Getting started with order flow

Price is a result, not a cause

A candle is a report. It tells you that between nine o'clock and five past, price opened here, closed there, and touched those two extremes along the way. The account is accurate, and it is silent about the one thing you actually care about: what had to be spent to obtain that move.

Take two twenty-point rises on the same instrument, an hour apart. The first happened because almost nobody was on the other side: a handful of orders swept the available sellers and price slid as if on glass. The second happened against a queue of sellers reloading at every level: ten times as much had to be consumed to travel exactly the same distance.

On your chart those two rises are the same candle. In the market they are nowhere near the same event. The first was obtained through a void, and a void closes as easily as it opened. The second cost real substance, and that substance will not be found in the same place again.

Order flow is exactly this: the layer under the chart, the one where you see the spending and not only the outcome. Price goes back to being what it always was, the consequence of a sequence of transactions, rather than something that moves of its own accord for reasons of its own.

That difference leaves a trace, and the trace is what makes it usable. The sellers missing from the first rise sold nothing: they are still up there, with the same orders left to place. The sellers of the second rise were filled, their inventory is gone, and a price coming back for them will find a thinner book than before. A price chart knows neither of those two states, although they decide what happens on the next visit.

The single question it answers

Order flow does not answer “where is price going”. It answers one question, one only, and that question has the merit of being measurable: how much movement do you get for how much volume traded?

The idea is not new. Richard D. Wyckoff, a New York operator and publisher active from the 1910s to the 1930s, made it one of his principles under the name the law of effort and result. Effort is the volume traded. Result is the distance price covered. When the two agree, the move is coherent. When they diverge, something is going on that the price line does not show: a great deal of effort for very little result means somebody on the other side is filling everything that arrives without giving an inch.

The ratio can be worked out by hand. Take the number of contracts or shares traded during a move, divide by the number of points covered, and you have a cost of travel. Compare it with the earlier moves of the same session. You are predicting nothing, you are recording a fact written nowhere on the price chart, and one that most of the people staring at the same screen as you have never looked at.

⚠️ A ratio is never read on its own. Enormous volume for three points can mean a large seller is absorbing everything, or simply that the session has just opened and the whole world is trading at once. Order flow is always read relatively: same time of session, same instrument, same regime. An absolute volume figure means nothing on its own, ever.

An order that waits, an order that takes

Here is the distinction that carries the entire field, and it is far simpler than the vocabulary it is dressed in. There are two ways to get into a market, only two, and everything else follows from them. The first is to display your price and wait for somebody to come and fill you. You say: I am a buyer of ten at a thousand. Your order joins a queue at that price, behind everyone who arrived before you. You move nothing, you offer. This is a passive order, also called a limit order.

The second is to take what is already displayed, right now, at whatever price it costs. You say: buy me ten, immediately. Your order crosses the queue and consumes what it finds there, level by level, until it is filled. This is an aggressive order, also called a market order.

One supplies liquidity, the other removes it. The first is paid in patience: it gets a better price, it is never certain of being filled, and it can sit in its queue all day. The second is paid in certainty: it is filled at once, it pays the gap between the two best displayed prices, and more than that if its size exceeds what was waiting at the first level.

Hold on to the exact word: it is aggression that moves the traded price. A passive order can be gigantic, and no trade will print higher until somebody comes to get it.

⚠️ One precision that heads off a common misreading, because the short version above is almost always taught without it. The quoted price does move with no trade at all: it is enough for an order to be added inside the spread, or for the orders holding the best level to be pulled. A book emptying by cancellation slides the quoted price on zero volume, and that is exactly the kind of move the measurement in the next section will never see, since nothing changed hands.

Signed volume, the one real measurement

Since price travels towards the side that aggresses, counting volume without saying who took it throws away half the information. Hence the measurement every tool in the next section is built on: signed volume, credited not to a buyer and a seller, both of whom are always present, but to the one side that came to take.

