Volume & breadth analysis · a standing reference
Volume moves first.
Price follows.
Large institutional orders cannot be executed quietly. They leave their trace in the volume long before the trend gives way, and that trace is measurable. This is the reference for measuring it.
- 42 indicators, formula and failure modes
- 13 index constituent lists
- 3 calculators that run in the page
Three questions volume answers that price cannot
A price chart records where a market ended up. It says nothing about how hard that was: whether the move absorbed everything offered against it or drifted through an empty book on a quiet afternoon. Volume is the record of effort, and reading it turns on three questions that recur on every page of this reference.
Was there enough participation for the move to mean anything?
A breakout on half the usual volume and the same breakout on triple it are different events wearing the same shape. The first is a handful of orders finding no resistance; the second required someone large to be on the other side, repeatedly. This is the first filter, and it discards more signals than any indicator setting.
Who was doing the trading, and can they hide it?
A position that takes weeks to accumulate cannot be hidden, because the shares have to come from somewhere. The trace shows up as sustained above-average volume on days that look unremarkable on price alone, and again as the surge that appears when the same position is unloaded. The volume spread analysis section reads this directly from the bar; the indicator library measures it with numbers.
Is the whole market moving, or a handful of large names?
An index can rise while most of its constituents fall, carried by a few heavyweights. Price cannot show you that; breadth can. The advance/decline ratio, the new-high and new-low counts and the McClellan oscillator all answer the same question, how many issues actually participated, and a market that keeps rising on narrowing participation is the most reliably documented warning in this field.
None of these is a signal to trade. They are the conditions under which the tools below are worth consulting at all, which is why every indicator page here ends with the situations in which that indicator misleads.
Start with a number
The three tools below take the inputs you already have on a chart and return the levels everything else on this site refers back to. Nothing to install, nothing to sign up for.
| Tool | You supply | It returns |
|---|---|---|
| Pivot point calculator | High · Low · Close | R3–S3 levels, classic & Woodie |
| RSI calculator | Closing prices | Wilder RSI, any period |
| Fibonacci retracement | Swing high · Swing low | 23.6 / 38.2 / 50 / 61.8 / 78.6 % |
How this reference is organised
Five sections, divided by what the measures in them read rather than by when they were invented. The division matters in use: two measures that share an input will mostly agree with each other whatever the market is doing, and it is only when measures reading different inputs disagree that anything has been learned.
| Section | Reads | Answers |
|---|---|---|
| Indicator library | Price and volume, one instrument | How fast, how smooth, how heavily traded, with each formula and its failure modes. Start at RSI, the one most often read wrongly. |
| Market breadth | Counts of issues, whole market | How many took part, which no weighted index can express. The crash record is where the difference showed most. |
| Calculators | Your own inputs | Levels and readings with every intermediate step shown. |
| Indexes and exchanges | Index rules and listings | What a list contains and when it changes, a prerequisite for reading breadth. |
| Glossary | The vocabulary these pages assume | Definitions in the sense this reference uses them, not a general dictionary. |
What this site does not do
There is nothing to subscribe to. No signals, no alerts, no model portfolio, no newsletter, and no claim that any measure here forecasts anything on its own. The pages state what a number is calculated from, what it makes visible, and the situations in which the number is accurate while the conclusion drawn from it is wrong.
That last section is the one most references leave out, and it is the reason these exist. An indicator is a compression: it takes a history of prices or volumes and reduces it to one number per bar, and every reduction discards something. Knowing what a given measure cannot see is more useful in practice than another paragraph on how to read a crossing, and it is the only part of the description that costs money to learn the hard way.
Two consequences run through the whole site. Volume is compared against an instrument’s own history rather than against another instrument, because share counts carry the price level with them. And every chart is computed from data in the page’s own source, so the awkward sessions (the ones where the indicator did something the caption would rather it had not) stay in the picture.
Questions worth settling first
Why read volume before price?
Price tells you where the market went; volume tells you how much conviction it took to get there. The same one-percent move on triple the usual volume and on half of it are two different events, the first involves institutional size, the second is noise. That asymmetry is why a volume surge is read as a warning about the current trend rather than a confirmation of it.
Does a volume spike always mean a reversal?
No. A surge marks unusual participation, not a direction. It resolves into a reversal when it appears late in an extended move and the next sessions cannot make further progress, and into a continuation when it appears as a trend begins, a breakout on heavy volume is the same signal pointing the other way. The distinction is context, which is why every indicator page here also states where the indicator fails.
What is the difference between volume and market breadth?
Volume counts shares traded. Breadth counts how many issues participated, advancing versus declining stocks, new highs versus new lows. An index can rise on falling breadth when a handful of large constituents carry it, and that divergence is invisible in the index price alone.
Do these indicators work on intraday charts?
Most do, with one caveat that runs through every page here: raw intraday volume is distorted by the open and the close, when a disproportionate share of the session trades in a few minutes. Comparing a 10:15 bar against a 15:55 bar without adjusting for that shape produces signals that are artefacts of the clock rather than of participation.
Is volume comparable between two instruments?
No, and this is the most consistently ignored limitation in the whole field. Volume is counted in shares, so a two-dollar stock generates roughly fifty times the share count of a hundred-dollar stock for the same money committed, which means any ranking by share volume is substantially a ranking by low share price. Every comparison worth making is against an instrument’s own recent history, or made in currency rather than shares.
Does this site publish signals or recommendations?
No. There is no newsletter, no alert service, no model portfolio and nothing to subscribe to. What is here is a reference: what each measure is calculated from, what it can and cannot show, and where it misleads. That last part is the reason the pages are worth reading: an indicator described without its failure modes is being sold rather than explained.
Where do the numbers in the charts come from?
Every figure is computed at build time from a series held in the page’s own source, so the chart and the text it sits beside cannot contradict each other. Nothing is drawn by hand to demonstrate a point. One page goes further and uses real published market data — FINRA’s daily short-volume files — and it states its date range rather than pretending to be a live feed.
How much does any of this actually predict?
Nothing, on its own, and a reference that claims otherwise should be read with suspicion. Every measure here summarises what has already happened in order to make one property visible: how fast a move was, how broad, how heavily traded. Forecasts are separate claims laid on top of those summaries, and each has to earn its keep on its own evidence rather than inheriting authority from the arithmetic underneath it.