As in real estate, the three most important things in pattern analysis might be said to be location, location, and location.
Consider a pattern contained in Figure 1, which shows the price action in the S&P 500 ETF (SPY) for several days in early October 2011. If someone asked you to summarize the intraday price action for Oct. 12 (the highest bar on the chart), you might answer, “After the open, the market traded strongly higher but reversed at some point during the day to close very near the bottom of the day’s range, and also very near the opening price.” If someone asked you what that price action potentially implied, you might say, “The market retreated after testing higher levels, which suggests declining upside momentum or bearish sentiment.”
It’s as good a guess as any, and a reasonable explanation — given the absence of any other factors. But it must be said the pattern’s assumed implication is very likely influenced by the fact that Oct. 12 was the fifth consecutive visible up day, established the highest price level on the chart, and was followed by a day that made a lower high, lower low, and lower close. And this explanation is, in fact, precisely the one typically attributed to this pattern, variations of which are often referred to as reversal, or “key reversal,” days. (Candlestick pattern adherents have their own pattern names, although the implications are similar.)
Figure 2 illustrates why location is everything. The highlighted bars include Oct. 12 (far left) but also four more days with the same general characteristics: up movement after the open but a close near the open and in the lower portion of the day’s range. The lighter blue arrow marks the only other day of this group that made a high above the previous day’s high. But all the other highlighted days besides Oct. 12 (except the final one, Dec. 6) were followed immediately by at least a few more days of upside price movement, and all of them except Dec. 6 were followed by a higher open the next day (but even in that case the market closed higher the next day). Finally, SPY traded more than 6 percent higher after the Oct. 12 “bearish reversal” day despite the one-day correction that immediately followed it. So much for bearish implications.
However, perhaps there is something of value hidden in this apparent pattern failure — namely, the tendency for days that open and close at roughly the same level and in the lower portion of the day’s range to be followed by a higher opening price the next day. The middle three highlighted bars in Figure 2 were all followed by significantly higher opens. If this pattern proves to have an edge, it could potentially provide the basis for a useful overnight trade setup in SPY.
|
|