McClellan Summation Index for Indian Markets: Reading Real NSE Breadth

What the McClellan Summation Index Actually Measures

The McClellan Summation Index is a breadth indicator, which means it ignores price and instead counts how many stocks are participating in a move. It is the running cumulative total of the McClellan Oscillator, which Sherman and Marian McClellan built from advancing and declining issues. On the NSE this matters because the Nifty 50 is capitalisation weighted, so a handful of heavyweights like Reliance, HDFC Bank, ICICI Bank, Infosys and TCS can drag the index up even while most of the other 1,900-plus listed stocks are falling. The MSI is designed to expose exactly that gap between the headline index and the average stock.

Think of the Oscillator as today's breadth momentum and the Summation Index as the breadth bank balance. Each day the Oscillator value is added to the prior MSI total. A long run of positive Oscillator readings makes the MSI climb steadily, signalling broad, healthy participation. A long run of negative readings drains it lower. Because it accumulates, the MSI is slower and smoother than the Oscillator, which is what makes it useful for spotting the multi-week shifts that matter for swing and positional traders on Indian indices.

The raw input is the NSE daily advance and decline figure, which the exchange and every broker terminal publish at the close. On a typical NSE session somewhere between roughly 1,800 and 2,100 equity symbols trade actively. If 1,400 close higher and 600 close lower, that is 1,400 advances and 600 declines, a net breadth of plus 800 for the day. That single net number is the seed from which the entire Oscillator and Summation chain is grown.

The Two NSE Versions: Raw and Ratio-Adjusted

There are two ways to feed NSE breadth into the McClellan family, and choosing the wrong one quietly corrupts your back-history. The raw classic version uses (advances minus declines) directly. This was fine when the number of listed stocks was stable, but the NSE listed universe has grown over the decades, so a net breadth of plus 500 in 2008 is not the same proportional event as plus 500 today. The raw MSI therefore drifts and its old levels are not comparable with new ones.

The ratio-adjusted version (RANA) fixes this by using (advances minus declines) divided by (advances plus declines), usually multiplied by 1,000 to keep the numbers readable. Because it normalises by total issues traded, a reading from the 2008 crash is directly comparable with a reading from 2020 or today. For any serious historical study of Indian breadth you should use the ratio-adjusted form. All the historical levels discussed later in this page are stated on the ratio-adjusted basis and are illustrative reconstructions, because the NSE does not publish an official MSI.

• Raw classic: Oscillator built from (advances minus declines). Simple, but old and new levels are not comparable as the listed universe grows.
• Ratio-adjusted (RANA): Oscillator built from (advances minus declines) divided by (advances plus declines) times 1,000. Comparable across decades and the correct choice for NSE history.
• Neutral line: the MSI is conventionally re-centred so that zero represents neutral breadth; readings far above or below zero mark breadth extremes.

Step by Step Calculation From NSE Advance-Decline Data

The calculation has three layers and each builds on the one before it. First, compute the daily breadth input from the NSE advance and decline count. Second, smooth it into the Oscillator using two exponential moving averages, a fast 19-period and a slow 39-period EMA, where the 19-period uses a smoothing constant of 0.10 and the 39-period uses 0.05. The Oscillator is the fast EMA minus the slow EMA. Third, add each day's Oscillator value to the previous day's Summation Index to get today's MSI.

• Step 1: Daily breadth. Ratio-adjusted = (Advances minus Declines) divided by (Advances plus Declines), times 1000.
• Step 2: Fast EMA. Today's 19-day EMA = prior 19-day EMA + 0.10 times (today's breadth minus prior 19-day EMA).
• Step 3: Slow EMA. Today's 39-day EMA = prior 39-day EMA + 0.05 times (today's breadth minus prior 39-day EMA).
• Step 4: Oscillator = 19-day EMA minus 39-day EMA.
• Step 5: Summation Index = prior day MSI + today's Oscillator.

Worked NSE Breadth Example, Day by Day

Here is a fully worked, illustrative example using realistic NSE numbers. Assume on a session 1,450 stocks advance and 600 decline, so total issues are 2,050. Ratio-adjusted breadth = (1450 minus 600) divided by 2050, times 1000 = 850 divided by 2050 times 1000 = about 414.6. That single value of roughly 415 is today's breadth input. Suppose the prior 19-day EMA was 300 and the prior 39-day EMA was 250.

