The Pitfalls of Cycle Trading
MAGIC APPEAL OF CHANGE
There are technical analysis techniques that, for some reason, seem to have always been around, at least ever since the market stopped being associated with fish, daisy nosegays and the ugly pottery, manufactured by the local artisans. Why? Most probably because the trivial logic that governs these methods and their application is seemingly so hard-and-fast that most anyone trading the markets sooner or later gives them a try. A case in point is the legendary technique of cycle trading, extolled by classics (such as, for example, the rocket scientist Mr. James Hurst), revered by some, and ridiculed by others. So what is so attractive about this method that made it one of the methods of choice for a great many of our fellow-traders?
At first blush, if everything is calculated correctly, the technique seems to be so simple that any more or less thinking person who gets familiar with it for the first time can hardly help asking himself how come that there's still some money left out there. Indeed, what can be simpler - the markets move in cycles. The universe is boundless. Buy at rock-bottom, sell at the crests, and count the bucks. Simple, isn't it?
Well, don't get too elated, here is the full story.
CYCLES AND WHY THEY CAME INTO PLAY
It's pretty obvious to anyone who has as much as had a brush with the market that there is definitely something cyclical to its nature. However, the cyclical nature of the market is not governed by any clearly defined law - cycles form with constantly changing oscillating amplitude, frequency and phase displacements. Given the cyclical nature of the world around us, the human brain inadvertently looks for a "cyclical" explanation of the different aspects of life: the changes of seasons and governments, droughts and bumper crops, downturns and periods of economic growth.
The recurring processes in nature have been studied and used by humans since time immemorial. Our distant ancestry were able to use their observations of these phenomena to create things that are still capable of impressing us today. Sound scientific basis for the modern spectral analysis was provided in the 17-th century by Sir Isaac Newton. The field was contributed to by a number of other luminaries, such as Daniel Bournoulli, Norbert Wiener, John Tukey.
As far as trading is concerned, one cannot but mention the scientific legacy of the esteemed Jean Baptiste Joseph Fourier, the renowned French Egyptologist and Mathematician, whose paper Théorie analytique de la chaleur (1822) gave birth to the scientific notion of what is now known as the Fourier series. Another major, if the greatest, contributor to the field was Mr. John Burg, who viewed the issue from the angle of maximum entropy in his 1975 doctoral thesis. This approach laid the foundation for cyclic analysis in trading owing to the very small amount of data it required to provide spectral estimates.
In theory, to profit from the cyclic nature of the market, one needs to divide the price movement into the trend, cycle and noise. Then the periods when the trend doesn't exist need to be identified, when the cyclical component carries enough weight "to drown out" the market noise. Ideally, the trend should only be traded with when the level of the cyclical activity is low and it is only during those periods, when the cyclical activity is very strong that cycle trading can be applied.
Just like any technique called to describe the nature of the market cycles are a simplified model of the market. As the basis for the creation of a market model, the sinusoid is used. Simple sinusoids are combined to model the market's cycle nature using a range of mathematical methods, from simple cycle finders (for example, the determination of the average distance between two lows) to Fourier transforms or Maximum Entropy Spectral Analysis, reckoned to be the most efficient of the methods. As the parameters for this kind of modeling the amplitudes, frequencies and phases of the primitive cycles being combined are used. The art of cyclic analysis consists in the ability to correctly select the required combinations while taking into account the resonances and objectively determining the parameters.
The cyclic analysis approach can be used for describing classical chart patterns, determining trend channels, parameter determination methods for MAs and indicators, as well as for calculating stop loss signals.
Any more or less versed trader is fully cognizant of the fact that due to the nature of the market it is nearly impossible to identify the beginning and end of a cyclical period or correctly select the cycle frequency to trade with.
Basically, a market cycle is a repetition of a stock's or currency's average fluctuations - a low, a rally, and a new low, dividing the period of time, occupied by the cycle, into the following four stages: the initial rock-bottom/recovery stage, upward move, distribution stage, and downward move. However, even if we do wholeheartedly embrace the elegant theory of the market's cyclical nature, how do we identify the beginning of a cycle and its end? Probably, this is both the greatest challenge and deepest pitfalls awaiting anyone who treads on the risky path of cycle-trading. Besides, normally, the market rises and falls several times before any of the three major points is reached. This can confuse the trader and mislead him into entering or exiting a position too early, which, actually, happens rather often. Actually, any cycles-based trading offers an acceptable amount of risk only if the cycles are consistently repeated at least 85-90 % of the time and the smaller moves account for not more than the remaining 10-15 %. All those who attempt to come up with a viable classification of the different cycle stages and just another smart technique of how to identify them, either mentally or automatically with a specialized software, have so far failed to either convince or impress yours sincerely. Attempting to picturesquely describe the different market stages, lulls, and upsurges with the help of tenacious psychological phenomena is one thing, identifying these phenomena as they happen is a completely different matter. In fact, in our opinion, virtually the only moderately risky and realistically profitable way of trading here is what they sometimes refer to as the "greater fool trading," applied short-term within the second 1/3 of the upward move - the second and only feasibly detectable stage in a market cycle. The beginning of the second stage is signaled by the end of a prolonged recovery lull and appearance of a growing number of early buyers. If, having waited long enough to make sure the trend is getting stronger, you manage to both buy and sell short close to the middle of the upward move, there will always be someone to buy from you.
