Tools for
Confirmation
THE MOVING AVERAGE—THE STILL CENTER
Many trading books will tell you to use a specific moving average like a 21-period exponential or a 50-period simple moving average. Anytime anyone tells you to use a certain
average length, the logical question is: “Why that one and not another?” There is no magic
to any moving average—they are all pretty much created equal, which is to say that all
work sometimes and do not work other times. If you are viewing moving averages as support or resistance, or price crossing a moving average as a significant signal, you might
want to reconsider your approach to these tools.
There is great value in reducing everything to simple elements, to first principles, and then building the analytical structure
up from that foundation. To that end, Appendix B takes a deep look at the differences
between two commonly used moving average types and examines how they behave in
different market environments. Though this understanding might not be useful for all
traders, mastery rests on a perfect grasp of basic elements.
Ideas for Using Moving Averages
Traders have many ideas and approaches to using moving averages, but many of the
things that are supposed to happen around moving averages are based on market lore and do not pass simple statistical tests. People will claim that certain averages are important because “hedge funds buy there,” “everyone watches those averages,” or “it becomes
a self-fulfilling prophecy,” but it is extremely difficult to support these claims with quantitative tests. Tests of statistical tendencies around moving averages fail to find that any average is objectively better than another, or in fact, that there is any verifiable edge around
these averages at all. Regardless, there are practical ways to use these tools in trading.
As a Trend Indicator One potentially useful application of moving averages is to
use them as confirming trend indicators. For instance, when scanning many charts, a
quick glance at a moving average—which summarizes the price structure, smoothes out
the bumps, and highlights the overall trend—can give a quick read on the overall market
structure.
There are two broad ways to use moving averages as trend indicators, which can be
used separately or combined. The first is very simple: when price spends a lot of time on
one side of a moving average, this suggests the market is trending in that direction; see
Figure 7.1 for an example. Note that price may occasionally dip to the average, or even
cross a bit to the other side, but it remains mostly above the average.
The second idea, which is closely related, is that a moving average will slope in the
direction of the trend. A flat moving average is more indicative of a trading range, while
up-sloping and down-sloping averages suggest uptrends and downtrends. As Figure 7.2
shows, this tool is very dependent on the length of the moving average chosen; in this
diagram, the three averages each suggest different trends at the same spot. This is not a
problem, for the market can actually be in an uptrend, downtrend, and trading range at
the same time—it depends on what time frame is being considered—but the trader must
understand these issues. Again, it is not necessary to use this as a mechanical tool—for
instance, by executing only at the precise points where the line shifts slopes. A trader
making decisions based off price structure will almost always find entries before they
are confirmed by a moving average.
Avoid Markets in Equilibrium Markets are usually in equilibrium, and at those
times markets are efficient in the academic sense—random walks prevail. There is no
consistent edge possible in such an environment; it is not an exaggeration to say that the
essence of technical analysis is to identify markets that have a temporary imbalance of
buying and selling pressure, and to limit our trading to those environments.
One way
to identify markets in equilibrium is that they tend to stay close to an average price
(moving average), which represents a rough area of consensus. If the market continues to
trade around that average, chopping back and forth on both sides of it, it is probably best
to move on and to look for better trading opportunities. This might seem like common
sense, but a rule like this can keep traders from making multiple attempts at trading a
flat market.
To be sure, nothing works all the time. Even random walk markets will show significant departures from moving averages; a significant move away from a moving average is
not, in itself, sufficient evidence to declare that a tradable buying and selling imbalance
exists. However, the absence of that condition virtually guarantees that an imbalance
does not exist, and markets that remain close to intermediate-term moving averages
almost always present challenging trading environments with no significant edge. Figure 7.3 shows a period where Consolidated Edison, Inc. (NYSE: ED) traded back and
forth around a moving average with no clear trades to either direction. For most traders,
simply avoiding this type of price action can add to the bottom line, as many small losing
trades will be eliminated.
As a Reference for Trading Pullbacks Expecting a moving average to truly provide support and resistance is probably misguided, but there is a valid way to use moving
averages as a crutch for trading pullbacks. Markets typically move by an alternation of
momentum moves and consolidations, and it is usually a bad idea to initiate a trade when
the market is overextended either to the upside or the downside in a momentum move.
In almost all cases, it is better to wait for the market to work off this condition and to
return to a short-term state of balance before entering in the pullback. It might seem to
be a joke, but using a moving average like this “keeps you from doing something stupid”
(i.e., buying or shorting an overextended market). A simple but effective trading rule
could be instituted that simply prevents you from buying or selling pullbacks that are far
away from a moving average, assuming that far away has been precisely defined. It is
also worth considering that a moving average is a good reference in a normal trend; very
strong trends may not pull back as much. In general, it makes sense to build a trading
plan that first addresses the most common situations that will be encountered, dealing
with more extraordinary trending environments later. Figure 7.4 shows an example of
some possible pullback entries at a moving average.
Trade management and context are
important, as there is no inherent edge to simply executing a trade at the moving average.
Slope of a Moving Average as a Trend Indicator The slope of a moving average
is another subjective application that does not test out well in quantitative testing, but
it can be useful as a scanning tool or as a cue for the developing discretionary trader.
The concept is that the moving average’s slope can give confirmation of trend direction,
with the time frame of the trend roughly corresponding to the time frame of the moving
average.
Longer-term (100-period or more) moving averages will address longer-term
trends, while very short-term (less than 10-period) moving averages will give indications
for the shortest intervals on the chart. An important point is that an attentive trader will
almost always be able to identify inflection points based on price structure long before
the moving average changes slope, as in Figure 7.5. It is probably a bad idea to filter trades
based on this criterion—for instance, taking long trades only when the average slopes
up—because the lag effect will cause you to miss many good trades at the beginning, but
you will be green-lighted to take all the losers at the end of a trend. Focus on learning to
read the price structure, but the slope of a moving average can be a useful aid, especially
when scanning many charts rapidly.
