Thomas N. Bulkowski - Encyclopedia of Chart Patterns

If the stock breaks out downward, the measure rule computation is nearly the same. Subtract the pattern's height from the lowest low, giving a target price of 37.50. Be aware that upward breakouts in bull markets are more likely to reach their targets (65%) than other combinations.

The lower portion of the table shows how often price will reach the target based on various heights. For example, if you slice the height of the pattern in half and use that in the measure rule computation, you'll find price will reach the target 82% of the time if your pattern behaves like the average pattern.

Once you've computed a target, change the difference between the target and the current price into a percentage of the current price. Then use Table 10.3to see how often the stock will fail to exceed the move.

Using our example, the target is 6 points away from the top of the pattern (49.50, which we assume is the current price), or 12% above A (or 100 × 6/49.50). Table 10.3shows that 32% will fail to see price rise more than 10%, so the failure rate would be higher for a 12% target. That also means 68% of the trades will see price exceed a 10% rise. That sounds like a reasonable number.

Table 10.10 Trading Tactics

Wait for breakout.Because the breakout direction isn't known until price closes outside one of the trendline boundaries, don't try to anticipate the breakout direction. Price may reverse at the trendline, handing you a loss.

Stops location.Let's say price breaks out upward. Compute the target price and make a profit‐and‐loss assessment of the potential trade. What is the likely downward move compared with the target price? Does the potential profit justify the risk of the trade? For Figure 10.7, there is support in the 46‐to‐47 area.

Looking for prior peaks and valleys helps determine support and resistance levels. In March 1995 (not shown in the figure), there was an area of congestion bounded by a symmetrical triangle with an apex at about 46. Additional resistance appeared in July and October, as shown. Together, the 46‐to‐47 area made a good location for a stop‐loss order.

Let's say the stop price you select is at 45.75, just below the bottom of the support area. If the breakout price is 50.50 (which is the close the day after the upward breakout), that gives a potential loss of less than 10%. With a target price of 55.50, or 10% upside, the win/loss ratio is an unexciting one‐to‐one. In such a situation, you could either tighten your stop by moving it higher (and risk getting taken out by normal price action) or look elsewhere for a more profitable trade.

Remember there is no rule that says you have to place a trade. Let me also say that I'm not a fan of win/loss ratios. If you trade patterns well, the profit should come.

Intraformation trading.If the broadening pattern is tall enough, go long after price rebounds off the lower trendline and hope for an immediate upward breakout. Only try this if you're an experienced swing trader and only after the pattern passes all of the identification guidelines.

Partial decline.A partial rise correctly predicts a downward breakout 47% of the time. However, a partial decline correctly predicts an upward breakout 75% of the time. So if you can tell when a partial decline is in place, meaning the broadening pattern is fully formed (see “Partial decline” in the Glossary for details), then consider buying the stock and hoping for an upward breakout. The partial decline might be a pause that happens as price moves to the lower trendline, so if price breaks out downward, then close out the position. Otherwise hold onto the trade and hope price starts to head back up.

Busted trade.With a 60% average rise after a single busted downward breakout, busted patterns might be the way to profit from this chart pattern. See Table 10.9for details.

I have traded this chart pattern a number of times. Let me tell you about what I found in my trade review.

Let's start with three trades in Cisco Systems. All of them traded the same broadening pattern that formed in June to September 2000.

In the first trade, the stock looked to be doing a partial decline. The pattern had a flat top with three trendline touches and three touches of the lower trendline, too, just as you would expect. The pattern looked ripe to break out upward in the high‐tech boom of the 2000 market (but the general market turned bearish in March, before I bought. I'm not sure I knew that at the time).

I bought the stock after it broke out of a small congestion area (it was a small downward trending pennant with an upward breakout buried within the broadening pattern). I expected the stock to break out upward from the broadening pattern. Instead, it touched the top trendline and headed lower. As a swing trade, it would have been good to sell then, but I didn't. I rode the stock lower and sold after a downward breakout.

The second trade was also within the same pattern, but I bought as price dropped before breaking out downward. This was a “hip shot,” a trade I just glance at and make, confident that it'll work. Each time I pen those words in my trading notebook, I know the trade is going to be a loser (that's what I learned over the years, perhaps well after this trade). Hip shots just don't work for me. Now, I take my time to analyze the situation before trading.

Anyway, I thought the stock was going to turn, but it didn't. It kept going down.

With the third trade, I bought at the lower trendline, expecting an upward bounce. Instead, it closed the day outside the pattern's boundary, breaking out downward from the pattern. The next day, I closed out all three trades and took a loss of 15%, 8%, and 4% on them, respectively.