Why Volume? Price Is the Result

≈32 min read · Real gold futures examples with XAUUSD equivalents · Free

In this chapter: You will learn what volume-based trading adds to ordinary price-chart trading, and what it does not add. We follow the chain from decisions to orders to trades to price; see why moving averages, RSI and MACD always lag; treat the market as a two-sided auction; read the order book; and learn why every trade has one buyer and one seller, of whom only one was in a hurry. Then we compare a pattern trader and a volume trader on the same gold chart, clear away popular myths, separate real exchange volume from "tick volume", and map the platforms and data feeds. The theme in one sentence: candles show WHAT happened; volume shows WHO, WHERE and HOW HARD.

---

Price Is the Result, Volume Is the Cause

Most people meet markets through a price chart. A line or a row of candles wiggles up and down, and it is natural to think of price as something that moves by itself, like the weather. That picture is wrong, and fixing it is the first step towards understanding volume.

Every price on your chart is a trade

On a futures exchange, the number your chart plots is the last traded price, often just called the last. It is the price at which the most recent trade took place. A trade happens only when two orders match: someone willing to buy at a price meets someone willing to sell at that same price. If nobody trades, the last price does not change, no matter how many people are watching, hoping or posting opinions on social media.

So when you see gold futures (GC) print 4000.1 after 4000.0, what you are actually seeing is this: at least one contract changed hands at 4000.1. The chart did not "decide" to go up. Two parties agreed on a price, and the exchange recorded it.

Gold note: Throughout this chapter we use round GC prices near 4,000 for the examples. Spot gold CFDs (XAUUSD) usually trade a little below the futures price, because a futures contract includes a "carry" cost until delivery. In recent months the gap has been roughly $27–30, so GC 4000.0 corresponds to approximately 3,970–3,973 on XAUUSD. The gap changes over time and from broker to broker, so always treat the equivalent as approximate.

The chain: decisions → orders → trades → price

Think of what happens behind every printed price as a chain of four links:

1. Decisions. People and algorithms decide to buy or sell. These decisions come from many places: a central-bank statement, a fund rebalancing, a jewellery manufacturer hedging, a trader's stop being hit, an algorithm following a rule.
2. Orders. Each decision becomes an order: an instruction sent to the exchange to buy or sell a number of contracts, either immediately or at a specific price.
3. Trades. When a buy order and a sell order meet at the same price, a trade happens. The sum of contracts traded is volume.
4. Price. The price of the latest trade becomes the new "last", and that is what your chart draws.

Candles and indicators live at the very end of this chain. They are summaries of the price link. Volume and order flow live one step earlier, at the trades link, and the order book (which you will meet later in this chapter) shows part of the orders link. Each step back along the chain brings you closer to the reasons price moved, although you never reach the decisions themselves.

Key idea: Price is the result of orders meeting. Volume, the record of those meetings, sits one step closer to the cause. That is the entire reason volume analysis exists.

Two ways to move up one tick

Here is a concrete example that shows why the trades link matters.

Imagine GC is quoted with buyers waiting at 4000.0 and sellers waiting at 4000.1. (In the next sections you will learn to call these the bid and the ask.) Suppose 60 contracts are offered for sale at 4000.1. Now the price "ticks up": the best price at which you can buy moves from 4000.1 to 4000.2. There are two very different ways this can happen.

Story A — the push. Impatient buyers arrive and buy everything on offer at 4000.1. Over a few seconds, aggressive buy orders totalling 200 contracts hit the market. The first 60 consume all the sellers at 4000.1, and the rest start filling at 4000.2. The offer at 4000.1 has been "eaten", so the best ask moves up. Price rose because buyers pushed it.

Story B — the pull. Nobody buys much at all. Instead, the sellers resting at 4000.1 change their minds. Perhaps a news headline makes them nervous, or an algorithm decides to step back. They cancel their sell orders. Now the cheapest seller is at 4000.2. A trader buys 5 contracts there, and the last price prints 4000.2. Price rose because sellers pulled away.

On a candle chart, both stories look identical: one tick up. But they mean different things. In Story A there were 200 contracts of aggressive buying; in Story B there were 5. Volume, and the split between aggressive buying and aggressive selling (called delta, which Chapter 3 covers in detail), tells these stories apart. The candle cannot.

