Custom Search

Traditional Moving Average Systems: Assets and Liabilities

Moving averages have been used with varying degrees of success by stock and futures traders for many years. Although there are literally hundreds of possible variations on the theme of moving averages, the fact is that moving-average-based trading systems, regardless of their specific type, have some distinct limitations and only a few advantages.

Moving-average systems are trend-following systems. They get traders on board once a move has started based on the averages, and they either exit positions (i.e., "go flat") when a move has ended or reversed. MAs perform well when markets are in trends. However, in trendless or sideВэways markets, moving averages experience their greatest limitations and downfall. Since markets are in strong trends perhaps only 30 percent of the time, you'll find that most moving-average-based systems are accuВэrate (i.e., correct) between 20 and 50 percent of the time. This relatively low accuracy rate, however, does not mean that MA-based systems canВэnot make money. They can, provided the essential rules of risk manageВэment and good trading are carefully followed. Before I discuss these, however, I'll give you some of the pros and cons of MAs.

As previously stated, the problem with moving-average-based sysВэtems is that they are not very accurate. Moving averages are lagging indicators. A lagging indicator, as its name suggests, is an indicator which follows the market and which, by its very nature, does not change direction until after the market has changed direction. The positive aspect of such indicators is that they frequently do not change direction until a new trend is under way.

The downside, is, of course, that what the moving-average perceives to be the start of a new trend may actually prove to be a very brief trend or it may be nothing more than a random variation or "hiccough" withВэin the existing trend. But most moving-average systems aren't smart enough to distinguish between a real trend or a brief variation within the existing trend. Hence, they generate many false signals. Moving averages are basically deaf, blind, and dumb indicators. They say to the market, "lead me and I will follow." Provided the market establishes a meaningful trend, moving averages will do well. Yet, in the absence of a meaningful trend, or in a whipsaw type market, moving averages will suffer terribly, taking loss after loss after loss.

Another severe limitation of moving averages is the fact that in order to immunize them from false turns, they must necessarily sacrifice senВэsitivity. It is not, therefore, uncommon for a considerable amount of profit to be given back in awaiting an MA signal to exit an existing posiВэtion. Similarly, new positions will often be entered well after a trend has changed, therefore, also sacrificing a good deal of the potential profit. Although there are some ways to minimize this problem period, it is, nevertheless, a serious one.

Yet another significant problem with moving-average-based systems which arises from the two foregoing limitations is that the systems are frequently incorrect. Most moving-average systems are correct between 20 and 40 percent of the time, with the upper end of this range being the exception rather than the rule. Yet, in spite of these limitations, moving-average-based systems continue to be very popular among technical traders and in particular, among fund managers. Why? There are severВэal reasons: First, the use of moving averages requires very little in the way of sophisticated mathematics. Basic moving averages can be readily computed without the assistance of a computer and can be determined quickly, both on a daily as well as on an intraday basis. Second, moving-average-based systems provide specific trading signals which are a function of moving averages and/or prices crossing above or below one another. This is ideal for the strictly mechanical trader. Third, moving average systems in most cases are always-in-the-market systems. This means that with few exceptions, these systems go from long to short and short to long and rarely maintain a neutral position. The idea of always having a position in the market appeals to many traders who know that some of the most spectacular and potentially profitable marВэket movements occur when traders do not have positions. Fourth, many traders fear that by not having a position, they take the risk of not being able to establish one at a reasonable entry price once a major price move has started. (Of course, commodity fund managers who share in the commission income of their funds love MA systems which trade freВэquently. After all, they can't help but have a vested interest if they parВэticipate in the commissions.) Yet a fifth reason for the popularity of movВэing averages is found in their simplicity. Moving averages do not require a great deal of understanding about the markets, yet they offer the attraction of a mathematical model which rings the bell of money managers who have a public image to maintain.

Because moving averages are so amenable to quick calculation, and because the signals derived from moving averages are so precise (do not confuse precise with accurate), they lend themselves readily to applicaВэtion for short-term and day trading.

This chapter will focus on a number of specific moving average appliВэcations which day traders may use, frequently with considerable sucВэcess, provided they follow certain very basic rules of application. Although some of the methods are traditional, you will find some new and interesting ideas in what follows.