The relative strength concept can also be expanded to the commodity area. This is achieved by comparing the price of an individual commodity, such as corn, to a commodity index, such as the Thompson Reuters Commodity Research Board (CRB) Composite or the Dow Jones UBS Commodity Index. Our examples will focus on individual stocks relative to the market, but this and other concepts are equally acceptable. In this respect, Figure 19.1 shows the closing price of the stock in the upper panel and its RS in the lower one.
FIGURE 19.1 RS and Price
When the line is rising, it means that the stock is outperforming the market. In this case, the denominator is the S&P Composite, so a rising line means that the stock is outperforming the S&P. Later on, it continues to rally, but the RS line peaks out. This means that the stock is now underperforming the market. Another possibility might involve the comparison of an individual country’s stock or index to a global indicator, such as the Morgan Stanley World Stock Index. As long as the appropriate currency adjustments are made, the concepts are the same.
The interpretation of relative trends is subject to exactly the same principles as the price itself. It is important to note that an RS indicator is just what the name implies: relative. A rising line does not mean that an item, such as a stock, is advancing in price—merely that it is outperforming the market, or rising relative to the market average. For example, the market as measured by the S&P Composite may have fallen by 20 percent and the stock by 10 percent. Both have lost value, but the RS line would be rising because the stock retreated less than the market.
Major Technical Principle Relative strength moves in trends, just like the absolute price.
RS Interpretation
Because the RS line moves in trends, it lends itself to trend-reversal techniques such as price patterns, trendlines, and moving-average (MA) crossovers. Momentum indicators may also be calculated from relative relationships.
Relative trends can be interpreted in similar ways to trends in absolute prices. However, the introduction of relative analysis into the equation presents an additional dynamic. This arises from a comparison of the two series that often throws up subtle differences in much the same way as a comparison between the price and an oscillator.
Since RS trends tend to experience more random noise than absolute price trends, we generally find that charts based on weekly and monthly data tend to be more reliable than those constructed from daily RS data. This same principle is true for the absolute price, but more so, I believe, for relative action.
Positive and Negative RS Divergences
When both the price and the RS are rising, they are said to be “in gear.” Important trends usually begin with both series acting in concert, but eventually, the RS line fails to confirm new highs being set by the price itself. This type of situation indicates that the odds favor the stock beginning a period of underperformance against the market. However, weakness in RS is not an absolute sell signal, i.e., one indicating that the price itself will go down; it is merely a relative signal, i.e., one implying a switch from an issue that has started to become out of favor to one that is coming into favor.
Quite often, though, a divergence or series of divergences between the price and RS represents an early warning sign of trouble, which is later confirmed by a trend-reversal signal in the price itself. In Figure 19.2 the two are in gear at the start, but later on, the RS line diverges negatively with the price on three occasions. Finally, the price itself completes a top and declines.
FIGURE 19.2 RS and a Negative Divergence
The opposite set of circumstances holds true in a declining market, in which an improvement in RS ahead of price is regarded as a positive sign. An example is shown in Figure 19.3. This time, the confirmation comes from a trendline break in the price.
FIGURE 19.3 RS and a Positive Divergence
Trend-Reversal Techniques
1. Moving-Average Crossovers Sometimes it’s a good idea to run a moving average through the price using the crossovers as legitimate signals of a change in trend. It’s also possible to do the same thing for a relative strength line, but because the RS line tends to be much more volatile, this technique often proves unprofitable because of the numerous whipsaws that are generated. This is especially true for short-term trends, but even long-term moving averages, such as a 40-week simple or 65-week exponential, often result in more whipsaw signals than we might like. Figure 19.4 offers an alternative. It involves plotting two moving averages, a short-term and a longer-term average, using the crossovers for signaling trend reversals. This approach definitely eliminates whipsaws, but the trade-off is that several signals are less timely.
FIGURE 19.4 RS and MAs
An example is shown in Chart 19.1 for General Motors. Note the numerous whipsaws for the 65-week exponential moving average (EMA) being flagged by the ellipse.
CHART 19.1 General Motors, 1993–2001 and RS MA Crossovers
Chart 19.2 shows an alternative where the 65-week EMA is used with its 10-week EMA as a method of signaling trend changes. Note that nearly all of the 1996–1998 whipsaws have been eliminated. We are still left with a couple, but the 1996–1998 pounding is totally avoided as the 65-week red line remains below the solid blue line during the whole period. It is a good idea when you have settled on a time span that makes sense to take a look back historically to make sure it has worked consistently in the past. This should include as much historical data as possible so that its performance during a variety of market conditions can be assessed. For example, a chart plotted with weekly data should ideally cover at least 8 years (two complete cycles) of data.
CHART 19.2 General Motors, 1995–2000, and RS MA Crossovers
2. Trendline Violations In my view, a better alternative to the moving-average approach is to construct trendlines against the relative strength line. The concept is to construct a trendline for the RS line and when that is violated to look around for a legitimate trend-reversal signal in the price itself. In that way we can see that the price is responding to the relative strength or weakness. Figure 19.5 shows an example of a reversal from an uptrend to a downtrend.
