the Fast Line crossing the Slow Line or vice versa. I found that, in many cases, the move was over or well under way when the crossover occurred. I wanted a way to anticipate the crossover. When momentum stalls, the histogram ceases to continue in a given direction. I thought if I could identify turns in the histogram, I could enter the market earlier than if I waited for an actual crossover of the lines.




Figure 3

In figure 3 I colored the histogram green when rising and red when falling. For me, this makes it more visually apparent which way momentum might be heading. I included red arrows to show where momentum has peaked. The blue arrows show that downward pressure is slowing and that a reversal is possible. On the figures to follow, the charts will have yellow dots above price to indicate a sell signal and blue dots under price to indicate buy signals. They will approximately agree with the red and blue arrows I used in Figure 3.

There are only two problems left to solve:
• What if the market isn’t moving?
• How do I reduce false signals?




Figure 4

Figure 4 shows how both of these problems are addressed in the code. In the first third of the chart and the last third of the chart no signals are given (no yellow or blue dots). Sideways market movement is filtered out by requiring that the MACD histogram be above or below a certain height before any further calculations are permitted. The horizontal lines show the value above or below that is necessary to register a signal.

The lower vertical line shows the first occurrence of the histogram beginning to fall. The upper vertical line shows the price at which the actual sell signal was generated. The code records the height of the histogram. When the value of the histogram begins to fall it must exceed a preset differential before a signal is given. This prevents a signal from occurring too early or a spurious bar from generating a signal.

Figure 5 represents a typical chart setup for me.




Figure 5

The yellow dots are sell signals, the blue ones are buy signals. The blue moving average is the 15 min 62 period EMA (this is a 15 min chart) and the white dashed line is the 62 period EMA from a 1 hour chart. I use a time of 4 times higher than the chart I am looking at to get a sense of the overall trend. It is very important to know if you are trading with or against the trend. Money can be made in either direction, but when going with the trend the probability of success is greatly improved.

The bottom indicator is an oscillator. I use one called RSX but any oscillator that is capable of giving overbought or oversold indications will work well. If a buy signal is generated concurrent with an oversold condition, or a sell signal is concurrent with an overbought condition, the odds are again improved. One last discretionary pattern that I suggest using are candle patterns. As price nears an extreme, the candles should become smaller and reversal candles should begin to show up. These are doji, hanging men, hammers, inverted hammers, shooting stars, and spinning tops.




At this point it’s not much of a mystery how these dots can be traded as a system. There are additional indications that can be used to increase the odds of success. These are:

• Trading concurrently with an overbought or oversold condition
• Trading with the prevailing trend
• Looking for reversal type candle patterns
• Looking for obvious levels of support and resistance where a reversal is likely to occur
• Looking for divergence and hidden divergence. There is an example of this directly in the middle of figure 5.
• Waiting for the signal candle (the one with the dot) to be exceeded prior to taking a trade. Sometimes this can keep you out of a lousy trade. If you are only looking for a quick profit, waiting may not be prudent since several points could be a large percentage of the gain you are looking for. But, if you are looking for larger gains waiting to see that the trend has indeed reversed can often be helpful.

My trading partners and I have recently experimented with scaling into these positions. It is foolish to think that any of us can pick the exact top or bottom of any reversal. If price improves in a way that you can add additional positions (or goes against you if you are a pessimist) and improve your effective entry price, then it might be a good idea to do so.

I have several people to thank for helping me with this concept. My girlfriend Sherry who suggested no less than 50 grammatical “improvements” in this paper (even though it is mostly charts) and forgives me when I occasionally trade too much and forget to shower. My friend Holden who actually knows how to write code and converts ideas into strange languages that my computer will understand. And my trading partners and dear friends Rob and Max who have helped develop this idea into a profitable system.

The code is written in EasyLanguage which is proprietary to TradeStation. The code is available for free upon request. If there are questions, comments, or improvements to this system I can be reached at pmcgrew@myacc.net.

Also, this code will implemented into Xtick from version 3.24.3