Breakouts, Pullbacks, And Gaming The 2B

Writing about the 2B test in his book, Trader Vic: Principles Of Professional Speculation, Victor Sperandeo notes of what he has come to call the "2B Indicator":

In an uptrend, if prices penetrate the previous high, fail to carry through, and then immediately drop below the previous high, then the trend is apt to reverse. The converse is true for downtrends.

This observation holds true for all time periods.... If a 2B is to occur in a short-term trend (3-5 days), then after the new high is established, prices will usually fail below the previous high or low within a day or two. In an intermediate-term trend, the 2B often marks the beginning of a correction and usually occurs within 7 to 10 days after the new high or low is made.

Looking at how time factors into the 2B trade (I can't bring myself to call it an "indicator") as Sperandeo describes it encouraged me to consider another aspect of this setup: How long after entry should a trader expect to be profitable if, in fact, the 2B has caught a true change in trend. While later Sperandeo suggests that "even if you get whipped out two, three or even four consecutive times, the one you catch will end up making you a bundle," that would not necessarily have been the case with the trader trying to sell 2B tops against the semiconductor index from July 2006 through December. Rather, it appears that the trader willing to take short-term, countertrend plays against the $SOX rally could have done fairly well -- if he or she had an idea of how long to let the 2B play out before the uptrend was restored and the short play stopped out (or, worse, exited at a loss).

FIGURE 1: PHILADELPHIA SEMICONDUCTOR INDEX, DAILY. Three of the five key breakouts in the $SOX bull market were followed by reversals that were relatively sharp and sudden. But insofar as those reversals did not lead to significant lower lows, none of them truly threatened the rally.

In all three of those failed breakouts, traders would have been provided sell signals based on the 2B and Turtle Soup methodologies. And while none of those sell signals would have correctly anticipated a reversal in trend, all three would have given fair warning for traders who were long semiconductor stocks to take profits, or for short-term, hourly chart traders to make a little money, perhaps with carefully placed two- or three-day short plays.

FIGURE 2: FEDEX, DAILY. A classic 2B top in the spring of 2006 sent shares of FDX tumbling some 13% before bottoming in the summer. The 2B top is approximately 20 trading days wide. Using the 50% time target "rule of thumb," a trader shorting the 2B top would have made 6% in 10 days. The highlighted candlestick marks the short entry day.

Consider the three 2B tests in Figure 1 -- and recall that the $SOX is tradable as an option play, even if there is no way to trade the $SOX outright (SMH, the HOLDRS exchange traded fund, would be one alternative). In the first instance, there are two highs, August 17 and August 30, that make up the 2B test of top. There are 10 trading days between them. Entering short at 441.15 -- the low of the initial high -- on September 5, we find that by the fifth day after the second high -- September 7 -- the $SOX is closing at 435.58. That ends up being the low close of the correction, as the $SOX reverses shortly afterward and begins moving up again.

FIGURE 3: PHILADELPHIA SEMICONDUCTOR INDEX, DAILY. This 2B top appeared after a near-vertical move higher in the $SOX. Ten days wide, this pattern found its correction low approximately five days after the second higher high, putting 440 puts on the $SOX in the money before reversing and resuming the uptrend.

FIGURE 4: PHILADELPHIA SEMICONDUCTOR INDEX, DAILY. The 2B top in September and October developed as part of a consolidation range. Twenty trading days wide, the closing correction low arrived after five days, but the 50% rule of thumb still would have allowed traders to catch at least a two-strike move as the 460 and 455 strikes were taken out. The horizontal blue line represents the level at which the short signal would have been triggered.

An approach like this that lets traders take advantage of short-term tops but allows for the possibility of those short-term tops becoming more long-lasting is a worthwhile strategy to consider. While Sperandeo insists that, sooner or later, a lasting 2B top will arrive, this method of tempering expectations somewhat and looking to lock in whatever gains the market provides is ideal for countertrend bets that, inevitably, will prove to be losing propositions from time to time. This approach is also suitable for trading around a position such as a longer-term bullish stance, which (using the semiconductors as an example) might have been adopted shortly after the July bottom.

SUGGESTED READING
Connors, Laurence, and Linda Bradford Raschke [1996]. Street Smarts, M. Gordon Publishing Group.
Penn, David [2003]. "Building A Better Breakout," Working Money.com, July 8.
Sperandeo, Victor [1997]. Trader Vic: Principles Of Professional Speculation, John Wiley & Sons.