New 19-day EMA = 300 + 0.10 times (414.6 minus 300) = 300 + 11.46 = 311.46. New 39-day EMA = 250 + 0.05 times (414.6 minus 250) = 250 + 8.23 = 258.23. Today's Oscillator = 311.46 minus 258.23 = about plus 53.2. If yesterday's Summation Index stood at plus 1,200, today's MSI = 1,200 + 53.2 = about 1,253. A positive and rising Oscillator pushing the MSI higher confirms that breadth is broad and improving, not just a Reliance-and-HDFC-Bank rally.

Reading the Index: Levels, Slope and Zero-Line Crosses

Three things matter when you read the MSI: its level, its slope and its zero-line crosses. A high and rising MSI says broad participation is fuelling the trend, which is the healthiest backdrop for staying long Nifty or Bank Nifty. A high but flattening MSI is an early warning that the engine is losing cylinders even if price is still ticking up. The slope change often leads the price top by a few sessions, which is the whole point of watching breadth.

Zero-line crosses are the cleanest mechanical signal. When the MSI crosses from below zero to above, it marks a shift from net distribution to net accumulation across the market, historically a constructive sign for the following weeks. A cross from above to below zero is the mirror image. On Indian indices these crosses are most reliable when they coincide with the Nifty reclaiming or losing a well-watched level such as its 200-day moving average, because then breadth and price are telling the same story.

2008 and 2020: What NSE Breadth Actually Did

The original version of this page claimed the MSI went negative in 2008 and 2020 without any figures, which is not useful. Here is the real context. In the 2008 global financial crisis, the Nifty 50 fell from a January 2008 high near 6,357 to roughly 2,524 by 24 October 2008, a drop of about 60 percent over the year. NSE breadth was brutally negative for months: sessions with 1,500-plus declines against only a few hundred advances were routine, so the ratio-adjusted Oscillator stayed deeply negative and dragged the Summation Index into one of its most extended negative runs on record. The MSI did not bottom on the exact price low; it began curling up from its deep negative reading in late 2008, weeks before the Nifty found its durable bottom near 2,539 in late October and began the 2009 recovery.

The March 2020 COVID crash was faster and sharper. The Nifty fell from a January 2020 high around 12,362 to an intraday low near 7,511 on 23 March 2020, roughly a 39 percent fall in weeks. Breadth collapse was near total: on the worst sessions of March 2020 advances on the NSE were in the low hundreds while declines ran past 1,500, producing some of the most negative single-day ratio-adjusted breadth readings ever seen, well past minus 700 on the times-1000 scale. The Summation Index plunged accordingly. Critically, the MSI turned up in the last days of March 2020, ahead of the V-shaped recovery that carried the Nifty back above 11,000 by year end.

The Nifty levels above are real and verifiable from NSE history. The MSI values are illustrative reconstructions, because no exchange publishes an official Indian McClellan Summation Index, and your exact figures will depend on whether you used the raw or ratio-adjusted version and where you anchored the start of your cumulative total. The behaviour, deep negative readings that turned up before the price low, is the durable and repeatable lesson, not any single number. None of this is a promise of future returns.

Breadth Divergence: The Highest-Value Signal

The trade setup that justifies watching the MSI at all is divergence. Because the Nifty is cap-weighted, it can grind to fresh highs on the back of four or five mega-caps while the broader market is already rolling over. When the Nifty prints a higher high but the MSI prints a lower high, breadth is diverging: fewer stocks are supporting each new index high. This thinning participation has repeatedly preceded meaningful Nifty pullbacks, because once the few leaders wobble there is nothing underneath to hold the index up.

The opposite, bullish divergence, is just as useful near bottoms. If the Nifty grinds to a marginal new low but the MSI makes a higher low, selling pressure is broadly exhausting even though the headline index looks weak. That is often the first objective hint that a base is forming. Combine this with a momentum tool to time the actual entry rather than acting on the divergence alone.