Anyhow, on a more close inspection, it becomes clear that what you can reckon to be the beginning of an exemplary and profitable market cycle at one time is nothing but too much market noise over a longer period of time. For example, a day's high is not more than plain noise if you are trading a year-long cycle. This makes the very notion of a market cycle vague and unconvincing - after all, a cycle can take as long as a couple of decades to complete. Was that high of yesterday the promising beginning of what will turn out to be a windfall fortune in 2026? The realization of the cyclic nature of flue pandemics is unlikely to be of help while answering this question.
Today, cycles remain one of the most spoken about trading-related topics. This can be chalked up to (let's admit it) most markets having cyclical nature, as well as to the marginal simplicity of cycles' structure. The seemingly irrefutable statements about the cyclical nature of life itself and just as convincing "discoveries" of some major cycles the world seems to be pillared upon (such as, for example, the four -year presidential term), do their bit too.
In addition, in theory, cycle-trading is a real cinch: some grasp of math, a little knowledge of geography (they don't raise coffee beans in the North Pole - who could know), and some general erudition will be all you will need to strike it rich. Alas, the whole thing gets a just a bit more complicated when it comes to practice.
The relative popularity of cycles in trading can most probably be attributed to yet another reason: there is large number of seasonal commodities, being, oftentimes, strategic industrial products and the staple products of the areas from which they originate. The prices of these products are cyclic in nature as are their sales. A perfect example would be some kinds of industrial fuels, consumed for heating purposes, whose prices, which is only natural, peak in the winter and toboggan in the summer.
This has allowed many of the cycle enthusiasts to develop the so called seasonal models for different seasonal commodities, which seemingly, are simply destined to make the learned minds who concocted them the movers and shakers of this world - after all, if you know when teddy bears can be flogged at double their regular price, you can do a land-office business.
But as luck would have it, the world in which these would -be movers and shakers live is round and diverse, and the effect the magic cycles produce can sometimes differ drastically from what their worshippers expect it to be. The different time zones, harvest seasons, droughts and overproduction are all there to shatter the immaculately prepared forecasts and dearly cherished dreams of cycle worshippers. All the above can hardly ever be taken into account and prevented for terms sufficient for the formation of a seasonal commodity cycle: in any event, any heart-felt attempts to eliminate a Colorado bug populace hard at work on destroying your potato-trading cycle several months after its beginning will be fated to fail. However, if you feel that you are able to take into consideration all the existing multitude of fundamental (political, geographical, psychological) factors that influence the market, you can use the cycles as an auxiliary means that shows you the general direction in which the market will move, bearing it in mind that the actual price may (and, most often, will) be significantly different. In fact, the number of the market forces is so large that the average fundamental and calendar (weekly, yearly) and production cycles turn out to be a lot more robust than the multi-layer ones, artfully calculated with the help of mathematics.
Another thing which should be remembered is that cycle-trading is much less suitable for stock-trading than it is for futures-trading.
Obviously, cycle trading can hardly be recognized a viable means of profiting on the market in the majority of cases. However, it can be applied by well-skilled professionals in some kinds of trading situations and for certain kinds of commodities as a complementary technique, for example, for long-haul agricultural futures trading.
Very often those traders who trade and "think" cycles use the same trading techniques as those who consider cycles to be a bizarre figment of their colleagues' imagination, the difference being that for these market players cycles provide a "general framework" of the market, an additional angle from which the market can be viewed. The division of the market movements into the trend and sideway periods whereby different techniques are applied during the different periods is a good example of this kind of approach.
As already mentioned above, the beginning and end of a cycle are hard to predict. This is highly disadvantageous, as this very divide is the cornerstone of cyclic analysis and even the simple realization of it's existence by a trader has a practical meaning. Often, the fluctuations of an amplitude fade just as the trend reaches the required height and the model needs to be adjusted with new parameters either manually or automatically. This takes a while and, normally, the kind of the model you receive after the adjustment can only provide a good picture of the market - alas, the trend can no longer be used. In other words, a cycle with other parameters can become dominant at any moment in time.
Modeling the market's cyclic nature, despite its individual character, is based on the gargantuan task of approximating the price time movements to analytical functions. And although the market has long stopped being a place where only medieval cooking utensils can be bought, the present-day computerized means and mathematical achievements do no allow us to use its cyclic nature to our satisfaction.