Footprint Chart: The Bar Taken Apart
A footprint chart expands each price bar into rows, one for every traded price, showing how much volume executed there and splitting it by which side crossed the spread. It is the volume of a single bar taken apart, and it is an inference about urgency rather than a record of who bought.
Covered on this page: TradingView.
How it works
A footprint chart is an ordinary bar with its inside shown. Instead of four numbers — open, high, low and close — the bar becomes a column of rows, one per traded price, each carrying the volume that executed there.
The rows are the whole idea. A candle gives the range a bar covered; the footprint gives where inside that range the trading happened, and how much at each price.
Each row is then split in two. Trades executed at the offer are counted as buying and trades executed at the bid as selling, so every cell holds a pair rather than one number.
That split is an inference, and worth saying plainly. Every trade has a buyer and a seller. The classification records which side crossed the bid-ask spread, not who bought.
Delta, imbalance and absorption
Delta is the difference between the two counts. Subtract selling from buying in a row for that row’s delta, or across the bar for the bar’s. Cumulative delta sums it across a session.
An imbalance is a row that is badly lopsided. One side substantially exceeds the other, usually measured against the adjacent price on the opposite side.
Every threshold for one is relative. What counts as lopsided depends on the instrument and the hour, so a number copied from elsewhere marks everything or nothing.
Heavy buying with no progress means somebody is selling. A row with far more executed buying than selling, at a price that did not move, says resting sell orders absorbed it. That is the clearest reading the chart offers, and still an inference.
It is also the reading most often applied afterwards. Once price has turned, the bar where it turned contains one; finding it there is not the skill.
In practice
Running one requires tick-level data with side classification. That is a paid feed on most instruments, plus a platform that renders it.
It is primarily a futures tool. Those markets run one consolidated tape, so every execution is counted once.
What you are looking at is the volume bar taken apart. Volume analysis gives how much traded; this gives where.
None of the detail survives to a higher timeframe. On a daily bar each row averages a session, and the sequence that produced it is gone.
And there is nothing to count across a gap. An opening gap is a stretch of prices at which nothing traded, so those rows do not exist.
It never tells you where the stop loss belongs. The grid describes what executed; a stop decides how wrong you are prepared to be.
Reading it quickly also means trading quickly. Say a scalp targets four ticks and the round trip costs one tick in spread and commission: a quarter of the move is gone before it starts — a toll on every move scalping tries to capture.
Against its neighbors, the difference is what gets counted. The order book, shown as level 2 or depth of market, lists intent that can be canceled.
The tape shows executions in sequence; the footprint shows the same ones organized by price. One is time order, the other is price order.
Learning to read one
Start by watching, not by trading. Pick one instrument, put its footprint beside its price, and follow it for several sessions before drawing a conclusion from it. The first job is calibration.
You are learning what normal looks like on that instrument. A row that is wildly lopsided in one market is unremarkable in another, and the same market changes character between the open, the middle of the day and the close.
Note where the heavy rows sit, then ask whether the price held. Write down what you saw. After a few sessions the answer stops being theoretical, because you have your own baseline instead of somebody else’s threshold.
Only then is it worth putting on a live decision. Reading a grid you have no reference for produces confident sentences about nothing.
What a footprint chart is not
- Not a record of who bought. Every trade has both sides; the split records who crossed.
- Not an order book. It counts completed executions, which cannot be canceled.
- Not a signal. A lopsided row describes the last few seconds, not the next few.
- Not a higher-timeframe tool. Above a few minutes the rows blur into averages.
How this connects
A footprint shows buying and selling volume at each price inside a bar, which makes it the detailed view of the same data cumulative delta summarizes.
It is order flow, so it needs tick data with quotes attached - not something a standard candle feed can produce.
Volume profile is the close relative. Both organize volume by price rather than by time; the footprint adds which side was aggressive.
The tape is the raw version underneath both - individual prints, unaggregated.
And the classification is inferred, not reported. No feed states who initiated a trade; it is deduced from where the print landed relative to the quote, and that deduction can be wrong.
When it fails
In a quiet market the cells are too empty to read. Volume has to be spread across enough prices for the rows to differ. Inside a slow trading range each cell holds a handful of trades and the reader supplies the pattern.
- When the classification is backwards. A resting buyer filled at the bid — often market makers doing their job — is counted as selling, because the seller crossed.
- When size is being worked. An algorithm slicing a parent order produces steady two-sided rows that look like broad participation, by design.
- When the market gaps. Prices that never traded have no rows, so the grid is silent across exactly the move that mattered.
- When the reading arrives late. Absorption is obvious at the bar where price turned, ambiguous at every bar that looked the same and did not.
- When you are already positioned. Every imbalance then reads as confirmation, and a grid this detailed will confirm almost anything.
The original data
Seventeen videos carry “footprint” in the title; 228 carry “volume profile”. The scan of 24,971 titles in the finance search study this site keeps, logged in this site’s search study, puts their medians at 22,330 and 13,034 views. Two magnifications of one dataset, and the coarser one — the one that photographs, rather than the one you have to watch live — is covered thirteen times more often.
Why does the cost matter more here than elsewhere? A tool used to trade small moves has to clear a spread and commission that are a large fraction of them. Use it to judge whether a level is holding, not to predict which way the next bar goes.
Related
Order flow is the subject this page sits inside — 110 videos at a median of 22,195 views, and the footprint is one way of drawing it.
The tape is the same executions in time order rather than price order, which is why the two are usually read side by side.
Volume profile is the coarse version of the same count, stacked across a session rather than a bar, and where the value area comes from.
More resolution is not always more information. Past a certain point, extra detail tells you more about your own attention than about the market, and you can read a footprint and still be no clearer about whether to take the trade. Use it to check a level you have already chosen, rather than to go looking for one.
— Michael Whitman
This page is educational, not financial advice. Test every idea on your own charts before risking money. Some links on this page earn a commission if you buy through them. It costs you nothing and it does not decide what appears here or in what order — how these pages are made is set out in our methodology.