Trendlines And Consolidations

When an equity is traded it is either in an uptrend, a downtrend, or it is moving sideways. There are two types of sideways movements. One is what might be called an accumulation phase, which is a series of daily price changes that vary less than 5%. If the price was $50, then price would vary no more than $2.50 for several weeks.

The other type of sideways price movement, a consolidation, is one where price moves up and down by 5% or more in a channel. Keep in mind that patterns you want to trade should have high average daily volume. Some would like to see at least 500,000 shares per day, but I prefer the number to be closer to one million.

I have found the key to defining trendlines is to use pivots; doing so makes consolidations more apparent. It also allows you to practice a useful analysis technique called the Andrews pitchfork*. Pitchforks deserve their own discussion, but let's focus for now on drawing trendlines using pivots, and the benefits of doing so.

A DOWNTRENDING EQUITY

Figure 1 is a simple illustration of an equity whose price is in a downtrend. Each of the peaks and valleys is a pivot. To define a pivot, look for two lower trading bars before and after a pivot peak, and similarly, higher bars on each side of a valley pivot. (See my Working Money article "What Are Pivots?")

Figure 1: Idealized downtrend with consolidation. Each peak and valley, A through E, is a pivot, with a clear downtrend evident from point A to F and a sideways move from B to E.

You can put more risk in your analysis by having just one lower bar before and after a peak, or you can be more conservative and require three bars (a valley is similar, with either one or three bars being higher than the pivot valley). Let's apply this method to the Nasdaq Composite, a popular index that is analyzed extensively (see Figure 2).

Figure 2: Daily Nasdaq Composite prices. The red trendline satisfies the condition that a straight line connects two or more peaks, but it misses the consolidation in October. The blue lines are an alternative set of trendlines connecting several pivots.

The natural tendency is to just start connecting peaks with a straight line (red line). But first you should identify the pivots. Before that, you must decide what constitutes a pivot in terms of meaningful price action. Again, I prefer at least two bars before and after a high or low to define a peak or valley. Using this two-bar rule, we can identify the pivot point labeled B in Figure 3.

By connecting points A and B (blue trendline) and extending the line to October, it becomes apparent that the downtrend was broken in mid-October. This leads you to wonder what direction price will take next. The ensuing price action is a sideways movement defined by pivots X, Y, Z, and C (Figure 3).

Figure 3: Nasdaq daily price. The price bars leading up to and following pivot B have been identified in the blue square. Pivots Y and Z share bars, since the two bars lower than pivot peak Y are the same two bars higher than valley Z.

Before continuing, let's consider another way pivots can be significant. A pivot is a point where a price turnaround occurs, and by definition some traders profit and some don't (especially if they don't use stops to take themselves out of a position). The market has yet to return to these levels (Figure 3; prices at pivots X, Y, Z, and C), and it will probably "remember" them, because those who bought on the dips at pivots X and Z have seen prices go on down. When prices return to these levels now, those who were caught before are likely to want to sell. The longer it takes to return to these levels, the more likely it is that market participants have given up in disgust and taken their losses. The volume matters in this analysis as well, since the volume at any given pivot gives you a measure of how many traders were caught long or short, and thus, how strong the market's "memory" (resistance or support) will be at this level.

By late November it is clear that prices have finished consolidating. Your next question might be, "What is the likely pivot point for the next major move?" This is where consolidations become significant.

CONSOLIDATIONS

Consolidations often occur halfway through major moves. This allows you to estimate the duration of a trend by finding the distance between the start of the move and the consolidation. On September 1, 2000 (pivot A), the Nasdaq Composite closed at 4234. The midpoint of the consolidation is at 3278, so the continuing downtrend should see the pivot for the next major move at 3278-(4234-3278)=2322. The chart confirms this estimation the next move is an upswing at around 2322 in early January 2001 (Figure 4).

Figure 4: Nasdaq daily price.

Using a definition of two bars before and after a bar peak or valley allows you to define additional pivots at X, Y, and Z. Since a consolidation marks the halfway point of a move, then the prediction would be that the Nasdaq should see a turning point at about 2322, which is what happened.

Again, the rule is that consolidations mark the halfway point in major moves. A major move could consist of trends interrupted by several consolidations. The longer and more volatile a consolidation, the more likely it will result in a strong move to continue the trend.

Consolidations can also occur at tops and bottoms, but for this method you need to look for prices that are breaking out of the consolidation pattern in the same direction they entered. That is why you would wait until late November to determine that you haven't seen a bottom but a consolidation. The key is the bar circled in red (Figure 4). Two bars prior to that is a pivot peak, which tells you the market has decided where it's going to go — that is, down. You would place an order to short Qqq on November 17, 2000, after seeing the high of that day was not going to be any higher than the midpoint of the previous day. This gives you an idea of how far to ride this trend.

Consolidations also exhibit some other behaviors that are useful. Figure 5 identifies some trendlines, consolidation lines, and one symmetrical triangle.

Figure 5: Daily price chart of Home Depot (HD). When you analyze a chart you should look for both trends and consolidations. The market considers trends in three ways — first, a gradual change is tolerated and not likely to have consolidations because they are not needed; second, faster changes will be challenged with consolidations to test the price rate change; and third, extreme rates of changes are going to end up as either intermediate or major tops or bottoms.

