SPY dropped from $653.18 to $631.97 over four sessions in late March 2026. That is a 3.2% slide in under a week. But the damage under the surface started before the index broke. On March 24, advancing issues on the NYSE still outnumbered decliners. By March 26, the ratio had flipped hard, and by March 30, breadth was deeply negative. If you only watched the S&P 500, you saw the breakdown on March 27. If you watched breadth momentum, you saw it a day earlier.
That is the point of the McClellan Oscillator. It measures the momentum of market breadth, not just whether more stocks are going up or down, but whether that balance is accelerating or decelerating. Price indices are market-cap weighted. Five mega-cap names can drag the S&P 500 higher while 400 stocks in the index are falling. The McClellan Oscillator ignores market cap entirely. Every stock counts as one vote. And when those votes start shifting, this oscillator catches it.
What the McClellan Oscillator Measures
Sherman and Marian McClellan introduced this indicator in 1969. The concept is simple: take the daily count of advancing stocks minus declining stocks on an exchange (usually the NYSE), and apply two exponential moving averages to that number. The difference between the fast EMA and the slow EMA is the oscillator value.
If that sounds familiar, it should. The structure is identical to the MACD. The MACD takes the difference of two EMAs applied to price. The McClellan Oscillator takes the difference of two EMAs applied to breadth data. Same architecture, completely different input.
The result is an oscillator that fluctuates above and below zero. Positive readings mean breadth momentum is expanding. Negative readings mean more stocks are falling and the decline is gaining speed. I find this more useful than the raw Advance-Decline Line because the A/D Line is cumulative. It tells you direction but not momentum. The McClellan Oscillator tells you whether participation is accelerating or losing steam.
The Formula
Start with Net Advances:
Net\ Advances = Advancing\ Issues - Declining\ IssuesThen apply two EMAs to that daily Net Advances figure:
McClellan\ Oscillator = EMA_{19}(Net\ Advances) - EMA_{39}(Net\ Advances)The 19-day EMA uses a smoothing factor of:
k_{19} = \frac{2}{19 + 1} = 0.10The 39-day EMA uses:
k_{39} = \frac{2}{39 + 1} = 0.05In the original McClellan literature, these are called the 10% Trend and the 5% Trend. Same math, different label. When the 10% Trend (fast) crosses above the 5% Trend (slow), the oscillator turns positive. When it crosses below, it turns negative.
Some platforms calculate the oscillator using a ratio-adjusted version. Instead of raw Net Advances, they use:
Ratio\ Adjusted = \frac{Advancing - Declining}{Advancing + Declining} \times 1000The ratio-adjusted version normalizes for changes in the total number of listed stocks over time. On modern platforms, most default to the ratio-adjusted version. The interpretation is the same.
How to Read It
The zero line is the dividing line. Above zero, breadth momentum is positive. More stocks are rising, and the trend of participation is expanding. Below zero, more stocks are declining, and that weakness is accelerating.
But the zero line crossing alone is not a trade signal. I treat it as a condition, not a trigger. When the oscillator is above zero, I look for long setups. When it is below, I either sit on my hands or look for short setups. The crossing itself often happens mid-move, not at the turn.
Extreme readings matter more. When the McClellan Oscillator pushes above +100, the market is overbought on a breadth basis. When it drops below -100, it is oversold. These levels are guidelines, not walls. In strong bull markets, the oscillator can stay above +50 for weeks. In crashes, it can plunge to -200 or worse. But readings beyond +100 or -100 flag conditions where a reversal becomes more likely.
I pay more attention to what happens after the extreme than to the extreme itself. An oscillator that reaches -150 and then starts climbing back toward zero is telling you that sellers are exhausted. The decline is losing participation. That information often arrives before the index turns.
Divergences Are Where It Gets Interesting
The highest-value signal from the McClellan Oscillator is divergence with the price index.
Bearish divergence: the index makes a new high, but the oscillator makes a lower high. Fewer stocks are participating in the rally. The index is being carried by a shrinking group of leaders. This pattern preceded several notable corrections. It does not tell you when the top arrives, but it tells you the foundation is weakening.
Bullish divergence works in reverse. The index makes a lower low, but the oscillator makes a higher low. Selling pressure is fading. Fewer stocks are making new lows even as the index drops. This is often the first sign that a bottom is forming.
