Market Breadth Indicators, and Why Forex Has None of Them

Market breadth is a count of how many members of a defined set of stocks rose and how many fell. The set has to exist before the count can, and that is the whole of the problem for anyone who trades currencies: a currency pair has no members. There is nothing to tally.

That does not make the term irrelevant to a trading account. Breadth readings move index prices, and an index CFD is a product plenty of currency traders hold. It means something narrower and more useful: every breadth number met in market commentary is a statement about an equity market, computed over a list somebody else maintains, and the list decides what the number can be compared with.

What Market Breadth Actually Counts

A breadth reading needs three things before it can be computed. A defined set of instruments. A rule that sorts each member of that set into a bucket for the session. And a period over which the buckets are added up.

The buckets are usually advances, declines and unchanged, measured against the previous closing price. Nasdaq publishes all three as separate daily fields for Nasdaq-listed securities, and the third one is real: a security that closes exactly where it closed yesterday is neither an advance nor a decline.

Most commentary drops it. That matters more than it sounds, because a ratio of advances to declines and a percentage of the set that advanced are different quantities whenever the unchanged bucket is not empty, and it is rarely empty on a listing that includes thinly traded issues.

Market breadth is therefore a counting statistic, not a price statistic. It says nothing about how far anything moved. A set in which every member gained a fraction of a per cent and a set in which every member gained ten per cent produce an identical breadth figure.

Index Constituents Against a Whole Exchange Listing

The single largest source of confusion around the term is that two unrelated sets are described with the same word, and the sentence quoting a reading often does not say which one was used.

The first is index breadth. The set is the membership of a named index, and the membership is a published, rule-bound list. The Nasdaq-100 takes the hundred biggest non-financial companies with a Nasdaq listing, so a breadth reading over it excludes every bank and insurer by construction, whatever the banks did that day.

The second is exchange breadth. The set is everything that traded on a venue, which includes securities belonging to no index at all, and includes issues that a reader picturing well-known companies would not recognise. The Nasdaq daily fields are of this kind.

These disagree for a structural reason rather than an occasional one. An exchange listing is numerically dominated by small and thinly traded issues, while an index membership is a size-filtered subset. On a day when large companies rise and small ones fall, index breadth is positive and exchange breadth is negative, and both are correct measurements of different populations.

So the first question to ask of any breadth number is not what it says. It is which set it was computed over.

Why No Breadth Reading Exists for a Currency Pair

A currency pair is a price of one currency expressed in another. It has no members, no constituents and no components. Nothing about it can be counted the way index members are counted, and this is not a data-availability problem that a better provider would solve.

There is no listing to fall back on either. The Bank for International Settlements describes foreign exchange as an over-the-counter market, and its April 2025 turnover survey collected reports from more than 1,100 dealing institutions in 52 jurisdictions rather than from an exchange tape. There is no single venue whose issues could be tallied at the close, because there is no close and no venue.

A substitute is sometimes offered: count how many pairs involving one currency rose today. That is a real number and it is not breadth.

Every pair in that count shares a leg, so the members are correlated by construction, and a result of eight up and one down is close to a restatement of the single fact that the shared currency strengthened. Reading currencies against each other is a legitimate exercise, and it belongs with intermarket analysis rather than with a participation count.

The Common Readings, and What Each One Samples

Four readings account for most of what circulates. They are not variations on one number, because they sample differently and respond to different things.

Two of them are cumulative. An advance-decline line adds the net of advances over declines to a running total each session, so its level carries the entire history of whatever set it was computed over. A ratio and a percentage are not cumulative and describe one session only.

That distinction decides comparability. A cumulative series is only comparable with its own past if the set underneath it did not change, and index memberships change on schedule. A single-session ratio has no such exposure.

The fourth column is the one that gets left out of published readings, and it is the one that decides whether two figures can be placed side by side.

What a Divergence Between Price and Breadth Does and Does Not Say

A capitalisation-weighted index can rise on a day when most of its members fall. That is arithmetic rather than a market anomaly: weights are unequal, so a handful of large members can outweigh a majority of small ones.