FIGURE 19.5 RS and Up Trendlines
Figure 19.6 shows that this is also a useful way to identify “buy” candidates. The first thing to do is wait for a violation of the RS line.
FIGURE 19.6 RS and Down Trendlines
Then, when the price also confirms with a trend-reversal signal, you can take some action. These joint violations don’t occur that often, but when they do, it’s usually a signal of an important reversal. In this case, the strength of the signal has been enhanced because it was preceded by a positive divergence. The divergence did not represent a signal to buy, but it did set the scene for some positive action later on by indicating that the technical position is improving. Incidentally, the confirmation by the absolute price does not have to be a trendline break—it could be a price pattern completion, a reliable, (and I emphasize the word “reliable”) moving-average crossover, or even a reversal to a series of rising peaks and troughs. Remember at all times that the size of the new trend will depend principally on the time frame being charted and the length of the lines. Intraday breaks, for example, reflect small trends and will have nowhere near the significance of trendline violations on the monthly charts.
3. Price Patterns Price patterns can also be employed to analyze trends in relative strength. In Figure 19.7 the RS line completed a head-and-shoulders top.
FIGURE 19.7 RS and Price Patterns
This certainly indicates that the RS trend has reversed and provides enough evidence to justify a switch from this stock in favor of one where the RS trend was emerging in a positive way. However, it does not signal that the price itself is going to decline, though in many instances, that will prove to be the case. In this particular example, the absolute trend reversal is signaled when the short-term low flagged by the small horizontal trendline is violated, because this confirms that the series of rising bottoms and tops h
as now been reversed. Note that even though the price subsequently rallies back through the line, this does nothing to reverse the peak/trough progression to the upside, so the trend is still regarded as negative.
Figure 19.8 features a reversal from a downtrend to an uptrend. First, the RS line diverges positively with the price. This is our initial indication that both trends may be about to reverse. Then the RS line traces out a rectangle and breaks to the upside, later to be followed by the price completing a broadening formation with a flatish top.
FIGURE 19.8 RS and Price Patterns
Divergences and Trend Signals Sometimes a study of the relationship between the price and RS line can be quite revealing. Chart 19.3, for example, featuring Stanley Works, shows the price reaching a new high in April 2006 but the RS line fails to confirm. This provided some advance warning that the technical structure was not as positive under the hood. When both trendlines were violated later on, this confirmed the weakness and lower prices ensued. Later on, the price began to rally and everything was in gear until June 2007. This was the warning that when both trendlines were penetrated that price weakness would continue.
CHART 19.3 Stanley Works, 2005–2007 RS, Divergences and Price Confirmation
Chart 19.4 shows another RS example. This time, it was the Turkish Akbank versus the Turkish Index. At the start of 2005, the divergence between the RS and absolute price was quite serious, as the RS line was falling quickly. It violated its trendline early on, but when the price followed suit, a very nasty decline followed. Quite often, when we see a discrepancy as large as this, the weakening RS line is offering a loud warning that despite what we see on the surface (rising absolute prices), the technical picture is really quite weak. The price is being dragged up by the market and when it finally turns, the deteriorating relative strength chickens come home to roost. Later on, we see a positive RS divergence as the RS line bottomed in early March, but the subsequent late-March higher low was not confirmed by the price. Later, both series violated down trendlines and a rally was confirmed.
CHART 19.4 Akbank, 2004–2005 RS, Divergences and Price Confirmation
Long-Term Relative Strength
Chart 19.5 features a monthly close of the S&P Domestic Oil relative strength line. This is a very long-term chart that encompasses much of the twentieth century. It is useful in that it demonstrates that the RS line lends itself to price pattern and trendline construction. These formations are not completed very often, but they are usually followed by a relative price move that lasts for many years. It’s important to bear this in mind because most of the patterns look small on the chart but actually extend over considerable periods of time. Their completion, therefore, signals a change in the environment that typically lasts for many years, even decades. For example, it traces out a 12-year reverse head-and-shoulders top in the 1970s. A break of this magnitude signals a change in sentiment for a very long time. Indeed, a huge rally took the RS line to its secular peak in 1980.
CHART 19.5 S&P Domestic Oil Index, 1960–2012 Long-Term RS Trend Analysis
Note the fact that the Oil Index itself experienced a substantial rally between 1980 and the end of the century. Does this mean that oils would have been a good place to invest in this 20-year period? Hardly, because the persistent decline in the RS line indicated that they were consistent underperformers. Only in the next decade did the relative action come into its own as it broke out from a triangle base, resulting in a rally by both series. In 2008, this strong relative rally was halted as the 2003–2008 up trendline was violated.
Obviously, we would not make a practice of studying these long-term charts every week or so, but say, once a quarter, it does make sense to review the bigger picture to see whether any major trends might be emerging.
Major Technical Principle It is the longer-term trends that dominate the characteristics of their shorter-term brethren.
On a more regular basis, we would look at monthly charts covering shorter periods of time, later moving down to the weekly and daily ones.
Individual Stocks and Relative Strength Analysis
Chart 19.6 features a daily chart of Hewlett-Packard. It starts off in early 2005 with the simultaneous completion of two inverse head-and-shoulder patterns.