The convention is mechanical and calls for no judgement. A trade done at the best offer, the price you have to pay to be filled straight away as a buyer, goes to the account of impatient buyers. A trade done at the best bid goes to the account of impatient sellers. The difference between the two totals carries a name you will meet everywhere, the delta, and its running total across the session is the cumulative delta.

What that changes shows up straight away. Two candles displaying the same total volume, so the same bar under the chart, can carry opposite deltas: in one, buy aggression took four fifths of everything that traded, in the other, sell aggression did. No candlestick chart tells them apart, because it adds the two sides together before showing you the result.

⚠️ A delta is not a signal, and reading it as one is the most expensive mistake in the field. A heavy buy delta alongside a price that fails to rise usually says the opposite of what a beginner reads into it: plenty of people bought, somebody filled every one of them without flinching, and that somebody has not shifted an inch. A delta is always read next to the movement it actually produced.

The understanding order flow lesson picks this thread up and runs it out, showing why the sentence “price rises because there are more buyers than sellers” is arithmetically impossible. It is wrong for the reason described here, and it is worth having that in mind before you get there.

The four tools, named once and for all

The field comes with a small kit that is almost always handed to you in a heap, with busy screenshots and a lot of colour. There are four objects, and they answer four different questions. Knowing which is which is enough to untangle most of what you will read elsewhere.

The order book, also called market depth or the ladder, displays the passive orders resting at each price, above and below the last trade. It answers: what is waiting right now? ⚠️ It is the most deceptive of the four, so much so that the understanding order flow lesson draws its reading rule from it: a book is judged on what fills in it, never on what is displayed in it.

The tape, or time and sales, lists the trades that actually happened, one by one, with the time, the size and the side that aggressed. It answers: what has filled? It is the oldest of the four and the only one that already existed in 1900: reading the ticker tape is the craft described by Edwin Lefèvre in Reminiscences of a Stock Operator, published in 1923.

The footprint takes each candle and opens it up. Instead of a body and two wicks you see, for every price crossed, how much traded on buy aggression and how much on sell aggression. It answers: where inside the candle did the volume happen, and on which side? It is the densest of the four in information, and the longest to learn to read.

The volume profile stacks volume by price over a whole period, a session or a week, and draws a horizontal silhouette. It answers: where did the market accept to trade, and where did it pass through without stopping? Its modern form descends from the market profile devised by J. Peter Steidlmayer at the Chicago Board of Trade in 1984, which counted time spent at each price rather than volume.

Of those four, the last is by far the most reachable for somebody starting out. It is read on a long timeframe, it demands no continuous attention, it comes free with most charting platforms, and it is the one the auction market theory lesson leans on most.

Which markets carry a real measurement

Everything above assumes one thing, and that thing is not a given everywhere: that the volume shown on your screen is volume genuinely traded. Nobody will point it out while selling you a course, and it is nevertheless the first question to ask. The condition is fully met on futures. A given contract trades on one venue, in one book, kept by the exchange that lists it. Every transaction is recorded once, with its size and its price, then broadcast to everyone at the same instant. The number you see is the number that dealt, for everyone watching.

It is met differently on listed shares, and the nuance is worth knowing because nobody volunteers it. A share has not traded in a single book for the best part of twenty years: it trades in parallel across competing venues, plus a portion done away from any venue at all. The book your broker shows you is one venue's book, not the market's. What rescues the measurement is the obligation to report every execution to a consolidated feed: the total volume is real and complete, reaching you with a slight delay on the part dealt off-venue. A volume profile on a share is therefore sound; an order book on a share shows one slice of the market.

The condition is not met on the spot currency market. There is no single book: there is a network of banks, platforms and brokers each dealing on their own side. Your broker only sees what passes through its own hands. The “volume” your chart displays is most often a count of price changes per interval, known as tick volume. It is a measure of activity, correlated with real volume, and it is not a volume.

