An index can rise while most of its members fall. When that happens for long enough, analysts start using the word breadth, usually with an ominous tone. The measurement is real and worth understanding. The predictive claims attached to it are considerably weaker than the confidence with which they are made.
This article is for informational and educational purposes only. It is not financial advice, investment advice, or a recommendation to buy, sell, or hold any security, cryptocurrency, or financial product. Always verify data with official sources before making financial decisions.
What breadth measures, in one paragraph
Short answer: Breadth measures how many stocks are participating in a move, as opposed to how far the index travelled. A market where 400 of 500 members rise has broad participation. A market where the index rises the same amount because eight megacaps rallied while 380 members fell has narrow participation. Cap-weighted indexes report identical returns in both cases, which is the entire reason breadth indicators exist.
The three standard measures
| Indicator | Construction | What it shows |
|---|---|---|
| Advance-decline line | Running cumulative total of advancing minus declining issues | Whether participation is expanding or contracting |
| New highs minus new lows | Count of 52-week highs less 52-week lows | Whether leadership is broadening or deteriorating |
| Percent above 200-day average | Share of members above their own long-term trend | How much of the index is in an uptrend |
These are not interchangeable. The advance-decline line responds to daily direction regardless of magnitude, so a market where most stocks fall fractionally while a few rise sharply looks bad on the A-D line and fine on the index. Percent above the 200-day is slower and describes trend rather than daily flow. New highs minus new lows is the most volatile and the most sensitive to the anniversary of a prior extreme — readings distort mechanically when a sharp move from a year earlier rolls out of the lookback.
The 2026 setup
The current configuration — indexes near records with thinning participation — is the one that generates the most commentary and the least reliable inference. Concentration in a handful of very large companies means the cap-weighted index reflects the fortunes of those companies more than the fortunes of the market.
That is a description of index construction rather than a warning. A cap-weighted index is designed to be dominated by its largest members; observing that it is dominated by its largest members is not a finding.
Why cap-weighted indexes hide the problem
The mechanism is arithmetic. In a float-adjusted cap-weighted index, a company representing 7% of the index contributes seven times the index impact of a company representing 1%, per unit of return. When the top ten members represent an unusually large share of total capitalisation, the index becomes a levered bet on those ten.
The cleanest way to see this is the ratio of an equal-weighted index to its cap-weighted equivalent. A falling ratio means the average member is losing to the largest members — the same information the advance-decline line carries, expressed as a price series you can chart directly.
Breadth thrusts vs breadth divergence
These two terms describe opposite conditions and are frequently confused because both are called breadth signals.
A breadth thrust is a sudden expansion in participation from a depressed base — a large share of members moving from below to above a threshold in a compressed period. It is a bullish construction and the well-known versions have small sample sizes measured in single digits over decades.
A breadth divergence is the slow decay described above: the index making new highs while the A-D line or the percent-above-200-day fails to confirm. It is the bearish construction, and it is the one with the poor timing record.
What narrow leadership has and has not preceded
The honest version of the historical record: breadth divergences have preceded some significant declines and have also persisted for extended periods without one. The signal has no reliable lead time, which makes it close to useless for timing even when it eventually proves directionally right.
There is also a structural argument that deserves weight. Index concentration has trended higher over decades as the largest companies became genuinely larger relative to the economy, and comparing today’s concentration to a mid-century baseline compares two different market structures. Some portion of what looks like a deteriorating breadth signal is a changed composition of the investable universe.
What this article does not conclude
Narrow breadth is a factual description of current market internals. It is not a forecast, it does not carry a timing signal, and the historical cases in which it preceded a decline are not numerous enough to establish a base rate.
Breadth data is published daily by the exchanges and is available from most charting providers. Where a specific reading matters, check the construction — providers differ on whether they include all listed issues or index members only, and the two produce noticeably different lines.