Try it: Next time you watch a gold chart with a Time & Sales window open, find two moments where price moved by one tick. Note how many contracts traded in each move. You will often see that some ticks are "bought" with hundreds of contracts and others happen almost silently.

What "cause" does and does not mean

It is tempting to jump from "volume is closer to the cause" to "volume predicts the future." Be careful. There are three honest limits.

First, explaining is not predicting. Volume tells you what forced the latest move; it does not tell you what the next people will decide.

Second, part of the cause is invisible in volume. Cancelled orders, like the sellers in Story B, never become trades and leave no mark on a volume bar.

Third, the biggest decisions are made away from the chart. Rates, the dollar, geopolitics and fund flows drive gold over days and weeks; order flow shows how those decisions hit the market, not the decisions themselves.

In our own testing, order-flow signals on their own did not predict gold's direction once realistic costs were included. So this book treats volume as a window into the mechanism, not a crystal ball: more information and better questions.

Common mistake: "Volume always moves before price." This is a popular claim, not proven. Sometimes heavy volume comes first; often it arrives together with the move or after it.

Common mistake: "Price can only go up if people buy." As Story B shows, price can rise when sellers simply withdraw.

---

Why MA, RSI and MACD Lag (and What They Throw Away)

Most traders start with indicators: a moving average on the chart, RSI or MACD in a panel underneath. These tools are popular for good reasons, they are simple and they smooth out noise, but they have a built-in property that every volume trader should understand. They are mathematical transformations of past closing prices. That makes them late by design, and it means they ignore most of what happened in the market.

The moving average

A simple moving average (SMA) of length N is the average of the last N closing prices. A 20-period SMA on a 5-minute chart averages the closes of the last 20 five-minute candles, which is 100 minutes of history.

Here is a small worked example on gold. Suppose GC has closed at exactly 4000.0 for five candles in a row. Then it jumps to 4010.0 and stays there. Watch a 5-period SMA:

The price moved $10 in one candle (that is 100 ticks, worth $1,000 on one GC contract or $100 on one MGC). The average needed five candles to catch up fully. With a 20-period average it would need 20 candles. This delay is called lag, and it is not a flaw in a particular indicator: it is what averaging is. You cannot remove noise by averaging without also delaying the signal.

An exponential moving average (EMA) gives more weight to recent closes, so it reacts faster. But faster is not the same as instant. An EMA still blends the present with the past, so it still lags.

RSI in one formula

The Relative Strength Index (RSI) compares the average size of up-closes with the average size of down-closes over N candles (14 by default):

• RS = average gain ÷ average loss
• RSI = 100 − 100 ÷ (1 + RS)

Suppose over the last 14 five-minute candles, the average up-move in gold's close was $2.00 and the average down-move was $1.00. Then RS = 2.0 and RSI = 100 − 100 ÷ 3 ≈ 66.7. If the average gain grows to $3.00 with the same average loss, RS = 3 and RSI = 75.

Notice what the formula uses: closing prices only, and only how much each close changed from the previous one.

MACD

MACD (Moving Average Convergence Divergence) is the difference between a 12-period EMA and a 26-period EMA of the close. A 9-period EMA of the MACD line itself is drawn as the "signal line". MACD is sometimes described as a "leading" indicator. It is not: it is built entirely from moving averages, so it inherits their lag, twice.

What these indicators throw away

All three reshape one series, the closes; they create no new information. Picture a funnel: at the top are all the trades in a candle (every price, size, timestamp and aggressive side); one step down, the candle compresses them into open, high, low, close (OHLC); one more step, most indicators keep only the close. Each step throws information away. In particular, classic price indicators ignore:

1. Volume. A close reached with 10 contracts and a close reached with 10,000 contracts have exactly the same weight.
2. The path inside the candle. The open, high, low and the order in which they happened are dropped.
3. The aggressive side. Whether buyers were lifting offers or sellers were hitting bids is invisible.
4. Location. At which prices most of the trading took place inside the candle is invisible.

Volume tools put those four things back. That is the honest case for them: not that they are magic, but that they use data the classic indicators never look at.

Key idea: Price indicators are the smoothed past. Order flow is the raw detail of the present. Both are information about what has already happened.

None of this makes moving averages useless: a long average is a reasonable way to define "trend" or filter market regimes. Nor is order flow lag-free; your feed, platform and reaction time all add delay. The real difference is the type of information.