• Bearish divergence: Nifty higher high, MSI lower high. Participation is thinning, treat new longs with caution and tighten stops.
• Bullish divergence: Nifty lower low, MSI higher low. Selling is broadly drying up, watch for a basing pattern.
• Confirmation: pair the divergence with RSI or MACD on the index and with a price level such as the 200-day moving average before acting.

A Worked Bank Nifty Trade Using a Breadth Turn

Here is an illustrative options example that shows how a breadth signal converts into rupees, including costs. Assume the MSI has been deeply negative and, after a broad NSE selloff, it ticks up for two straight sessions while Bank Nifty stabilises near 48,000. You read this as early breadth recovery and buy one lot of a Bank Nifty monthly 48,000 call. The Bank Nifty lot size is 30. Say the call premium is 300 per unit, so your cost to enter is 300 times 15 = Rs 4,500 plus charges.

Over the next two sessions breadth keeps improving and Bank Nifty rallies to 48,700, lifting the call premium to 520. You sell at 520, so the gross gain is (520 minus 300) times 15 = 220 times 15 = Rs 3,300. Now subtract costs. STT on options is charged at 0.15 percent on the sell-side premium, so 0.0015 times 520 times 15 = about Rs 11.70. Add broker flat fees of roughly 20 per order on entry and exit (about Rs 40), plus exchange transaction charges, SEBI fee, stamp duty and 18 percent GST on brokerage and transaction charges, which together come to roughly another 30 to 60 rupees on a trade this size. Call total charges roughly Rs 95 to Rs 115. Your net profit is approximately Rs 3,300 minus about Rs 100 = around Rs 3,200, illustrative only.

Combining MSI With Price and Momentum Tools

The MSI works best as a filter on top of price, never as a standalone trigger. A practical workflow is to let the MSI tell you whether breadth supports being long or short, then use a price and momentum tool to time the precise entry. If the MSI is rising and above zero, you favour long setups and trust Nifty breakouts more. If it is falling and below zero, you fade rallies and respect resistance. This keeps you on the right side of broad participation.

A clean confluence example: the MSI crosses above zero, the Nifty reclaims its 200-day moving average, and MACD prints a bullish crossover in the same week. Three independent tools agreeing on a broad accumulation phase is a far stronger context for risk than any one of them alone. When they disagree, the disagreement is itself information and usually a reason to size down or stand aside.

Limitations, False Signals and Indian Market Quirks

The MSI is a lagging, smoothed indicator by construction, so it will not call exact tops and bottoms and it can whipsaw in choppy, range-bound markets. During a sideways Nifty phase the index can flip back and forth across zero, generating signals that go nowhere. In those conditions you should widen your interpretation and lean on price structure instead of acting on every minor cross.

There are also India-specific quirks to respect. Breadth can be temporarily distorted on big expiry days, on days dominated by a single sector event, and during periods when index inclusions, exclusions or fresh IPO listings change the composition of the traded universe. A burst of new small-cap listings, for example, can inflate the advance count for reasons that have nothing to do with the health of the Nifty. The ratio-adjusted version softens but does not fully remove these effects, so always read the MSI alongside the actual NSE advance and decline numbers rather than blindly.

• Lagging by design: do not expect pinpoint tops or bottoms, expect early-warning context.
• Whipsaws in range-bound Nifty phases: zero-line crosses lose reliability when price is sideways.
• Expiry-day and single-sector distortions can produce one-off breadth spikes.
• Listing and index-reconstitution changes alter the traded universe, so prefer the ratio-adjusted version and cross-check raw A/D counts.

Sources and Further Reading

For authoritative data and contract specifics, verify on NSE India for daily advance-decline data and lot sizes, Zerodha Varsity for breadth and charges, and Investopedia for the original McClellan methodology. The Nifty price levels cited for 2008 and 2020 are historical fact and verifiable on NSE; the McClellan Summation Index values are illustrative reconstructions because no exchange publishes an official Indian MSI. Always confirm current rules, tax rates and STT before you trade, and treat every number here as illustrative and not a promise of returns.

Sources and Further Reading

For authoritative data and further reading on this topic, refer to NSE India, Zerodha Varsity and Investopedia. Always confirm current rules, rates and contract specifications on the official source before you trade.