Market breadth
A headline market cap can rise on a day when most coins fell. Breadth is the answer to the question that number cannot settle: is the move broad, or is it a handful of large assets carrying a market that is otherwise going the other way?
Every figure here describes what price has already done. Nothing on this page is a forecast, and nothing on it is a recommendation to buy or sell anything.
The universe
Every reading covers the 100 largest assets by market cap that have an active Binance spot market, excluding stablecoins and tokenised funds. When an asset in that list leaves the exchange or drops out of the top 100, the list refills from rank 101 downward, so it always holds a hundred.
Every figure states how many assets it measured
This is the page where that matters most, because the readings on it genuinely describe different numbers of assets, and all of them are correct:
- The advance/decline split and the equal- vs cap-weight pair are taken over a 24-hour price change that exists for every asset in the universe.
- Above the 20-day EMA needs twenty days of closes.
- Above the 200-day EMA needs two hundred. An asset listed six months ago has an answer for the first and no answer at all for the last.
So "50 of 95" under one card and "91 of 99" under another are two true statements, not a contradiction. Each card names its own sample, and none of them borrows another's.
Where fewer than half the assets could be measured, the figure is withheld entirely and reads "—". It is not shown weakly and it is never filled in with a zero: a percentage drawn from a handful of survivors reads exactly like one drawn from the whole market, and nothing on the page could tell them apart.
Advancing vs declining
How many assets closed up over the window, how many closed down, and how many closed exactly flat.
The flat count is reported in its own right rather than folded into the decliners. A day on which nothing moved and a day on which everything fell are different market days, and a two-colour bar has to put the flat assets in one of the two colours. Here they get their own segment, and the count appears under the ratio whenever it is above zero.
The ratio is advancing divided by declining. On a day with nothing declining there is no ratio to state, and the page says so instead of printing a very large number.
Equal-weight vs cap-weight
The same 24-hour move, counted two ways.
- Equal-weighted treats every asset the same. It answers "what did the typical coin do today?"
- Cap-weighted weights each asset by its market cap. It answers "what did the money do today?" — and it is dominated by the largest few assets.
When cap-weighted is ahead, the move is concentrated in the big names. When equal-weighted is ahead, it is broad. The gap between them is the single clearest statement of breadth on the page, which is why they sit side by side rather than in separate panels.
EMA breadth
The share of the universe trading above its 20-, 50- and 200-day exponential moving average.
An exponential moving average is a rolling average of closing prices that weights recent days more heavily than older ones. An asset above its 50-day EMA has been closing higher lately than it averaged over the recent past; below it, lower.
Read across the three windows rather than at any one of them. A market where most assets are above their 20-day average but below their 200-day is one that has bounced inside a longer decline. A market above all three is a broad advance on every timescale.
The colour ladder on each card is a reading aid, not a classification:
| Share above | Reads as |
|---|---|
| 65% or more | Broad |
| 50–65% | Majority |
| 35–50% | Narrow |
| Under 35% | Weak |
These four steps exist so the three windows separate from one another at typical readings. They are a property of this page's presentation and are not used in any calculation.
The history line
The chart under the three cards is the share above the 50-day EMA on each of the last ninety days.
It is computed from the daily closes we already hold for every asset in the universe, so a past day's crossing is a fact about candles we have rather than an estimate of a figure nobody observed. Nothing on it is back-filled from a guess.
Two things follow, and the page states both:
- The line reaches back only as far as the shortest-listed asset in the universe allows, so the caption names the days it actually covers rather than implying ninety.
- A day we could not measure leaves a gap in the line, not a straight segment across it. An interpolated line would say the market sat still on a day we simply did not measure.
Sector breadth
The cap-weighted return of each sector over the selected period.
The period control governs this table and nothing else. Everything above it — the advance/decline split, the weighting pair, the three EMA readings — is a reading of right now, and changing the period does not change any of them. That is enforced by the server rather than by the page.
Each row states how many assets it covers, and a sector whose coverage falls below half over the selected period reads "—". This is common on the longer periods and it is not a fault: the 1h, 24h and 7d columns cover every asset in the universe, while 30d, 90d and YTD are computed from exchange candles and cover only the assets with that much history on the exchange. One table, two sources, two honest denominators.
Sectors are curated by hand from a fixed list, not fetched. An asset that has not been filed under a sector yet is in every reading above the table and in no row of it, so the footnote says how many of the hundred the table describes.
How fresh it is
The advance/decline split and the two weighted returns move with the market feed, which refreshes every ten minutes. The EMA readings, the history line and the sector table are recomputed hourly.
The page is dated by the older of the two, because that is how old the combination is. If either falls behind, a banner says when the figures were last computed — the readings stay on screen, clearly dated, rather than disappearing.