Confirmation Destination

Not only this, but many of the methods and strategies for analyzing the financial markets that were popularized decades ago, methods such as Elliott wave and Dow theory, have had many of their main tenets assimilated into the technical consensus. How many traders who scoff at Elliott wave analysis nevertheless look to exploit A-B-C corrections as the distance between point B and an undetermined point C begins to match the distance between point A and B? How many traders who consider themselves Dow theory agnostic still find themselves pointing giddily at a chart of the Dow transports when that chart is revealing a bullish reversal not yet mirrored in the Dow industrials (Figure 1)?

FIGURE 1: DOW JONES INDUSTRIAL AVERAGE, DAILY. Secondary corrections in the primary bull market in the industrials are marked with a support line. The high points of these corrections, the resistance beyond which the market must move in order for the correction to be deemed over, are marked with resistance lines.

There's no need to become an expert in every analytical or interpretive style when it comes to technical trading and analysis. But if you can find the elements in different strategies — the way trends are defined here, the way bull and bear markets are defined there — and incorporate them into your core trading and analytic approach, you can often find techniques that compliment and confirm your own work. For me, when it comes to taking what is most useful from Dow theory, it is without question the theory of confirmation and the relationship between primary trends and secondary corrections and reactions.

In Dow theory, the idea of confirmation between the Dow industrials and Dow transports can be summed up in the phrase: Yes/Yes and No/No. When both averages are doing the same thing — whether that is advancing or declining — they are said to confirm each other. Confirmation, as Robert Edwards and John Magee wrote decades ago in their classic book, Technical Analysis Of Stock Trends, is a positive signal — meaning that it represents information that can be acted on directly. Actionable market intelligence, if you will.

This is in contrast to nonconfirmation, which occurs when one average is advancing and the other is declining. Nonconfirmation, it is important to note, does not produce a positive signal. The intelligence provided when the two averages are not mirroring each other or confirming each other is not actionable and does not provide specific trading information.

Edwards and Magee warned traders that "no valid signal of a change in trend can be produced by the action of one average alone." Too often, non–Dow theory traders note when one average is rising and the other falling and assume that this divergence between the two averages is bearish — or bullish, for that matter. This is the wrong way to think about nonconfirmations between the industrials and the transports.

That is not to say that a nonconfirmation or a divergence between the transports and the industrials is a cause for celebration; it does indeed represent a reason for caution. But too many traders and analysts misread that caution. A nonconfirmation or divergence between the two averages is a cause for concern insofar as (1) the divergence will be resolved sooner or later in the form of a confirmation and (2) the divergence itself does not necessarily reveal whether the confirmation will result in resumption or a change of trend.

These are the reasons why traders should be cautious when nonconfirmations between the averages appear, not because nonconfirmations are inherently bearish.

PRIMARIES AND SECONDARIES
The other Dow theory concept that non–Dow theorists can easily adapt and include in their own analytical methods is the notion of primary and secondary trends. While larger concepts such as secular and cyclical bull and bear markets are often helpful "big picture" frameworks to understand markets historically, they lack the specificity to be truly useful for most intermediate-term traders and investors. To this end, the idea of primary markets and secondary trends can be a superior substitute.

What is a primary market? "Trader Vic" Sperandeo described primary markets this way in his book Trader Vic: Methods Of A Wall Street Master:

The primary movement is the broad basic trend generally known as a bull or bear market extending over periods which have varied from less than a year to several years. The correct determination of the direction of this movement is the most important factor in successful speculation. There is no known method of forecasting the extent or duration of a primary movement.

For the purpose of this discussion, a secondary reaction is considered to be an important decline in a bull market or advance in a bear market, usually lasting from three weeks to as many months, during which intervals the price movement generally retraces from 33% to 66% of the primary price change since the termination of the last preceding secondary reaction.

I like to broaden this notion of secondary corrections and reaction to include sideways movement as well as exclusively the sharp 33% to 66% corrections Sperandeo mentions. The market experience, post-2003, also has me allowing for corrections half as severe as those he notes.

What is important for me is less the mathematics of the correction and more the way the market reacts to the correction. Markets are generally considered to have recovered from their secondary corrections and reactions when they rally above the high point of the correction (bullish case) or decline below the low point of the reaction (bearish case). But when do we know whether a correction or reaction is merely that — and consequently an excellent opportunity to place a bet in the direction of the primary trend — or is instead the beginning of a new primary trend in the opposite direction?

FIGURE 2: DOW JONES TRANSPORTATION AVERAGE, DAILY. The most recent secondary correction in the Dow transports is noted here with a support line at the correction low and a resistance line at the correction high. The failure of the transports to hold above the correction high is indicative of the weakness the average has experienced of late.

This is often where a trader's core set of analytic and trading tools can be especially helpful. Because Dow theory is silent between the moment of nonconfirmation and confirmation, there is an analytic space that can be best assayed by other techniques and methods to determine the likelihood of a top in the positively diverging market or a bottom in the negatively diverging one.

A FAILURE TO CONFIRM
With all this in mind, let's look at the recent action in the Dow industrials and the Dow transports leading up to and since the August 2007 correction.

Both the industrials and transports created bullish positive divergences at the end of their recent July-August corrections. But the similarities may end there. While the relatively milder decline in the industrials was met with a rally that tested the all-time highs near 14,000, the bounce in the transports has been decidedly lackluster.

In fact, on a closing basis, the transports have made their first higher high only a few days ago in contrast to the industrials, which accomplished the same feat back in August (Figure 3). What is all the more daunting for the transports is the fact that they remain pinned below their 50-day exponential moving average — again compared to the industrials, which broke out above their 50-day exponential moving average (EMA) back on September 13.

FIGURE 3: DOW JONES INDUSTRIAL AVERAGE, DAILY. The positive divergence in the MACD histogram in August was a clear sign that momentum to the downside had waned. The resulting bounce has brought the industrials within range of the all-time highs just north of 14,000.

The bullish case for both averages is buttressed on the sizable positive divergence in the MACD histogram. Positive divergences like those after significant declines must be respected. And even if they do not result in strongly bullish price action, they often accurately signal the end of significant momentum to the downside.

Which type of confirmation is more likely at this point between the averages, a bullish one or a bearish one? In some ways it is a difficult call because from a technical perspective the industrials appear as likely to set new all-time highs as the transports are to set new year-to-date lows.

As crazy as it may sound, given the fact that the transports are flirting with year-to-date lows (and crude oil is flirting with record highs), it appears far more likely to me that the transports will confirm the advance of the industrials than the industrials confirming the recent weakness in the transports. (See Figure 4.)

FIGURE 4: DOW JONES TRANSPORTATION AVERAGE, DAILY. The MACD histogram also shows a positive divergence in the transports. However, this divergence has not led to the sort of strong upward movement that the divergence in the industrials did. The highlighted area represents a zone of resistance the transports have yet to overcome.

The industrials, it goes without saying, are nowhere near their correction lows from the summer of 2006. And in the same way that the transports outperformed the industrials from late 2005 into summer 2006, it may be the industrials that will now outperform the transports as the markets recover from their most recent correction.

SUGGESTED READING
Edwards, Robert D., and John Magee [2001]. Technical Analysis Of Stock Trends, 8th ed., W.H.C. Bassetti, ed. Saint Lucie Press.
Sperandeo, Victor [1991]. Trader Vic: Methods Of A Wall Street Master, John Wiley & Sons.