How broad is the market, and once we answer that question, how can we use that answer to our advantage? Unfortunately, it’s impossible to tell how much money was spent on stocks in the market today as long-term investments and how much was spent for purposes of speculation. Still, it’d be awfully helpful to know. If more money is spent on long-term investments, then it's more likely that the market might be overheating and headed for a correction.
Over the decades, several of the industry’s most formidable analytical minds have attempted to figure out what’s in store for the market by looking at how stocks en masse have risen, fallen, and continue to rise and fall. One of the most enduring measures of that breadth of daily winners and losers is the McClellan oscillator, which is used widely among both professional and amateur technical analysts.
Look at Gainers and Decliners
The oscillator was developed in 1969 by Sherman and Marian McClellan, who were either the only husband-and-wife team in the history of technical analysis or one of the exceedingly few. Mrs. McClellan died in 2003, her widower and their son handling much of the analytical development since. The family’s oscillator is a somewhat intricate concept, but not so much that it can’t be explained in a few hundred words.
If you’ve ever seen even a single nightly stock market summary, you’ll notice that it typically lists the number of gainers and decliners on that particular day. For instance, on March 4, 2014, a total of 2,598 issues on the New York Stock Exchange (NYSE) advanced while 536 declined. The difference between the two, 2,062, is the market’s "daily breadth." Taking it one step further, one day’s breadth carries over to the next, making for a cumulative total called the "advance-decline line" (A/D line).
Of course, the numerical value of the advance-decline line depends on what day you began measuring. In practice, the current advance-decline line is not the running total of every single net advance since the NYSE opened for business in 1817. Instead, analysts can start on any day that’s convenient, because the last days of the line will take on the same shape regardless of which starting point you choose.
That’s what analysts are interested in when they study the advance-decline line: not the raw numbers, but what the line looks like and what direction it seems to be heading in. The idea is to rate how viable the current trend is; a falling market coupled with a rising A/D line can mean that the market is set to rebound and vice versa. The market and the A/D line normally move in tandem, but certainly not always.
Recent vs. Older Movements
The McClellans understood that when assessing a trend, the recent movements ought to count for more than the older ones. As to how much more, we determine the answer by using an "exponential moving average." Essentially, you take the most recent data point on the A/D line, then the second most recent (multiplied by a constant), then the third most recent (multiplied by the constant squared), then the fourth most recent multiplied by the constant cubed, etc. The greater the constant, the more the moving average is weighted toward the present.
Analysts use the term “smoothing” to describe that weight. For instance, a “10% smoothing constant” would mean counting today’s A/D line value at 10% and yesterday’s at 90% when calculating the trend. Don’t forget, yesterday’s exponential moving average itself was comprised 10% of that day’s A/D line value and 90% of the then-previous day’s, etc.
As to how far in the past to go to deduce a trend, the choices are overwhelming. You could go one day back, which would make for easy arithmetic if not very useful data, or you could go 72,000 days back to the NYSE’s inception. The McClellans found it practical to use both 19- and 39-day periods. Specifically, their oscillator takes the 19-day exponential moving average and then subtracts the 39-day exponential moving average. For each of those, the McClellan oscillator uses both a 5% and a 10% smoothing constant.
High Means Bullish
Once you calculate the McClellan oscillator, you’re left with either a high number that’s supposed to indicate bullishness, or a low one that you can probably figure out is a sign of bearishness. “Bullishness” in this context means that the trend shows that money is coming into the market disproportionately. At some point, the bullish numbers become so high that some technical analysts feel that a crash or normalization is inevitable. Similarly, a strong bearish number may indicate that stock prices are bound to shoot up.
The McClellan oscillator for a specific stock exchange might be useful with respect to that exchange, but it can’t easily be compared to that of other exchanges, nor can it be used to establish universal trends. The more issues that trade on an exchange, the greater the potential for large net advances or declines and the greater the possible range for the oscillator. In theory, you could even use the McClellan oscillator to measure the breadth in a grouping as small as your own portfolio of stocks.
The Bottom Line
The McClellan oscillator moves around a lot, and sometimes it can be hard to tell which of those movements are meaningful and which are just random noise. For the investor committed to technical analysis, the oscillator offers objective data treated in a unique way, and hopefully a profitable one, if interpreted correctly.