CHART 19.6 Hewlett-Packard, 2003–2012 Long-Term RS Analysis
In 2007 and some of 2008, the absolute price traces out a head-and-shoulders top, and at the breakdown point at A, the RS line is close to a high. The actual price starts to fall like a stone but the RS line continues its upward march. This shows that Hewlett-Packard was a relatively safe place to be. Its recovery bounce into 2010 continued to indicate that the price had weathered the storm and was likely headed higher. However, both series violated trendlines at B. The RS trendline was far more significant because it was much longer—6 years—and had been touched or approached on eight occasions. This indeed was a serious break, and anyone who had confidence in the stock because it had weathered the previous bear market quite well should have paid attention to this very serious break. In the following three years, the bottom literally fell out of the market. Even during the two small rallies at C and D, the relative performance continued to deteriorate.
Major Technical Principle It’s always better to avoid a stock when it tries to rally during a downtrend but the RS line does not.
On the other hand, an improving RS line is often an early clue that a turn may be at hand. In such cases, the clincher is the eventual confirmation by the price.
Relative Strength and Momentum
Long-Term Trends
Since classic trend-determining techniques can be applied to RS lines, it is a small step to expand the analysis to embrace momentum indicators derived from relative strength lines. While it is certainly practical to apply oscillators to short-term momentum derived from RS lines, by far, the best use of momentum in relative work, I believe, is to use oscillators based on long-term spans, especially if they have been carefully smoothed to limit unwanted fluctuations.
Chart 19.7 features the relative strength line of the Dow Jones Oil and Gas Sector ETF together with a long-term Know Sure Thing (KST) of relative strength. The waves against the RS line reflect those in the KST. Peaks in the KST roughly correspond with its peaks and troughs. As we shall learn in a later chapter on group rotation (see Chapter 22), RS lines are far more cyclical in their patterns than absolute prices. This makes the use of smoothed long-term oscillators such as the KST far more accurate in their reflection of the primary trend. Remember, absolute prices can be subject to strong linear trends, which means that even the best-designed smoothed long-term momentum indicator will offer premature buy and sell signals. We cannot say this will never happen with a long-term momentum indicator constructed from relative action, but it is certainly a lot less likely.
CHART 19.7 Dow Jones Oil and Gas RS, 2005–2012 and Long-Term RS Momentum
The principal objective is to identify a security when its long-term KST is below zero and starting to poke above its 26-week EMA. Incidentally, the average plotted against the RS line itself is a 65-week EMA. Note that even with a substantial time span such as this there were still numerous whipsaws, especially after the late-2008 bottom. This is one of the reasons why I prefer to use trendline violations of the RS line in conjunction with long-term RS KST reversals. They are certainly not perfect, but tend to be relatively more reliable, as demonstrated by three of them in this chart. This example used a KST, but it is possible to substitute any smoothed long-term momentum indicator. The KST just happens to be my preference. Alternatives could be a stochastic, a moving-average convergence divergence (MACD), or other trend deviation indicator. The basic idea is to use one that closely resembles the primary up and down waves yet turns reasonably close to the turning points. When experimenting, always try for consistency over a number of securities in different time periods; never aim for perfection because it just doesn’t exist.
Once the direction and maturity of the long-term tren
d of the RS has been established, it is then time to move to the shorter-term charts.
Short-Term Trends
Chart 19.8 features the relative action of the Spider Metal and Mining ETF (symbol XME) together with two rates of change (ROCs). The 10-day series registered a post-1989 (limit of my database) high. The 45-day ROC also traced out an extreme swing and broke out from a small base a little later on. The price eventually confirmed with the completion of a small-platform, double-bottom formation. In a strict sense, this was not a mega overbought because the price declined for only 6 months and did not reach my minimum benchmark of 9 months. However, the speed of the decline and the fact that it was a post-1989 overbought condition that was confirmed by the price qualifies it as one in my poetic license book.
CHART 19.8 Spider Metal and Mining ETF RS, 2004–2010 RS and Momentum Interpretation
Chart 19.9 features the relative action of Abbott Labs in the top panel, followed by a 14-day RSI of the RS line, and in the lower panel an MACD of the RS line.
CHART 19.9 Abbott Labs RS, 1998–2001 and Short-Term RS Momentum
It is fairly evident by looking at the chart that there are two main environments: a bear market between the end of 1999 and the start of 2000. This is then followed by a bull move. Now take a closer look at the MACD. During the bear market, it fails to reach an overbought condition, yet oversold readings fail to signal rallies. The opposite is true during the bull phase. This is typical of oscillators since they change their characteristics in primary bull markets. Just like birds in the Northern Hemisphere, they migrate to the south during the winter (or the bear market) and to the north in the summer (or the bull market). Whenever you can spot a situation where an oversold oscillator fails to trigger much in the way of a rally, this represents a tip that the prevailing trend may be bearish. It doesn’t happen every time, of course, but in most cases, this rule will work out. In this case, the failure of the January 1999 MACD oversold condition to generate a rally and its failure to register an overbought reading a little bit later pointed to a bear market environment.
Technical Analysis Explained Page 28