BROADENING BOTTOMS PATTERN
H
Surprises are many and a few appear in the snapshot. I discuss them in more detail later
Tour
You may be wondering what differentiates a broadening bottom from a broadening top. A broadening bottom has a price trend leading down to the start of
the formation; a broadening top has prices trending up. This differentiation is
an arbitrary designation I made to separate the two formation types. I ignore
brief dips or rises a few days before the pattern as Figure 1.1 shows.
Some maintain that a broadening bottom does not exist. They simply
lump every broadening pattern into the broadening top category. I decided to
separate the two on the chance that their behavior differs.
This particular chart pattern shows a partial decline. Prices move down
from 26 to 24.50, then reverse course and shoot out the top. The stock reached
a high of 38.50 just over a year later.
Identification Guidelines
Table 1.1 lists the identification guidelines for broadening bottoms.
Price trend. As mentioned earlier, a declining price trend precedes a
broadening bottom. Even if prices rise just before the formation begins, ignore
it. It is still a bottom. This arbitrary designation also makes intuitive sense: A
bottom should appear at the end of a downtrend, not when prices are climbing
to the moon.
Shape. The shape of the formation is distinct. It reminds me of chaos
theory where small disturbances oscillate back and forth, then grow unbounded, wreaking havoc. In the stock market, price reaches a new high then
crosses over and makes a new low, creating the broadening pattern. When you
draw a trend line across the minor highs and another connecting the minor
lows, the formation looks like a megaphone.
Trend lines. The two trend lines drawn across the minor highs and lows
are important. The top trend line should slope up; the bottom one should
slope down. The diverging trend lines distinguish the broadening bottom from
other types of formations, such as the right-angled broadening formation
(which has one horizontal trend line) or the broadening wedge (both trend
lines slope in the same direction). So it is important that each trend line has a
slope that is opposite the other.
Touches. A broadening bottom needs at least two minor highs and two
minor lows to be a valid formation. Anything fewer means you are incorrectly
identifying the formation. What is a minor high or low? A minor high is when
prices trend up, then drop back down, leaving a clearly defined peak. A minor
low is just the same thing flipped upside down: Prices move lower, then head
back up leaving a clearly defined valley. Figure 1.1 shows five minor highs or
lows, labeled by numbers. The odd numbers tag the minor highs and the even
numbers are the minor lows. Let me stress that the minor highs and lows need
not be alternating, as in Figure 1.1. Just as long as you can count at least two
peaks and two valleys—wherever they may appear—that is fine.
Volume. There is nothing magical in the volume trend. I performed linear regression from the start of each formation to the end point (not the breakout point that is usually a month beyond the end of the formation) and found
that volume rises about 57% of the time.
If you look closely at broadening bottoms, you will find that volume sometimes follows price. In Figure 1.1, the price decline between peak 1 and trough
2 shows a receding volume trend. When prices head up from point 2 to point 3,
so does volume. If you were to place a thin wooden plank between the volume
peaks in late November and late January, it would bow downward in a U shape.
A U-shaped volume pattern occurs slightly more often than other shapes.
Breakout. The breakout point is difficult to identify in a broadening formation as it develops. I look for the place where price pierces the up or down
trend line or makes an extended move. If price pierces the trend line, then the
penetration point becomes the breakout point. If prices move up and follow
along the top trend line without piercing it, then I backtrack to the prior minor
high and draw a horizontal line forward in time until prices cross it. When that
happens, that is the breakout point.
Let me give you an example. Consider the broadening bottom shown in
Figure 1.2. The price trend over the preceding month leading to the formation
is downward. The two trend lines outline a widening price pattern as you
would expect from a broadening formation. There are more than two minor
highs and two minor lows pictured, meeting another criterion mentioned
in Table 1.1.
Where is the breakout? This formation is particularly easy. If you extend
the top trend line upward, you find that prices rise well above the line, signaling an upward breakout. Then it is just a matter of backtracking to the highest
minor high and drawing a horizontal line to determine the actual breakout
price. Point A marks the highest high in the formation.
Focus on Failures
Figure 1.3 shows a broadening bottom failure. Prices head down and appear to
suffer a small dead-cat bounce lasting from April to August. I do not recommend taking a position in any stock that shows a dead-cat bounce regardless of
how attractive the formation looks. Obey this recommendation for 6 months
to a year while the stock recovers and management gets its house in order (or
solves the cause of whatever is ailing the stock).
In the 3 weeks before the formation appeared, prices were heading higher
in reaction to the dead-cat bounce. In June they moved horizontally from the
formation top for over a month before easing down. During this time, prices
rose above the high of the formation (see point A).
A breakout occurs when price closes beyond the formation high or low.
Point A is not an upward breakout because the close is at 33.88, well below the
formation high of 34.25. Two days later, price peaks above the high, but the
close is also below the formation high.
