Part IV: Chapter 1

THE DAILY OSCILLATOR

The daily oscillator, unlike the weekly oscillator, is placed on an absolute scale. I want to know when the oscillator is overbought or oversold.

I defined the long term trend as being how I see prices trending on the weekly charts. The intermediate term trend is the trend that prices are making on the daily charts, and the short term trend is the trend prices are making intraday. All I need to know about intraday prices are reflected by the open, high, low and close that I see reflected in the last bar appearing on my daily charts.

In keeping with the orders of magnitude I previously explained, it can be seen that the long term trend (1 week), is 5 times the length of the intermediate term trend (1 day). The short term trend represents one order of magnitude less than the intermediate term trend. The short term trend represents approximately 5 hours of intraday trading. this varies from market to market but is close enough for my purposes. There is nothing magic about orders of magnitude.. If I were trading on an extended long term basis, I would use the monthly charts long term, the weekly charts intermediate term, and the daily charts short term. In fact, when I am trading mutual funds, that's exactly what I do.

When trading intraday, I would use the daily chart as my long term chart, the hourly chart as my intermediate term chart and a 10 or 15 minute chart as my short term chart. This method will work in any time frame.

For a daily oscillator, it is possible to use any of the popular oscillators that are available, as long as they show overbought/oversold.

It doesn't matter whether stochastics, RSI, %R, etc., is used as long as it has an absolute scale. Whichever one is picked must be used consistently. No switching back and forth. Learn to use it and to read it with all its particular idiosyncrasies. For computer trading, pick one and stay with it.

There is one other requirement RSI, %R, etc., must be set to reflect 5 days, otherwise the signals will come too late. That means that most of these studies as shown on charts coming from chart services will be unusable. They tend to show longer term oscillators for their daily charts.

If a computer program containing Stochastics is used, then set the parameters for the Stochastic as closely as possible to 5,3,3.

The object of using this methodology is to learn to see the movement of prices without the need for such an oscillator. Stochastics measures the grouping of closes. When closes group in the upper portion of a series of bars' individual trading ranges, then Stochastics will rise. When closes group in the lower portion of each bar's trading range then Stochastics will drop. These are things that can be seen by a trained eye in a manner better than what an oscillator can show.

Without a computer, here's how I calculate my daily oscillator - it is one-half of the Stochastic called %D.

close - 5 day low
------------------------------------- =%K
5 day high - 5 day low

sum 3 %K / 3 = %D

Enter high in column 1. Enter low in column 2. Enter highest high of last 5 days in column 3. Enter lowest low of last 5 days in column 4. Enter today's close in column 5. Subtract 5 day low from today's close and enter it in column 6. Subtract 5 day low from 5 day high, and enter it in column 7. Divide the number in column 6 by the number in column 7. This is %K. Enter %K in column 8. Sum last 3 day's %K. Divide sum by 3 and Multiply by 100. This is %D. Place %D in column 9. Plot %D on the daily chart above and below a 50% line, on a scale from 0% to 100%. 70% or above is overbought. 30% or lower is oversold.

Here's an example (figures rounded to two places):

If desired, %K may also be plotted and then used as a full Stochastic. I don't use it that way because I'm not interested in when one line crosses the other. However there is nothing wrong with trading using the full Stochastic.

I utilize a short term daily oscillator because I want it to be responsive to the time frame in which I'm trading. I'm looking for a move that will last from two days to two weeks. If I can stay in longer, that's fine.

%D will be my daily oscillator. It tracks the position of daily closing prices within the recent range.

By trading from prices and the daily oscillator in conjunction with the weekly oscillator, I avoid the problems that come from blindly buying oversold signals and selling overbought signals as some traders do. The daily oscillator may give some good buy and sell signals when prices are in a trading range, but all too often they give too many signals and I end up getting whipsawed. Also, oscillators give premature signals when a market is trending.

I'm filtering the daily oscillator signals against the weekly oscillator signals. Then I will filter the daily oscillator signal against an intraday signal.

In effect, what I'm going to do is to give a market three tests before I enter any trade.

To show the daily oscillator and the intraday breakouts, I'm going to use a continuous chart for illustrative purposes, so that I won't have to keep changing contract months. The continuous charts accurately reflect the way the market looked at the time the trading was done. The only difference is that the prices have been adjusted to blend the months together. Each price bar from high to low is the same magnitude as it was during the months it was traded, and each price bar is exactly as it was in relation to the ones prior and subsequent to it.

