Technically Speaking Archives
Technical Indicators & Overlays
The McClellan Oscillator is a breadth indicator developed by Sherman and Marian McClellan in the late 1960s. It is derived from net advances, which is the number of advancing issues of an exchange or index less the number of declining issues. The oscillator is the difference between the 19-day exponential moving average of net advances and the 39-day exponential moving average. As such, the McClellan Oscillator functions as a momentum indicator. The oscillator is also a gauge of market breadth. If an index is rising but declining issues outnumber advancing issues, the rally is narrow.
One drawback of using net advances is that the total number of issues trading on an exchange such as the NYSE or NASDAQ will change over time. One way to account for the dynamic nature of the total number of issues is to ratio-adjust net advances as a percentage of total advancing and declining issues. By using ratio-adjusted net advances, we are able to compare McClellan Oscillator values over extended periods of time. Comparing McClellan Oscillator values today for the NYSE versus values 10 or 20 years ago without such ratio adjustments would be difficult, since the number of issues trading on the NYSE has risen from around 2,000 in 1990 to over 3,500 in 2000. Today, roughly 3,200 issues trade on the NYSE.
Calculating the McClellan Oscillator
As we mentioned, the McClellan Oscillator consists of several components. These include:
- Ratio-adjusted net advances (RANA): (advancing issues – declining issues) ÷ (advancing issues + declining issues)
- 19-day exponential moving average (EMA) of RANA = [(current day’s RANA – prior day’s EMA) × 0.10] + prior day’s EMA
- 39-day EMA of RANA = [(current day’s RANA – prior day’s EMA) × 0.05] + prior day’s EMA
Exponential moving averages reduce the lag of a traditional simple moving average by applying more weight to recent prices. The weighting applied to the most recent price depends on the number of periods in the moving average. There are three steps to calculating an exponential moving average:
1. All exponential averages first start as a simple moving average for the first period’s calculation, since there is no prior-period EMA
2. Calculate the weighting multiplier based on the number of periods of the EMA, such that:
Weighting multiplier = [2 ÷ (time periods + 1)]
In the case of the 19-day EMA, the weighting multiplier is:
= [2 ÷ (19 + 1)]
3. Calculate the exponential moving average.
To learn more about calculating various types of moving averages, see this Spreadsheet Corner article.
Table 1 shows the advance and decline data for the NYSE between May 9 and June 6, 2014. Over this period, the “raw” net advances ranged from +1,791 to –1,278, while the ratio-adjusted net advances (RANA) ranged from +0.571 to –0.405. Remember, RANA is the percentage of net advances to total advances and declines. Therefore, the 0.571 RANA reading on May 12, 2014, means that 54.1% of all advancing and declining issues were net advancers.
Columns F and G of the spreadsheet are the 19-day and 39-day exponential moving average calculations from RANA. Column H, the McClellan Oscillator calculation, is simply the 19-day EMA less the 39-day EMA. Note that the McClellan Oscillator multiplies the difference between the two EMAs by 1,000 to do away with the decimals.
Figure 1 is a three-month daily chart of the ratio-adjusted McClellan Oscillator for the NYSE from StockCharts.com. Using the same advance/decline data that StockCharts.com is presumably using, we discovered that our McClellan Oscillator calculations differ slightly from those on the website. We can only assume that is due to rounding when calculating the RANA or the underlying exponential moving averages.
The moving averages on the bottom panel of Figure 1 are the 19- and 39-day exponential moving averages for the “raw” net advances, not the ratio-adjusted net advances. While the values of the raw net advances and RANA will differ, the shape, relative location and direction are the same. For example, on June 6 the 19-day EMA of the RANA (from Table 1) is 0.1594 (or 159.4 when multiplied by 1,000) while the 19-day EMA of the “raw” net advances from Figure 1 is 498.06.
Interpreting the McClellan Oscillator
The McClellan Oscillator is sometimes described as MACD (moving average convergence/divergence) of the advance-decline (AD) line, which is an apt descriptor. The AD line is a cumulative measure of net advances and the MACD is a momentum indicator that deducts a longer-period moving average of prices from a shorter-period one.
Looking at the area chart for the NYSE ratio-adjusted McClellan Oscillator in Figure 1, we see that it is positive whenever the 19-day EMA is greater than the 39-day EMA. Furthermore, the area chart moves between positive and negative territory whenever there is a crossover of the two moving averages.
When interpreting the McClellan Oscillator, you can apply the signals used for MACD as well. Traders use the McClellan Oscillator to identify bullish and bearish periods in the market; look for bullish and bearish divergences to identify potential reversals; and look for “breadth thrusts” that may signal the start of an extended market move.
Positive vs. Negative
Even though it is based on two moving averages, the McClellan Oscillator can be quite volatile. This is because it uses exponential moving averages, which are more reactive to individual data points as compared to a simple moving average, where all data points are equally weighted. There are times, however, when the McClellan Oscillator will remain positive or negative for extended periods of time. This typically corresponds with a strong uptrend or downtrend in the underlying index.