However, look what happens when prices begin sinking in mid-July.
They drop below the formation and close even lower. The price needs to drop below 30.38. At its lowest point, it closes at 29.88. That is just fifty cents below
the low, but it is enough to signal a downward breakout. Within a week of
moving below the formation low, prices shoot to 33 and continue up using a
slower trajectory.
Figure 1.3 represents what I call a 5% failure. Prices break out lower but
fail to continue moving in the breakout direction by more than 5% before
heading back up. The reverse is also true for upward 5% failures: Prices move
up by less than 5% before turning around and tumbling.
Statistics
Table 1.2 shows general statistics for the broadening bottom chart pattern.
Number of formations. I found 237 patterns in 500 stocks using data
from mid 1991 to 2004.
Reversal or continuation. You can see that more patterns act as reversals than continuations. By definition, a bottom pattern has prices entering
from the top and exiting any way it darn well pleases.
In a bull market, reversals outperform continuations, but in a bear market, the situation reverses: continuations outperform reversals.
Average rise or decline. Notice how upward breakout performance
improves in a bull market and downward breakout performance is better in a
bear market. Think of it as a rising tide that lifts all boats. This is an example
of how trading with the prevailing market trend will improve your results.
Rises or declines over 45%. Outstanding performance is rare for BBs.
The best showing comes from upward breakouts in a bull market, with 20% of
the patterns climbing more than 45%. That may sound like a lot, but other
patterns do much better. Thus, do not expect a large move. Downward breakouts almost never fare well in this category.
Change after trend ends. In a bull market, once prices reach the ultimate low after a downward breakout, prices rise an average of 51%. Even in a
bear market, the climb measures 46%. Thus, if you can determine when the
downtrend ends, buy the stock and surf the rising tide.
Busted pattern performance. Few patterns bust, so the performance
numbers are not solid. Still, they show how much better a busted pattern does
than one that works. If you see prices move less than 5% after the breakout and
then return to the pattern, consider trading the new direction, but only if it
breaks out the other side.
Standard & Poor’s 500 change. Compare the change in the index with
the average rise or decline. Large moves in the index associate with large moves
in the stock. For best performance, trade with the market trend (bull markets,
upward breakouts and bear markets, downward breakouts).
Days to ultimate high or low. It takes between a month to 4 months to
reach the ultimate high or low. The numbers say that the move in a bear market takes less time than in a bull market. Thus, the decline after a BB in a bear
market must be at a steeper slope than is the rise in a bull market.
Table 1.3 shows failure rates for BBs. The 5% break-even failure rate is
lowest in a bear market (9%). The failure rates climb quickly as the maximum
price rise or decline changes. For the 10% failure rate benchmark, BBs fail
between two and three times more often than at the 5% rate. Over half the patterns with downward breakouts fail to drop more than 15%. For upward
breakouts, patterns cross the halfway mark between rises of 15% and 25%.
These results suggest that you should not depend on a large rise or large
decline after a BB. Scanning down each row, we also see that BBs in bull markets
with upward breakouts have the lowest failure rates. Those are the ones to buy.
Another way to use Table 1.3 is to assess how likely it is that your trade
will fail. If you have a cost of trading of 5% and you want to make 20%, how
many patterns in a bull market with upward breakouts will fail to rise at least
25% (5 + 20)? Answer: 59%. Just 41% of the patterns will meet your profit
margins. Since the majority will fail, your winners must do quite well to compensate for the losses.
Table 1.4 shows statistics related to BB performance after the breakout.
Formation end to breakout. There is a delay between price touching
the trend line the final time and closing above it. This delay averages between
3 weeks and 5 weeks. It takes longer to break out in a bull market than a bear
market because the decline in a bear market is often steeper.
Yearly position. Most of the breakouts occur near the yearly low. This
makes sense for downward breakouts, especially those in bear markets. The
only exception is in bull markets. They show 44% of the breakouts occurring
in the middle of the yearly price range.
Yearly position, performance. In most entries, the sample counts are
few, making the results unreliable. Those BBs with breakouts following the
market trend (bull market, up breakout and bear market, down breakout) do
best when the breakout is near the yearly high.
Throwbacks and pullbacks. The throwback/pullback rate hovers
around 43% except for BBs in a bear market with a downward breakout. They
pullback 56% of the time. Still, the rates are too low to use in forming a reliable trading strategy.
The average time for the stock to return to the breakout price is less than
2 weeks. In most cases, when a throwback or pullback occurs, performance suffers. Look for overhead resistance to an upward move or underlying support to
a downward breakout before investing. If the congestion zone is nearby (less
than 5% away), prices will likely push through, but overhead resistance will
retard momentum and performance will suffer.