The first test examines the long term trend, or flow of the market. I will enter a trade in the direction in which the long term trend is pointing as labeled (a,b), if it is not flat (c). I will never trade against the long term trend. It is my primary filter. It shows direction. I will examine each weekly segment (distance from one vertical line of the histogram, to the next vertical line of the histogram) of the oscillator, computed current to date as was shown in Part III. This first test will be my primary filter. If there is no trend long term, I will not enter any trade based upon this oscillator. I will examine the overall weekly oscillator looking for divergence - divergence of trend and divergence of tops and bottoms as was shown in Part III, Figures 11 and 12.

The second test will examine the intermediate term trend. Criss-crossing the oscillator, I've shown what the weekly oscillator (WO) segments were showing while the intermediate term trend was oscillating.

The long term trend indicates the flow of the market. The intermediate trend indicates the waves that make up the flow. I will be looking at the daily oscillator. What I will be looking for here is overbought/oversold, and direction. The daily oscillator indicates the waves that go with the major trend and those that go against the major trend. As long as the long term trend is indicating the same direction, I will take trades in which the daily oscillator is in conformance with the weekly oscillator. However I will look carefully at the waves that go against the major trend (labeled a-f), because they are my alert to get ready to enter a trade or add to my position in the direction of the major trend. The probability is that the long term trend will reassert itself, and so daily oscillations that go against the long term trend are excellent signals to get ready to enter a market. I've marked the approximate entry points with the letter E.

The third test comes from the last price bar on the daily chart, each day, as they occur. I'm looking for a breakout of either the high or the low of the last price bar for my entry. I've marked the days when that occurred in the past with the letters h(igh) and l(ow). A breakout constitutes a passing of all three tests, and gives me the highest probability for success. Amazingly, as will soon be seen, a large percentage of the trades will approximate 1-2-3 breakouts, and breakouts of Ross hooks, but in many instances I will get a jump, a head-start on the breakout. This head-start means a great deal of extra money in my overall trading. I've marked the chart to show which breakouts were actually used to enter the trades.

This same advantage of getting in ahead of the 1-2-3 breakout is just as easily accomplished by learning to read the detail on the price bars of a chart. I suggest beginning to study the relationships of closes to overall price action as soon as possible, as well as studying the relationship of long, intermediate, and short term trends. The two oscillators I use should help in accomplishing that. Keep in mind that such oscillators are no more than a crutch. The objective is to throw away the crutches as soon as practical.

REVIEW OF THREE-WAY TESTING

My strategy will be that when the oscillator segment is pointing up, but the daily oscillator falls, it will give me a signal to get ready to enter a buy order. When the weekly oscillator segment is pointing down, but the daily oscillator rises, it will give me a signal to get ready to enter a sell order. In other words, every wave that goes against the major trend alerts me to an opportunity to get aboard that trend if it continues.

A segment of the weekly oscillator is up or down when it moves .01 or more in value.

If the daily oscillator has already been ob/os when the signal comes from the weekly oscillator segment, I will immediately enter the trade provided that the daily oscillator has not already gone ob/os at the other extreme, that is, in the direction of the trend.

The short term trend is the third and deciding factor in entering a trade. By examining the breakout of the latest intraday trading range, I can get help in making a timely entry into the market. In this case, I'm using intraday action to point out logical entry points.

The weekly oscillator shows me the direction of the flow of prices - the major trend. The daily oscillator alerts me to the wave that is moving counter to the flow of prices - the intermediate trend. Now the intraday breakout highlights a logical entry point and shields me from the risk of entering the trade too soon.

My orders are placed strictly at the breakout points - either high or low. Initial stops are placed as protective stops, or even reversing stops, and when going long are placed one or two ticks below the low of the day, or the low of the previous day, according to amount of risk I am willing to take. When going short, protective or even reversing stops are placed one or two ticks above the high of the day or high of the previous day according to the amount of risk I am willing to take.

As a rule, these stops will not be hit because I'm trading in the direction of the major trend. If they are hit, it reflects a fundamental change in the market and I will want to try to make a profit from the new direction, or at least recoup some of my losses. I don't want to give the trade much room to work itself out. It has to succeed quickly or I want out, either with a reduced loss or a profit.

Taking profits. I move my protective stop to breakeven as soon as possible. If the trend continues to move in my favor, I move my stop progressively to the number 3 (III) points (Part II of the manual). What I'm doing then is waiting for a retracement to take place, and then when the trend continues, I place my stop in proximity to where the retracement ended and the trend began to continue. I can also add to my position at the continuation of the major trend.