Stocks spend most of their time going sideways, versus being in uptrends or downtrends. From early 1997 to early 1998, Home Depot (HD) is in a long gradual uptrend. The rest of the chart, starting with the first trendline break, is more typical price action, with steeply sloped trendlines and consolidations. The steeper the slope of a trend, the more unsustainable it is, whether it's moving up or down.

Let's analyze this chart, going from left to right:

1 By middle or late May 1998, there is a confirmed trend break, as prices rise steeper. This faster price rate should alert you that the market is looking for a place to retrace, since it is not comfortable with changing its slope.

2 By late November to early December you can see that prices have gone through a consolidation; you would estimate that 30+(30-24)=36 (the upper limit on the consolidation plus the width of the consolidation) is going to be a price at which the next decision is going to be made. 3 In going from the first consolidation to the second one, the uptrend, 24 to 40, is even steeper than the previous uptrend, and by early October 1999, you can see you have gone through a consolidation between 36 and 44. Now you make a rough calculation where price is going next by calculating 40+(40-24)=56 (the midpoint of the second consolidation plus the price movement from the trend break to the midpoint of the second consolidation).

4 At 56 price makes a move. This is an uptrend with an unsustainable slope, going from 56 to 69 in approximately two weeks of trading (December 10, 1999, to December 27, 1999).

5 Using the same type of calculation as above shows that 44+(44-36)=52 is a price that forms a degree of support.

As an aside in this discussion, let's look at the symmetrical triangle on the chart in the second half of 2000. Is there any predictive value in this pattern? This triangle pattern calculation is worth mentioning because it is similar to the rules for consolidations.

The rule for triangles is to use the length of their base as the amount of movement predicted when a stock breaks out of the triangle pattern. In this case the lower end of the base is at 47 (May 22, 2000), while the upper end is an extrapolation of the upper trendline to May 22 with HD at about 64. The base is therefore 64-47=17, and the axis of the triangle about 52, so the move down from the triangle pattern should end at 35 (52-17), which it does.

NOW ON THE NASDAQ

What would you learn if you were to apply the same consolidation pattern analysis technique to the Nasdaq Composite?

By mid-March 2001, the chart of the Nasdaq Composite displayed in Figure 6 exhibits three trends — the last two are roughly parallel, and the first one has a slope that's a bit steeper than the other two. In fact, the steepness of the first trendline (A) only portends a rebound at some unspecified, future time, and of course it comes initially at pivot X in mid-October 2000 (Figure 4). By mid-March 2001 it is possible to draw a third trendline. When you see three roughly parallel trendlines, each displaced to the right, you know that the trend will be displaced because price has moved sideways. So what is this big "consolidation" area?

Figure 6: Applying consolidation pattern analysis to the Nasdaq. When you see consecutive parallel trend lines shifted to the right, you have to assume price has gone sideways.

The pivot peak of 3482 and a pivot valley of 2297 define this area, since they are the pivots defining the extremes. You may think this doesn't look like price moving sideways, but it is. It moves sideways, then down, and then sideways again. What you should see is a downswing in prices leading to the "consolidation" and a downswing leaving it. This variation works often enough to warrant taking the time to construct it. My midpoint is 2890, (3482+2297)/2, and my estimate of a turning point is 2890-1185=1705, where 1185 is the width of the "consolidation." After hitting 1638 for a low, the next day the Nasdaq Composite opens at 1706.

And how is 2297 holding up as resistance? The highest the Nasdaq composite has reached is a close at 2313 on May 22, 2001. So far it is resisting strongly — it seems the market is remembering. The close of 2313 is followed by a close of 2243 on May 23.

Dennis Peterson may be reached at DPeterson@Traders.com.

SIDEBAR

ANDREWS PITCHFORK

Curious about the Andrews pitchfork? Refer to the daily chart of the Nasdaq Composite displayed in sidebar Figure 1.

Sidebar Figure 1: Applying the Andrews pitchfork. Daily price is overlaid with a zigzag using a 6% change. An Andrews pitchfork is drawn using the bottom at 4/4/01 and two pivots at 4/20/01 and 4/26/01. The two dashed magenta lines form the envelope for prices to move within.

Take a pivot high or low and find the next two pivots (B and C). Next, draw a line from the pivot high or low (A) through the midpoint of the other two pivots (solid purple line). Construct two lines parallel to the line going through the midpoint, but place these two lines at each of the two pivots to form a channel (dashed purple lines going through pivots B and C). You now expect price to move within this channel.

One of the features of this device is that you can foresee a problem. After constructing the pitchfork, price would have to follow the red line to hit the upper channel, which is not even remotely likely. The price does something more rational and moves outside the envelope, which reveals the most likely case for price movement. By looking at the pitchfork, you can make a good estimate of what's likely to happen next. This is one of the reasons that chart technicians are wary of fast-climbing stocks. — DDP

SUGGESTED READING

Peterson, Dennis [2002]. "What Are Pivots?" Working Money, January.

Star, Barbara [1995]. " Support And Resistance With The Andrews Pitchfork," Technical Analysis of Stocks & Commodities, Volume 13: November. MetaStock (Equis International)

*See our online cumulative glossary

Current and past articles from Working Money, The Investors' Magazine, can be found at Working-Money.com.