Gaps. Gaps help performance in a bear market, but the result may change
with additional samples. In a bull market, breakout day gaps either hurt performance or have no influence.
Partial rises and declines. The next four lines of Table 1.4 concern how
often a partial rise or decline occurs and if they do occur, how the pattern performs. A partial decline correctly predicts an upward breakout 80% of the time
in a bull market. The other variations are less successful, being right about
two-thirds of the time. Figure 1.1 shows a good example of a partial decline
and Figure 1.4 shows a partial rise. See the Glossary and Methodology chapter for more information.
Performance improves when the breakout duration agrees with the general market (upward breakouts, bull market or downward breakouts, bear market) and a partial rise or decline occurs. For countertrend trades, performance
suffers after a partial rise or decline.
Intraformation partial rises and declines. Intraformation partial rises
and declines appear as small loops hanging from the trend lines. They are partial rises or declines that failed to breakout: Prices made another crossing of the
broadening pattern. Intraformation partial rises or declines occur 13% or less
of the time.
Table 1.5 shows a frequency distribution of time to the ultimate high or
low. Using numbers, it tells how soon your chart pattern will likely top or bottom out. For example, in a bear market with a downward breakout, over half
(53%, or 25 + 28) bottom out within 2 weeks. At the other end of the table,
almost half the patterns (45%) in a bull market take longer than 2 months (over
70 days) to reach the ultimate high.
Notice the slight blip in the numbers around days 35 and 42. Bear markets with upward breakouts and bull markets with downward breakouts show
more trend changes during that period. Thus, 5 to 6 weeks after the breakout,
look for a trend change.
Table 1.6 shows size-related statistics. The results suffer from low sample
counts, so keep that in mind. Height. Tall patterns perform better than short ones. Before you trade a
BB, compute its height and divide the difference by the breakout price. If the
result is above the median listed in the table, you have a tall pattern. It may not
outperform, but it places the probability on your side.
Width. Wide patterns usually perform better than narrow ones. I used
the median length as the separator between narrow and wide.
Average formation length. The average width is between 1 and 2
months long.
Height and width combinations. Tall and narrow patterns outperform
the other combinations. You will want to avoid short and wide patterns. They
perform worst.
Table 1.7 shows volume-related statistics for broadening bottoms. The
sample counts are few in this table, so the conclusions you reach may change
with a larger sample size.
Volume trend. Up breakouts do better when the volume trend is falling,
and downward breakouts do better with a rising volume trend.
Volume shapes. In all cases, a random volume shape—neither U- nor
dome-shaped—performed better than the other shapes. A random shape
includes flat, rising (trending up), and falling (trending down) volume trends.
Breakout volume. Patterns in bear markets perform better after a heavy
volume breakout. The other two columns either show no difference or show
light volume breakouts outperforming.
Trading Tactics
Table 1.8 shows trading tactics for broadening bottoms.
Measure rule. The first tactic is to determine how much money you are
likely to make in a trade. The measure rule helps with the prediction. Subtract
the highest high from the lowest low in the formation to give you the formation height. Then add the value to the highest high to get the target price for
upward breakouts and subtract the height from the lowest low for downward
breakouts.
In a bull market, this method correctly predicts an upward breakout target 59% of the time. The worst showing is from BBs in a bear market with a
downward breakout. The prediction is correct just 31% of the time. I consider
values above 80% to be reliable, so this prediction method is dismal.
Figure 1.4 makes the computation clear. Point A shows the highest high
in the chart pattern at 14.13. The lowest low is point B at 12. The formation
height is the difference between the two or 2.13. Add the value to the high to
arrive at the upward price target. This turns out to be 16.26. I compute the downward target by subtracting the height from the lowest low (that is, 12 –
2.13 or 9.87). You can see in Figure 1.4 that the price never quite reaches the
downward price target.
Go long at the low. Once you have uncovered a broadening bottom,
with two minor highs and two minor lows, you can think about trading it.
When the price bounces off the lower trend line, buy the stock. Sell when
prices turn down. The downturn may occur as a partial rise partway across the
formation, or prices may cross completely to the other side, touch the top
trend line, and head down. Remember, the formation may stage an upward
breakout, so do not sell too soon and cut your profits short.
Long stop. In a rising price trend, place a stop-loss order 0.15 below the
minor low. Should the stock reverse and head down, you will be taken out with
a small loss. As the stock rises to the opposite side of the formation, move your
stop upward to 0.15 below the prior minor low. The minor low may act as a
resistance point, so you will be giving the stock every opportunity to bounce off
the resistance level before being cashed out.
Go short at the high. The trading tactic for downward breakouts is the
same. When prices touch the top trend line and begin moving down, short
the stock.
Short stop. Place a stop-loss order 0.15 above the highest high in the formation, then pray that prices decline.
