When is selling overbought indices or buying oversold ones a bad idea? Many traders trying to short indices or stocks over the last three years would answer, "More often than you might think."
My own research has confirmed the truth of that answer. Breakout plays on Donchian channels sometimes perform better when oscillators such as stochastics indicate that the movement is already overdone.
Annotated Daily Chart of the OEX:
Breakdowns out of the Donchian channels also tended to perform better if stochastics were pinned at levels signaling that prices were overdone to the downside.
A recent article discussed overbought and oversold levels. Depending on the market being watched and the time frame in which it's watched, stochastics' oversold and overbought levels are sometimes indicated at 30 and 70, or perhaps 20 and 80. Most charting services set them by default at 30 and 70.
My own observations and those of many would-be bearish traders over the last several years confirm that it's not always a good idea to assume that a trend reversal is in the making when those overbought or oversold levels are reached. Instead, as Richard Donchian and others have proposed, trading in the direction of the trend often proves more lucrative, and that often means buying in the face of overbought indicators and selling in the face of oversold ones. An Internet search of trend-following tactics turns up any number of sites willing to enroll new subscribers and provide them with their proprietary trend-following signals.
The thrust of this article is different. This article's intention is to help traders, particularly newbie traders, identify those times when it's dangerous to act on overbought or oversold signals. The article focuses on keeping traders out of trouble rather than getting them into trades.
When prices continue to bounce from a key moving average (MA), overbought oscillator signals should probably be ignored except to warn those in long plays to protect profits. Overbought conditions can be worked off by sideways movement or even by a temporary slowing of upward momentum, but profit-protecting plans of course need to be in place in case prices plunge.
Annotated Daily Chart of the SPX:
After stochastics had signaled extreme overbought levels, a first bearish warning was given when there was a bearish cross of the stochastics line. Bearish divergence had been created, when comparing the September and early November stochastics highs. Traders who jumped in with a bearish trade suffered through a few days of choppy sideways trading into the rising 10-sma before the SPX bounded higher again and stopped them out of their plays. Neither the overbought readings nor the bearish divergence produced a significant downturn. The trend remained intact. As long as moving average support was being maintained, bearish trades did not prove to be a good idea.
One chart exhibiting a long-term uptrend is that of Boeing (BA). While moving higher, the company's stock price has maintained the support of the weekly 50-sma. In addition, BA's weekly chart--prepared as of March 15--demonstrates many other warning signals that, despite overbought setups, a reversal should not be assumed.
Annotated Weekly Chart of BA:
BA's chart exhibits many of the signs that warn that only adept traders should have acted on a stochastics sell signal for the last three years. Those warning signs of an intact trend include prices that still bounce from a key moving average, prices that bounce from a rising trendline or rising regression channel's support, and prices that do not retrace below the last consolidation zone's top resistance line. If any or all of those conditions exist, then traders should consider sell signals generated by stochastics or other oscillators suspect. As is clear above, swing traders might have benefited from bearish plays when BA pulled back from the top of the rising regression channel to retest the supporting line, but even that pullback could not be assumed. Last summer's coiling move required some skill to know when to get in and exit in order to capture any profit.
The previous examples have all concerned rallies, but the same cautions should apply when prices trend lower.
Annotated Daily Chart of GM:
As these charts have demonstrated, trusting oscillator evidence alone can be a risky endeavor, especially when that evidence suggests a countertrend move. Neither divergences, bearish or bullish crosses, extreme levels or any of the other typical oscillator signals is enough to prove that a trend is about to reverse. Entering countertrend plays on such oscillator signals can deplete trading accounts.
A first task for any trader, then, is to identify when a trend remains intact. The charts in this article have illustrated some signs. When a key moving average continually serves as support or resistance, the trend remains intact. The same is true of support or resistance from rising regression channels or trendlines. Prices that never pull back below previous resistance or bounce up past previous support show that uptrends or downtrends remain intact.
When these signals exist, knowledgeable traders avoid countertrend plays or
limit position size when they enter such plays. Let price prove that a trend
reversal is underway. Those who do otherwise risk shorting a rallying market all
the way higher or buying the dips all the long
time a market drops.