Cycle-based Market Timing

Traders today have a monumental task. Never before has there been so much volatility in the markets as we see today. Because of this reality, greater precision in market timing is a strong prerequisite to compete, survive and profit in these markets.

For speculators, those considered “short-term” traders as opposed to investors (longer-term), precise timing is imperative. Only those able to achieve effective timing with strict discipline (emotional control) will emerge victorious. Trading well requires that you be able to do several things well. All things required can be summed up into 3 main requirements: 1. Discipline (emotional component) 2. Money Management (financial component) 3. Market Timing (analytical component) It is the 3rd listed component above that this article focuses on. And since there are many different ways to time the markets, it is the one that I find the most valuable that will be addressed here; cycles.

For many traders, the subject of cycles is greatly misunderstood. The reason for this is that the most common cycle analysis taught by trading instructors is that of the ‘fixed-length’ variety.

What are ‘fixed-length’ cycles? Well, in reality, a cycle is technically ‘fixed’ at some interval, whether it be 10 days, 30 weeks, 90 minutes or 900Mhz (900 million cycles-per-second).

Most traders are taught to count the bars from one top to the next, or one bottom to the next, and note if it repeats. If it appears to repeat, then the trader will use that information to anticipate the next top or bottom, and hopefully the next.

The trouble with this approach, which is commonly taught, is that as soon as you find a repeating ‘fixed-interval’, it changes and can result in poor timing and trading losses. This is why when most traders hear about ‘cycles’, they picture how unreliable this type of analysis is and may not want to deal with it.

But that would be unfortunate. This is because all price patterns are the result of cycles. Most traders, even those who consciously do not want anything to do with cycles, find indicators such as the Stochastic, the RSI, the MACD and moving averages to be useful. In essence, these indicators often expose the dominant cycle for the period analyzed.

Figure 1 above is a Stochastic oscillator, a very popular analytical tool among traders. Notice how this indicator swings from top to bottom to top and so forth. It is clearly a cycle pattern, produced by the price action of the market analyzed.

If you note that market making bottom when this indicator makes bottom, and make top when the market makes top, you can with some minor degree anticipate when the next bottom or top will occur. Yes, it isn’t very accurate and precise, but it beats guessing and with experience you can get close and sometimes nail the turn.

The point of this discussion up to now is to bring you to the logical conclusion that, if market patterns are the result of underlying cycles, then it makes sense to use cycles to time the market.

Earlier in the discussion I brought up ‘fixed-length’ cycle analysis and how unreliable it can be. The reason it is unreliable is that every market pattern is not made up of just one fixed-length cycle, but several fixed-length cycles.

Imagine you had a rubber ball that when thrown to the ground would bounce and keep bouncing, never losing its height or distance per bounce. It would look something like this…

Now suppose you had another ball and you bounced it twice as high as the first one.

With the increase in height we also see an increase in distance between each new bottom and top. Now suppose that along the path of the bounce you were to mark how high it was from the ground in inches. In the first example, let’s say the height of each bounce is 36 inches (3 feet). And the second bounce example we’ll say each bounce is 72 inches (6 feet). If you were able to combine these two bounces together, what might the resulting bounce look like?

Figure 2 shows a rough draft of what it would look like. The dotted line is the resulting pattern from adding two patterns of different height and distance. In cycle-terms, we would refer this as two cycles of different frequency (cycles-per-second) combined to form a ‘distorted’ pattern. The resulting pattern would not look like any of its individual cycle components, but would have a pattern of its own.

In this example, we only added two different cycle patterns to arrive at our final distorted pattern. In the markets, price patterns are the result of more than just 2 cycle patterns combined, but several. This is because there are several external influences affecting the markets.

There are many external influences that occur with regular intervals. For example, every year there are 4 seasons. Every 24 hours there is day and night. For some commodities, there is the planting, the growing, and the harvesting intervals. For other commodities there is the birthing, the fattening, and the slaughtering. And for virtually everything, there is the accumulation and the distribution periods, the supply and demand periods.
With all different kinds of cyclic influences, it is no wonder that our price patterns look like this…

As you can see in Figure 3, there is clear evidence that cycles exist in market patterns. And in addition, note that the intervals are not exact but vary. If we were dealing with one cycle, these intervals would all be the same in distance. But with several cycles of various wave-lengths combined, we get price patterns that produce tops and bottoms at varying distances in time. Thus, fixed-length cycle timing is not very reliable, but if one could determine the varying distances based on the fixed-length cycle components that make up the pattern, that would be valuable.

Cycle extraction is a process that most would find no enjoyment in attempting. I have created powerful software programs that help me time the markets with dynamic cycles, and this is what is provided to my clients/members each week. But there are some simple techniques you may find useful in determining these points for yourself, and that is by using Fibonacci ratios.

For example, refer again to Figure 3.

Note the first bottom-to-bottom pattern of 11 bars. If you multiply 11 by .618, you get 6.8. Round that up to get 7 bars and that is the next bottom-to-bottom pattern you see on this chart.

Note the top-to-top pattern of 9 bars. Multiply 9 by .618 and you get 5.56. Round to 6 and you get the distance of the next top-to-top pattern.

Multiply the next top-to-top pattern of 6 bars by .618 and you get 3.7. Round to 4 and you get the next top-to-top distance in this sequence.

Of course if you kept doing this there would soon be no visible top-to-top or bottom-to-bottom patterns as the values would just get smaller and smaller. So it should be understood that this method of using Fibonacci has its limits like most other indicators. Also, you cannot expect exact results. For example, the 7 bar bottom-to-bottom multiplied by .618 equals 4.3. Normally rounded would be 4 and not 5. Yet, the next bottom-to-bottom is 5 bars and not 4. The lesson here is that you cannot expect to see the next top or bottom occur always on the exact calculated bar, but at times a bar off. That is acceptable for market timing purposes I’m sure you would agree. And note the bottom-to-bottom 11 bars pattern at the very top of Figure 3.

Before that it was 6 bars, and multiplied by 1.618 equals 9.7 (rounded to 10) and not 11. But again, 10 is just one bar difference and acceptable.

So while this is not perfect, it is one technique that can be used for simple cycle trade timing when applied with proper discipline and money management.

by Rick Ratchford
AmazingAccuracy.com