Common mistake: "RSI above 70 means price must come back down." This is a popular claim, not proven. In a strong trend, RSI can stay above 70 for a long time while price keeps rising.

Try it: On a 5-minute GC chart, find a sharp move during the New York morning. Mark the candle where the move began and the candle where a 20-period moving average first turned or crossed. Count the candles between them. Then look at the volume of the candle where the move began.

---

The Market Is a Two-Sided Auction

If price is the result of trades, what organises the trades? The most useful mental model comes from Auction Market Theory.

Where the idea comes from

In the 1980s J. Peter Steidlmayer, a Chicago Board of Trade trader, developed Market Profile, which organises trading activity by price instead of time. James Dalton, in Mind Over Markets and Markets in Profile, turned it into a broader framework now called Auction Market Theory. Chapter 7 covers it fully; here we need only its core.

The market's job is to facilitate trade

The central idea: the purpose of a market is to facilitate trade. Price moves in order to find the other side.

• When price looks cheap to many participants, buyers become eager. They keep raising the price they are willing to pay, and price rises until it finds enough sellers.
• When price looks expensive, sellers become eager, and price falls until it finds enough buyers.
• When both sides are content to trade within a range, a lot of business gets done there. That range is called value: the area the market currently accepts as fair.

A useful image is a pendulum swinging between "too cheap, buyers step in" and "too expensive, sellers step in". Most of the time it swings around the middle, where both sides agree.

Balance and imbalance

The market alternates between two states:

• Balance: price rotates around an accepted value area. On a chart this looks like a range. Lots of volume builds up at similar prices.
• Imbalance: one side is in control, and price travels directionally to look for a new value area. On a chart this looks like a trend.

A price that the market trades through quickly, with very little volume, suggests that the price was not accepted: few people wanted to do business there. A price where a great deal of volume builds up suggests acceptance: both sides were willing.

Why this matters for volume

Auction theory ties volume directly to the idea of acceptance:

• High volume at a price = both sides agreed to trade there.
• Low volume and fast movement = the market did not accept that price.

The tools in later chapters, such as the volume profile, the value area and the point of control (POC), come straight from this idea. When you read a volume profile in Chapter 5, you are reading where the auction found agreement.

Picture three days on a 30-minute GC chart. On the first day gold rotates between roughly 3990 and 4010 (about 3,960–3,983 on XAUUSD), and the volume piles up in a bell shape around 4000. That is balance. On the second day gold leaves that range and travels to 4045, with thin, stretched-out volume along the way. That is imbalance: the auction searching for a new value. On the third day it settles into a new range around 4040. The bulge of volume shows where agreement was found; the thin stretch shows prices the market passed through without interest.

Key idea: The market is not trying to go up or down. It is trying to find someone to trade with. Volume shows where it succeeded.

A framework, not a formula

Auction Market Theory is a descriptive framework: a way of organising what you see. It is not a tested law that produces numerical predictions. Some well-known rules attached to it have not been firmly established by careful statistical testing.

Common mistake: "The value area is 70% of volume because that is a law of the market." The 70% figure is a convention, chosen as a rough analogy to one standard deviation of a bell curve. It is a definition, not a law of nature.

Common mistake: "The 80% rule: if price enters the value area and stays for two periods, there is an 80% chance it crosses to the other side." This is a popular claim, not proven.

---

Bid, Ask and the Order Book

To understand who moves price, you need to see where the waiting orders sit. That is the order book.

Bids, asks and the spread

The order book is the list of all limit orders waiting to be filled, sorted by price. (A limit order is an order to buy or sell only at a specific price or better; the next section explains it fully.)

• A bid is a waiting order to buy. The best bid is the highest price anyone is currently willing to pay.
• An ask, also called an offer, is a waiting order to sell. The best ask is the lowest price anyone is currently willing to sell at.
• The spread is the difference: best ask − best bid.

In GC during normal, busy hours, the spread is usually one tick: $0.10 per ounce. For example, best bid 4000.0, best ask 4000.1. One tick is worth $10 on a GC contract (100 troy ounces × $0.10) and $1 on an MGC contract (10 ounces × $0.10).