Consider the late March 2026 decline. SPY closed at $645.09 on March 26 and then fell further to $634.09 on March 27 and $631.97 on March 30. Suppose the McClellan Oscillator hit -120 on March 27 but only -95 on March 30 even as SPY printed a lower price close. That would be bullish divergence: the index fell further, but breadth momentum was already improving. Fewer stocks were participating in the final leg down.
Using It With Other Breadth Tools
The McClellan Oscillator works best as part of a breadth toolkit, not in isolation.
Pair it with the Advance-Decline Line. The A/D Line shows the trend of breadth. The McClellan Oscillator shows the momentum of breadth. When both confirm, you have a strong signal. When the A/D Line is making new highs but the McClellan Oscillator is fading, the rally may be losing steam even though the cumulative count still looks healthy.
The Put/Call Ratio adds a sentiment layer. A deeply negative McClellan Oscillator reading combined with an elevated put/call ratio creates a classic oversold-plus-fear setup. Either signal alone is informative. Both together carry more weight.
I do not use the McClellan Oscillator for individual stock decisions. It is an index-level tool. If breadth is negative and I am looking at a long setup in a single stock, I want to know that the market environment is not hostile to longs. The oscillator gives me that context. If it reads -130, I know the environment is unfavorable for most stocks, even if one name looks technically attractive.
Where It Fails
The McClellan Oscillator is designed for broad exchanges like the NYSE. If you apply it to a narrow index (the Dow 30, for example), the data is too thin. Thirty stocks do not produce meaningful breadth momentum signals. You need hundreds or thousands of components for the oscillator to work as intended.
It also struggles in rotational markets. When money moves from tech into energy, the advance/decline count can stay balanced even though leadership is shifting dramatically. The oscillator might hover near zero while your portfolio gets destroyed because you are concentrated in the wrong sector. It measures breadth, not leadership quality.
Timing is another limitation. The oscillator identifies conditions, not entry points. A reading of -150 tells you breadth is deeply oversold. It does not tell you whether the market bounces tomorrow or continues lower for another two weeks. I have learned to use extreme readings as an alert to start looking for setups, not as the setup itself.
Practical Settings
Most charting platforms use 19 and 39 periods by default. I do not change them. Sherman McClellan chose those periods deliberately, and they correspond to the 10% and 5% exponential smoothing constants that have worked across decades of market data. Optimizing them for recent performance usually just curve-fits the indicator to a specific regime.
For the data source, use NYSE breadth if your platform offers it. The NYSE has the deepest universe of listed stocks. NASDAQ breadth works too, but be aware that NASDAQ includes many small-cap and micro-cap names, which makes the readings noisier. Some platforms offer combined NYSE+NASDAQ breadth. I prefer keeping them separate and focusing on NYSE.
The timeframe is daily. The McClellan Oscillator was built for daily advance/decline data. Intraday breadth data is available on some platforms, but the original research and the +100/-100 guideline levels were calibrated for daily closes. Applying the oscillator to weekly data compresses the signals too much to be actionable.
The McClellan Summation Index
One extension worth knowing: the McClellan Summation Index is a running cumulative total of the McClellan Oscillator. If the oscillator measures breadth momentum, the Summation Index measures the trend of that momentum. It functions the way a moving average relates to price. When the Summation Index is rising, the general trend of breadth is positive. When it crosses above or below specific threshold levels (typically +1000 and -1000), it flags major bullish or bearish regimes.
I use the Summation Index for the bigger picture and the McClellan Oscillator for shorter-term momentum shifts. They complement each other well.
When Breadth Momentum Earns a Spot on Your Dashboard
The McClellan Oscillator fills a gap that price-based indicators cannot. Every indicator on a price chart tells you about that one instrument. The McClellan Oscillator tells you about the entire market. When 70% of stocks are declining but the S&P 500 is flat because three mega-cap names had a good day, price-based indicators on SPY will not flag the problem. The McClellan Oscillator will.
It is not a timing tool. It is a context tool. Use it to know whether the environment supports the trade you are about to take. When breadth is healthy, trend-following setups work more reliably. When breadth is deteriorating, even good-looking setups fail more often. That context alone makes it worth tracking.
Educational content only. Not investment advice. Trading involves risk. You are responsible for your decisions.