A breadth divergence records exactly that and nothing more. The index level and the typical member disagreed over the period measured. It is a description of participation, and the honest reading stops there.

It carries no timing claim. Whether a participation reading tends to move before price or after it is the general argument about leading and lagging indicators, which is settled the same way for every indicator and is covered on that page rather than this one.

The point specific to breadth is narrower. A divergence observed on two dates is only the same measurement on both if the set was the same on both, and for an index that is a claim about the rebuild calendar. Across a reconstitution, the earlier and later readings were taken over different populations.

Where the Data Comes From, and Who Calculates It

The index provider maintains the membership and calculates the index level. It does not publish breadth. FTSE Russell publishes the rules by which the Russell indexes are built and rebuilt; Nasdaq publishes the rules for the Nasdaq-100. Neither publishes a daily count of how many members closed above an average.

That statistic is assembled by whoever collects the constituent prices, which is a data vendor. Two vendors can produce different numbers for the same index on the same day, and the differences come from choices none of them are obliged to disclose.

Share classes are one. FTSE Russell reviews each share class of a company independently for inclusion, subject to size, liquidity and float thresholds, so a company can appear in the set more than once, and a vendor may collapse those into one member or leave them separate.

Rebuild dates are another. Russell rebuilds land on two named dates each year, the mid-year one on the fourth Friday of June and the year-end one on the second Friday of December, with new listings folded in quarterly, and a vendor may compute history against current membership or against the membership in force at the time.

The methodology documents also count in two units. The Russell text describes the broadest US index as holding the largest 4,000 US companies in one section and the largest 4,000 securities in another, and states that if fewer than 4,000 eligible securities exist, the entire eligible set is included. The size of the denominator is a target, not a guarantee.

What an Index CFD Trader Can Actually See

An index CFD is a single symbol. The platform record behind it holds quote and contract information: bid, ask, trading session, contract size, margin terms. The MQL5 symbol property documentation lists what a symbol carries, and constituent membership is not among the fields, because a symbol is a tradable instrument rather than a basket the terminal can open.

So a trader holding a stock index position has the level without the ingredients. Breadth cannot be derived inside the terminal at all. Obtaining it means a third-party data source, and the first thing to establish about that source is which set it samples.

This puts breadth in the same category as several other readings a CFD account cannot generate for itself, including exchange-level liquidity readings and the volatility index, all of which are computed elsewhere and consumed as a quoted number.

Who This Is Not For

An account holding currency pairs and nothing else gains one thing here, which is the ability to discard the term on sight when it appears in commentary written for equity readers. No version of the measurement applies to what is on the screen.

Anyone looking for a threshold will not find one on this page either. Published thresholds circulate widely without the sample they were computed over, and a threshold without a sample cannot be checked or reproduced.

And anyone wanting a signal is asking the reading for something it does not contain. Breadth reports how many members participated. Deciding what that implies is a separate question, and one this page does not answer.

Frequently Asked Questions

What does market breadth measure?

It measures participation: how many members of a defined set of stocks rose, how many fell and how many closed unchanged over a stated period. It is a count of members, not a measure of how far prices moved, so a set that rose slightly and a set that rose sharply can produce the same reading.

Can market breadth be calculated for a currency pair?

No. A currency pair is a price of one currency in another and has no constituents to count. There is also no exchange listing to count over, since foreign exchange trades over the counter rather than on a single venue, so neither form of the measurement has an object to be computed on.

What is a breadth divergence?

It is the situation where an index level and the typical member move in different directions over the same period, which is possible because index weights are unequal and a few large members can outweigh many small ones. It records that participation was narrow, and it carries no claim about what happens next.

Which breadth reading is quoted most often?

The advance-decline line, which adds the net of advances over declines to a running total each session. Because it is cumulative, its level reflects the whole history of the set it was computed over, and it is only comparable with its own past if that set did not change in between.

Where does a retail platform get breadth data?

From a third-party data vendor rather than from the index provider, which publishes membership rules and the index level but not breadth statistics. A trading platform showing an index CFD carries the quote and the contract terms for that one symbol and no constituent feed at all.