The DOM: the order book as a ladder

Most order-flow platforms show the order book as a vertical price ladder, called the DOM (Depth of Market). Prices run down the middle. Waiting buy quantities sit on one side and waiting sell quantities on the other.

Here is a simplified GC ladder:

Reading it: the best bid is 4000.0 with 9 contracts waiting; the best ask is 4000.1 with 12 contracts waiting; the spread is one tick. Deeper in the book, 22 contracts are offered at 4000.3, 24 are bid at 3999.8, and so on. The quantity resting at each price is called depth.

Two names you will see:

• Level 1 data shows only the best bid and best ask (and their sizes), plus the last trade.
• Level 2, or market depth, shows several price levels on each side. CME's main market-data feed publishes aggregated depth for a number of levels on each side (typically ten for many products).

Worked example: what a market order does to the ladder

Using the ladder above, suppose a trader sends an order to buy 20 contracts immediately at the best available prices. What happens?

• The first 12 contracts fill at 4000.1, using up every seller there.
• The remaining 8 fill at the next price up, 4000.2, leaving 10 contracts still offered there.

The trader's average price is (12 × 4000.1 + 8 × 4000.2) ÷ 20 = 4000.14. If they had expected to pay 4000.1 for everything, the 0.04 difference per ounce is slippage. On 20 GC contracts that is 0.04 × 100 oz × 20 = $80. The new best ask is 4000.2, and the last trade printed at 4000.2.

This is the mechanism behind "push" in the previous section. Price moved because an aggressive order used up all the resting supply at one level.

Who gets filled first: price, then time

On CME Globex, most products, including gold futures, match orders by price-time priority, also called FIFO (first in, first out). Better prices are filled first; among orders at the same price, the one that arrived earlier is filled first. If you join a bid of 9 contracts at 4000.0 with your own order of 1, you stand behind those 9 in the queue.

Time & Sales: where trades are recorded

Every trade is recorded on Time & Sales, also called the tape: a running list of time, price and size. Most platforms colour each line according to where it printed relative to the bid and ask:

• A trade at the best ask usually means an aggressive buyer arrived and bought from a waiting seller. Traders say the buyer lifted the offer.
• A trade at the best bid usually means an aggressive seller arrived and sold to a waiting buyer. Traders say the seller hit the bid.

This simple rule is the foundation for classifying volume as "buying" or "selling" in Chapter 3.

What the DOM cannot tell you

The DOM is a snapshot of the current moment, not a promise.

• It shows only visible orders. An iceberg order shows a small part of its true size and refills automatically as it is filled, so the displayed quantity understates what is really there.
• Orders are added and cancelled constantly, often many times per second. A large quantity can vanish before price ever reaches it. Ordinary cancellation is perfectly legal and extremely common. (Deliberately placing orders you intend to cancel, in order to mislead others, is called spoofing and is illegal.)

Common mistake: "The price on my chart is the price I can buy at." The chart shows the last trade. If you buy immediately, you pay the ask; if you sell immediately, you receive the bid.

Common mistake: "The bid is the price I sell at, so it is my price." The bid is what other people are offering to pay. You can sell to them there with a market order.

Common mistake: "A big wall of orders in the DOM will hold price." This is a popular claim, not proven. Large resting orders can be cancelled or simply eaten through.

---

Market vs Limit Orders: Aggressor vs Passive

Now we can name the two roles that every trade involves.

The main order types

• A market order says: "Fill me now, at the best price available." It buys immediacy, and in return it pays the spread and accepts whatever slippage the book causes. A market buy matches against the best ask; a market sell matches against the best bid.
• A limit order says: "Fill me only at this price or better." If it cannot be filled right away, it rests in the book and waits. Resting limit orders are what create depth; they provide liquidity.
• A stop order is dormant until price reaches a trigger level. A stop-market order then becomes a market order. So at the moment it is triggered, it acts aggressively, and it can suffer slippage. (A stop-limit order becomes a limit order instead, which caps the price but may not fill at all.)

A limit order priced so that it can trade immediately, for example a buy limit at 4000.3 when the best ask is 4000.1, is called a marketable limit order. It behaves aggressively up to its limit price.

Aggressor and passive

The aggressor, also called the initiator, is the order that causes a trade to happen at the moment it arrives. CME's own market-data documentation uses essentially this definition: the aggressor is the order that, when entered into the book, immediately triggers a trade. Usually it is a market order or a marketable limit order.

