7 THE T TO PERFORMING BOTTOMS Leson 1:1f your mutual fund owns the stock, there is no need for you to buy it—you already own it. Lesion 2; Don't fall in love with a ‘tock just because it has a way-cool nme, Since those trades I have learned much about stocks and stock mar- ket behavior. In this chapter, I share some of that knowledge. Here are the top 10 buy signals—chart parteras—ranked according to their over- all performance. 1. High, Tight Flag + Average rise rank: | (best) + Breakeven failure rate rank: I (best) * Change after trend ends rank: 1 (best) I received an e-mail from a woman asking me average olen about a symmatical wiangle after | had wold her jordecine (ARP) about a high, sight flag (HTP) I saw forming on her fom the breakeat | chart. She bought at 80 and sold at 135 for a 69% preewineun | gain, Her eanaill alos: shouted, “Somebody swp er Big or ne : Brake ences ‘The brokerage firm she deals with protects the ultimate low, customer accounts up to $25 million. Can you me before I make too much money! surinenceerpute] imagine her teling them, “Ics nor enough. the average. you cover me for $50 million?” Trading the HTF juse might get you there one buck at a time. HITFs are the best performing chart pattern; they have the highest average rise (69%, based on perfect trading with no commissions), lowest breakeven failure race (0%), and best decline afver the uptrend ends (36%). None of the 253 HTFs I looked at failed to rise at least 5% (since writing this, I have found HTFs that moved less than 56). Just five failed to rise at least 10%. However, you have to know how to hunt this animal and how to ride i, or it could be as risky as running with the bulls in Pamplona. Identification Here's what to look for: © Prices should climb at least 96% in cwo months or less.Tight Flag, i * After che rise, find a place where prices pause—a congestion or consolidation area. * Volume should trend downward in the flag. Figure 6.1 shows three examples that appear as pennants. The First ‘one occurs in mid-August 1999 ard is a tight pennant, meaning prices touch the trendlines in a small, well-shaped pattem of price crossings. Prices climb 66% until reaching the September peak and dropping at least 20%, signaling a trend change. ‘The middle HTF occurs in December and tops out at High 2, the top of the loose pennant, for a rise of 74%. Ir also is a tight pattern with prices crossing from trendline 1 cendiine in a narrow cluster of action, ‘The final HF happens less than a month later and price soars 54% after the breakout (co High 3). Notice how the logarithmic scale makes the large price move appear small, On che arithmetic scale, the August HTF is a speed bump in the foothills of the Himalayas, “To find these three FTF, I looked for a rising price trend chat dou- bled in two months or less. The August pattern, for example, reaches its highest peak ac che start of the patern on August 16, at 7.63. On June Abgenix (Drug. NASDAO, 88630 C7 Betnine x High 1 Tight - \ Mma 5 pre s “Tend Start 2 and 3 Bee tei ee ee eee FIGURE 6.1 High, ught flags appea’ as flags, pennants, or random shapes after price doubles in less than two months.TO PERFORMING BOTTOMS 16, ewo months before the HF, prices reached a low of 3 resents a rise of 102% in two months. Prices pause and form a pennant shape, moving sideways then rest- ing, gathering strength for the climb co higher ground. Volume in this example trends upward and that is unusual because it happens just 10% of the time, and performance ofien suffers when ic does occur. The up- ward breakout confirms the pattern za valid one. | found the other two HTFs using the same method—a price doubling in less than wo months autended by price moving sideways. ‘The HTF need not appear as pretty as the ones shown in Figure 6.1. Sometimes the congestion area i irregular, looking like the loose pennant in kate December. With some HTFs, you cant draw a trendline along the tops because it is as ragged as discarded clothing. Still, if HT¥s breakout upward and obey the identification guidelines, chen they are valid pa- terns. For irregularly shaped HTFs, use a close above the highest peak in the pattern as the breakout price instead of a dose above the trendline, That rep- Trading and Trading Tips Once you have correctly identified a HITE how do you trade it? Here are the rules for bu + Wait for price to close above the upper trendline or above the pattern high if the pattern has no top trendline. *+ Buy the stock. * Place a stop below the prior villey, below the pattern itself, or use a volatility stop. With this pattern, the most importance rule is to wait for an upward breakout. In a test of 78 HTFs, I found 13 patterns that broke out downward. Thar might not sound lice much until a failure happens to you. Save your bucks and waic for the upward breakout, “The hardest thing to do when trading HTFs is to buy the stock. Chances are the stock is near the yearly high after doubling in price. How much higher can it possibly go? Buy in and find out! Remember, HITFs have che lowest failure rate and highest average gain of any chart pattern. Your selection may prove the exception and fail, so place a stap- loss order below the valley nearest the breakout. This may be below theTight Flag, 7 HT itself, Check to make sure the stop is not too close. You don't want volatility to stop you out. Larer in the book, I review two HTF trades, fone where I stole two grand from someone and another in which they stole it back. . . with inrerest. ‘When identifying or trading HTFs, what should you look for or avoid? Here is a list: * Avoid overhead resistance that may cause a throwback. HTFs with throwbacks rise just 499, bur the rise averages 100% for patterns without throwbacks. + Avoid loose patterns, Figure 6.1 shows one loose HTF and cwo tight ones. Loose patterns underperform (50% average tise versus 859), Your chance of having a losing wade increases with loose HTFs. * Patterns with breakout volume equal to or below the 30-day aver- age perform beter than de those with heavy breakout volume, 79% average rise versus 63%. + Keep the HTF retrace (the flag portion of the HTF) to less than 36% (the median) of the rior up move leading co the pattern HI Fs with smaller retraces show post breakout gains of 74% ver- sus 63% for patterns with larger retraces. + Avoid HTFs with nearly vertical rises leading to the pattern. Fig- ure 6.2 shows an example of this. Much of the vertical move lasts just afew days, Patterns with moderate rises (typically 45 degrees) climb an average of 70% after the breakout versus 64% for the vertical moon shots, * Wide patterns perform less well than narrow ones—65% average rise versus 7196—s0 keep the flag portion of the HITE width to less than the median 15 days + Select patterns wich a cop tendline thar slopes downward, HTFs with downsloping cop treadlines perform beiter (70% versus 65% average rise). You may decide to trade a stod: with an HTF that has all of the ele- ments suggesting underperformance. Good! Just be sure to watch the stock closely and raise your stop as price makes a new high.