Part IV: Chapter 2

A REFINEMENT

To this point, I've shown the basic concept. Now I will show a refinement that results in additional trading opportunities. This refinement requires repeated study in order to fully grasp what I'm doing.

The refinement is not necessary to the basic method. It not only increases the number of times I win, but also the number of times I lose. However, it increases the dollars won more than it increases the dollars lost. To win with this refinement, I have to be quick. Many times the trade is good only for a day or two.

Here are the rules for the refinement:

If the weekly segment is pointing up, and the daily oscillator has exited overbought territory and still pointing down, or is flat, I start placing day orders to buy just above the previous day's high. I simultaneously place a protective sell stop 1 tick below yesterday's low. In this case I have not waited for a full correction of the daily oscillator all the way down to oversold. Prices may not have fully corrected.

If the weekly segment turns up while the daily oscillator is in oversold territory and is pointing up, or is coming out of the oversold area, I place a buy stop order just above the previous day's high. I simultaneously place a protective sell stop 1 tick below yesterday's low. I have not waited for the daily oscillator to correct to oversold and then come back out. Prices may not have fully corrected.

If the weekly segment turns up while the daily oscillator is in oversold territory and still pointing down, or is flat, I start placing day orders to buy just above the previous day's high. I simultaneously place a protective sell stop 1 tick below yesterday's low. Again, I have not waited for a full correction of the daily oscillator to oversold and then back out of oversold. Prices may not have fully corrected.

If the weekly segment is pointing up, and the daily oscillator has previously entered and then exited the oversold area, but is still short of being in the overbought area, I place a buy stop order just above the previous day's high. I simultaneously place a protective sell stop 1 tick below yesterday's low. This may take place without a prior correction of prices other than that they have moved sideways.

The totality of these four refinements is that I will start placing orders to buy a breakout of today's high as long as, or as soon as, the weekly oscillator is pointing up, even though prices have not fully corrected yet as seen on the chart, or as shown by the daily oscillator.

Of course, I do just the opposite for shorting the market.

With the plain vanilla version of this method, when I want to go long, I wait for a full correction of the oscillator into oversold and then back out again before entering buy orders. As the daily oscillator approaches oversold, that is my signal to get ready to buy. With the refinement, I start trying to buy a breakout of the high as soon as the daily oscillator has corrected from overbought.

With the plain vanilla version of this method, when I want to go short, I wait for a full correction of the oscillator into overbought and then back out again before entering sell orders. As the daily oscillator approaches overbought, that is my signal to get ready to sell. With the refinement, I start trying to sell a breakout of the low as soon as the daily oscillator has corrected from oversold.

There is no absolute rule as to when everything seems right to do this, or as to just where the daily oscillator might be when I start entering the orders. I start placing orders as soon as I feel that the market will start correcting, and that it is safe for my account. Part of my decision will be according to how long the major trend has been going, how steep has been its ascent or descent, whether or not the daily oscillator is in overbought or oversold, and how long it has been there.

The refinement is as much a judgment call as you care to make it. The main thing is that you are trying to trade in the direction of the major trend. You can build your own set of rules around it based upon your own comfort level and experience.

The refinement will yield a great many more trades. Without the refinement many good moves will be missed.

Some have asked, "Will I ever see a weekly oscillator pointing up at the same time the daily oscillator has corrected and has also turned up?" The answer is yes, but it may happen only a few times in a year for any market -- maybe not at all for some. However, when it does happen, it means that the trend is relatively longer term for that particular market and the vast majority of your trades will be successful. If that is your comfort level, then I suggest you trade it that way.

Shows the continuous daily Crude Oil chart. The additional trades using the rules for the refinement are the ones marked 1-3. I've shown the rules that applied to these trades.

Rule #1, If the weekly segment is pointing up, and the daily oscillator is in oversold territory and is pointing up, or is coming out of the oversold area, I place a buy stop order just above the previous day's high. I simultaneously place a protective sell stop 1 tick below yesterday's low.

Rule #2, If the weekly segment is pointing down, and the daily oscillator is in overbought territory and still pointing up, or is flat, I start placing day orders to sell just below the previous day's low. I simultaneously place a protective buy stop 1 tick above yesterday's high.

Rule #3, If the weekly segment is pointing down, and the daily oscillator is in overbought territory and is pointing down, or is coming out of the overbought area, I place a sell stop order just below the previous day's low. I simultaneously place a protective buy stop 1 tick above yesterday's high.