Figure 2 is the McClellan Oscillator (ratio-adjusted) for the NYSE between July and December 2011. The lower panel of Figure 2 is the NYSE Composite index, which serves as the base index. We see that the McClellan Oscillator moved into negative territory in late July (2011) and remained there for nearly three weeks. In the lower panel, we see that during this same period the NYSE Composite entered a steep downtrend.
Immediately following this downtrend, the NYSE Composite entered a period of sideways trading for several weeks. During this time, the McClellan Oscillator entered a period of indecision as it “oscillated” around the zero line, moving between positive and negative territory. While some of the moves were quite strong, indicating strong short-term buying and selling pressure, none lasted long enough to indicate a meaningful trend in the NYSE Composite.
In October, the McClellan Oscillator went positive and stayed that way for roughly four weeks. Again, looking at the NYSE Composite in the lower panel, we see the index entered into a distinct uptrend. This time, the buying pressure held for a longer period of time. A healthy bull market has a large number of stocks making similar upward moves, which would mean a high number of net advancing issues.
Figure 2 ends with another period of indecision as the oscillator once again bounces between positive and negative territory while, again, the NYSE Composite enters a period of sideways movement.
Like many momentum oscillators, the McClellan Oscillator indicates possible overbought situations when it reaches the +70 to +100 range, and shows possible oversold conditions in the –70 to –100 range.
Divergences are a powerful tool in technical analysis and can be used to identify possible reversals in a security or index. A divergence takes place when the indicator moves in one direction while the index or security moves in the opposite direction. Bullish divergences occur when the indicator is making higher lows while prices are making lower lows. In contrast, a bearish divergence develops when the indicator is making lower highs as prices are making higher highs. In the case of the McClellan Oscillator, bullish and bearish divergences point to possible reversals in the underlying index.
We can attest that it’s much easier to identify a pattern after the fact than while it is developing. For this reason, it is important to avoid identifying a pattern where one really doesn’t exist. While they are often few and far between, there are the occasional “big fat obvious signals” that, if you are fortunate enough to identify them, you can profit from.
Divergences are relatively easy to identify, but finding those that result in reversals are more difficult to come by. Figure 3 is an example of a bearish divergence that did foreshadow a reversal in the underlying index. This chart shows the NASDAQ McClellan Oscillator (ratio-adjusted) and the underlying NASDAQ Composite index. The NASDAQ Composite entered a strong uptrend in early February 2014, and beginning in the middle of the month the McClellan Oscillator started forming noticeably lower highs while the NASDAQ Composite kept achieving higher highs. In March, building selling pressure pushed the McClellan Oscillator into negative territory. After this confirmation of the bearish divergence, the NASDAQ Composite lost nearly 6% before hitting a low in mid-April.
Figure 4 illustrates a bullish divergence between the NYSE McClellan Oscillator and the NYSE Composite index. The composite started a pronounced downtrend in early August 2013, but beginning in late August the McClellan Oscillator started making higher lows while the NYSE Composite continued to mark lower lows. In early September, the bullish convergence is confirmed when the McClellan Oscillator moves into positive territory. Following this, the NYSE Composite gained over 4% in less than three weeks.
A final way traders use the McClellan Oscillator is for identifying “breadth thrusts.” This shouldn’t be confused with the market momentum indicator developed by the late Martin Zweig, although the two are somewhat similar. Zweig’s Breadth Thrust indicator is a 10-day simple average of the percentage of advancing issues to the total number of advancing and declining issues. In contrast, the McClellan Oscillator uses exponential averages of the net advancing issues as a percentage of total advancing and declining issues.
In the context of the McClellan Oscillator, a breadth thrust occurs when the oscillator moves rapidly from extreme negative readings to strong positive readings. Most traders look for at least a +100-point move that starts below –50 and moves above +50. Such moves are bullish in nature and sometimes mark the beginning of a prolonged advance in the underlying index.
Breadth thrusts are more meaningful if they are preceded by a bullish divergence, which we discussed earlier. Figure 5 is the same period used in the bullish divergence example from Figure 4. As it turns out, this bullish divergence did develop in advance of a breadth thrust. The McClellan Oscillator bottomed out around August 19, but the NYSE Composite continued declining for another two weeks as the oscillator started forming higher lows. This bullish divergence signaled a possible reversal and the breadth thrust from –100 to +55 (+155) confirmed the reversal.
The McClellan Oscillator measures the momentum of net advancing issues, typically of indexes such as the NYSE and NASDAQ composites. As we have shown, you can use the oscillator in a variety of ways. Bullish and bearish divergences offer an indication of a possible reversal in the current trend. The same goes for breadth thrusts, which can be used to confirm such reversals.
It is important to keep in mind, however, that no single technical indicator is infallible. That is why technicians use a variety of indicators to confirm or refute signals.