The consequence is blunt, and it deserves to be written without softening. On spot currencies, a footprint or a volume profile gives you a plausible picture, never a measurement. You may use it as a relative activity indicator. Building an absorption read on top of it means confirming one thing with another thing nobody measured.

Crypto sits in between. Each exchange keeps a real book and publishes a real volume, but the whole is fragmented across dozens of venues. The volume of any single one is authentic and partial at the same time, which is workable provided you know which venue you are looking at and what share of the market it carries.

The learning cost, and the attention cost

Two very different costs hide behind the phrase “it is hard”, and the one people underestimate is not the one they expect.

The learning cost is real and unremarkable. Reading a footprint takes weeks of note-taking before the shapes start meaning anything, exactly like any other reading skill. It is paid in hours, it can be planned, and it depends on nobody but you. Of the two costs, it is the harmless one.

The attention cost is structural, which means no amount of effort will bring it down. Order flow is read live, on the scale of the second and the minute. A footprint reviewed in the evening has lost most of what it was worth, because what it showed was an intention in the act of being executed. That means sitting in front of the screen, available and fresh, during the hours the market is active.

⛔ If you trade alongside a job, say it to yourself honestly before sinking six months into this. A method demanding continuous attention during the hours you are paid to do something else is not a difficult method, it is a method you will not be able to apply. This is not a question of discipline or willpower, it is a question of what your working week actually allows, and no course will fix that for you.

That observation does not cancel the lesson, it moves what you should take from it. The order flow lens is useful to everyone, including someone who will never look at an order book in their life. The live exercise is not, and it is better to know that before than after.

What order flow does not replace

It does not replace the regime. Knowing that a candle was absorbed does not tell you whether you are in a trend or a range, and an absorption read against the regime is an absorption that costs you money with the comfortable feeling of having read well.

It does not replace zones. Flow tells you what is happening where price is; it does not tell you where to wait for it. Those are two separate questions, and the second is settled with structure and raw price, not with a book.

Above all it does not replace risk management. No amount of order book reading has ever taken a single pound off the loss on an oversized position. A trader with no plan who adds order flow does not obtain a plan: they obtain an absent plan, with more screens lit and far more reasons to click.

Its place is precise, and it is small: it refines an entry inside a plan that already exists. It turns “I am waiting for a reaction in this zone” into “I am waiting for a reaction in this zone, and I want it to cost real money to whoever is pushing the other way”. The zone itself was decided earlier, elsewhere, and without a book.

What you can take from it tonight, for free

Three things, in this order, and not one of them needs a subscription, specialist software or an extra screen.

First, put volume under your chart on an instrument whose volume is real, a future or a listed share. It is already there, free, on every platform. Spend a week looking at nothing but the relationship between the size of the candle and the height of its volume bar. Write down the three biggest mismatches of the week in each direction: large candle with small volume, and the reverse. Then go and see what price did in the hour that followed each one.

Second, work out three costs of travel within one session. Pick the three clearest moves of the day, note the total volume traded and the number of points covered for each, and put the division in a spreadsheet. Three numbers, ten minutes. None of the three means anything on its own; the gap between them is what speaks: you have just measured that the same instrument did not cost the same to move, depending on the hour and on the opposition it met. What a volume profile makes of those same numbers is the subject of the auction market theory lesson.

Third, get into the habit of asking the aggression question out loud. Faced with any move at all, ask yourself: did somebody pay to get there, or did price slide for lack of opposition? You will not always be able to answer, and that does not matter in the slightest. Asking changes what you see, because it forces you to look for a cause instead of commenting on a result.

After a fortnight those three habits have built something you did not have: a list of a dozen or so moves whose cost you know, on your instrument, at your hour of the session. Keep it next to your trading journal. That is your own scale of reference, and it, rather than any course, is what makes the next day's volume readable: it depends on the instrument and the hour that you, specifically, are watching, and nobody can sell it to you ready-made.