Move stop. If luck is on your side and the stock heads down, move your
stop lower. Use the prior minor high—place the stop 0.15 above it.
Other. If the stock makes a partial rise or decline, consider acting on it.
This is a reliable breakout signal. Take advantage of it but make sure you place
a stop-loss order in case the trade goes bad.
Once prices break out and leave the broadening pattern, consider selling
if the price nears the target. There is no guarantee that the price will hit or
exceed the target, so be ready to complete the trade, especially if there is a
resistance level between the current price and the target. The stock may reach
the resistance point and turn around.
Sample Trade
Susan likes to think of herself as the brains in the family. While her husband is
suffering in foul weather as a carpenter, she is hammering away at her keyboard,
a slave to her computer masters. She is an active position trader who is not
afraid to short a stock, given good profit potential and an especially weak fundamental or technical situation. It is a stressful life, but making money often is.
When she spotted the broadening bottom shown in Figure 1.4, she began
her analysis. The stock reached a high of 37.38 in early November 1991
and has been heading down ever since. Now, with the stock trading at 14, she wondered how much downside remained. She drew the two trend-line boundaries and counted the number of alternating touches (in Figure 1.4, three are
labeled as numbers and Point A is the fourth alternating touch).
Since most broadening formations tend to break out after four alternating touches and since the price was near the top of the formation heading
down, she guessed that the stock would break out downward on the next crossing. So she sold the stock short and received a fill at 13.88. It was a gamble,
sure, but one she was comfortable making. In any case, she immediately placed
a stop at 14.25, or slightly above the high at point A.
Susan was overjoyed to see the stock plummet 2 days later and race across
to the other side of the formation, touching the bottom trend line at point B.
Usually, her trades are not that easy. She decided to protect her profit and lowered the stop to the nearest minor high, shown as point C, at 13.75 or 0.15
above the high. Then she waited.
The stock bounced off the lower trend line instead of busting through as
she hoped. She decided to be patient and see what the stock did next. With her
stop-loss order in place at the break-even price, she felt protected and comfortable in letting the trade ride
The stock bounced off the 12 support level and did a partial rise before
meeting resistance and heading back down. Two days after cresting, she made
the determination that on the next touch, the stock would pierce the lower
trend line and continue down. She doubled her stake by selling more stock
short at 12.75. She was wrong. The stock continued down 1 more day before
moving up again. Susan adjusted her stop-loss order to include the additional
shares, but kept it at the same price level (13.75). Again she waited. The stock
slowly climbed and reached a minor high of 13.13 before heading down again.
This time the decline was strong enough to punch through the support zone
at the lower trend line.
When the stock descended below point B, Susan lowered her stop-loss
order to 0.15 above that point or 12.15. Then she looked at the measure rule
for the price target. She calculated a target of 9.88 and wondered if the stock
would really reach that price. To be safe, she decided to cash out if the stock
reached 10.15, or 0.15 above the common support price of 10 (a whole number typically shows support).
When the stock plunged to 10.38 on high volume, she wondered if she
was looking at a one-day reversal chart pattern. With those formations, it is
difficult to be sure if prices would reverse or not. She decided to hold on to her
original target.
Two days later, prices zoomed upward and her stop closed out the trade
at 12.15. She did not make much money (about 9% with a hold time of just
over a month), but she gained experience and a few pennies to put in the bank.
For Best Performance
The following list includes tips and observations to help you select better performing broadening bottoms. Refer to the associated table for more information.
• Use the identification guidelines to correctly select a broadening bottom—Table 1.1.
• Trade with the market trend. Select BBs that occur in a bull market
with an upward breakout or in a bear market with a downward breakout—Table 1.2.
• If you see a busted pattern, trade it—Table 1.2.
• BBs in bull markets with upward breakouts have the lowest failure
rates. Avoid trading downward breakouts in a bull market—Table 1.3.
• Performance usually suffers after a throwback or pullback. Search for
overhead resistance (upward breakouts) or underlying support (downward breakouts) before trading—Table 1.4.
• Use partial rises or declines to enter a trade sooner with little increase
in risk—Table 1.4.
• BBs in a bear market bottom quicker than BBs in a bull market, so
short BBs in a bear market for the quickest turnover (but the payoff will
be lower, on average)—Table 1.5
• Watch for the price trend to change 5 to 6 weeks after the breakout—
Table 1.5.
• Select tall or narrow patterns—Table 1.6.
• BBs both tall and narrow outperform the other combinations. Avoid
short and wide ones—Table 1.6.
• Select BBs with a falling volume trend for upward breakouts. A rising
volume trend works better for downward breakouts—Table 1.7.
• Pick patterns with a random volume shape (flat, rising or falling)—
Table 1.7.
Comments
Post a Comment