The passive side, or resting side, is the limit order that was already sitting in the book and got filled by the aggressor.

Worked example: the best ask in GC is 4000.1 with 80 contracts resting. A trader sends a market order to buy 80. All 80 fill at 4000.1. The buyer is the aggressor; the sellers whose limit orders were resting at 4000.1 are passive. Because the level is now empty, the best ask moves to 4000.2. On Time & Sales you would see a print of 80 at 4000.1, usually coloured as a buy, and the price ladder would shift up one tick.

Every trade has one buyer and one seller

This is the single most important sentence in order flow: every trade has exactly one buyer and one seller. Every contract bought is a contract sold. So it is impossible for "more contracts to be bought than sold." The totals are always equal.

What differs is which side was in a hurry. When order-flow traders say "500 contracts of buying came in," they mean 500 contracts where the buyer was the aggressor. The sellers of those 500 contracts were real too; they were simply waiting passively.

Key idea: "Buying" and "selling" in order flow always mean aggressive buying and aggressive selling. The number of contracts bought and sold is always equal.

How price actually moves

Putting it together:

• Aggressors consume liquidity. When they use up all the resting orders at a price, the best price shifts and the market moves.
• Passive orders provide liquidity. Large passive orders can soak up a lot of aggressive activity without letting price move. This is called absorption (Chapter 9 looks at it closely).
• Passive orders can also move price by leaving. When resting orders are cancelled or moved, the best price shifts without much trading at all (Story B from earlier).

Aggressive does not mean smart or big

It is easy to assume that the aggressive side is the "strong" or "informed" side. That is not necessarily so. Large participants often prefer to trade passively: resting limit orders avoid paying the spread and leave a smaller visible footprint. Their algorithms frequently split a large order into many small passive pieces. Meanwhile, a lot of aggressive volume comes from stop orders being triggered, which is forced, not chosen.

Common mistake: "The aggressor is always the smart money." This is a popular claim, not proven.

Common mistake: "Limit orders never move the price." Cancelling or moving limit orders can shift the best bid or ask just as surely as aggressive orders can.

---

Pattern Trader vs Volume Trader: Same Gold Chart

The best way to see what volume adds is to watch two traders look at the same moment. We will not declare a winner. The point is to see the different questions they ask.

The scene

It is the New York session. Yesterday's high in GC was 4035.0 (about 4,005–4,008 on XAUUSD). Gold rallies into that level and prints a 5-minute candle that opens at 4031.2, spikes to 4036.0, and closes at 4032.0. The candle has a long upper wick, called a wick or shadow: the thin line above the body showing prices that traded but did not hold.

The pattern trader's read

The price-pattern trader sees:

• price at a known resistance level (yesterday's high);
• a pin bar, a candle with a long wick and small body, often read as rejection;
• RSI(14) above 70.

Their conclusion is a hypothesis: "Resistance, rejection candle and an overbought reading. A pullback is more likely than not." Their data is the shape of the candle, the price levels and the indicators. Their underlying question is: "How has this pattern behaved in the past?"

The volume trader's questions

The volume trader looks at the same candle and asks different questions, using data the pattern trader does not see:

1. Was this candle's volume high or low for this time of day? Gold's volume has a strong daily rhythm, so a candle should be compared with candles at the same clock time on previous days (Chapter 3 shows how).
2. How much traded inside the wick? Did the prices from 4033 to 4036 trade 60 contracts or 1,450 contracts? A wick built on almost no volume means price briefly poked up where nobody did business. A wick built on very heavy volume means a lot of business happened there and price still failed to hold.
3. Who was aggressive? Was the candle's delta (aggressive buying minus aggressive selling) strongly positive? If many aggressive buyers arrived at the high and price still could not rise, someone large may have been selling passively against them: possible absorption. If almost no aggressive buyers arrived at all, the reversal may simply reflect a lack of interest.
4. Is the session's running delta moving with price? Is the cumulative delta (Chapter 3) rising with the rally, or flat while price climbs?

What the difference really is

The two traders might reach the same conclusion. The difference lies in the quality of the explanation and in how precisely each can say what would prove them wrong. The volume trader can state a sharper invalidation condition: "If aggressive buyers return in size and this time price moves up through the high, my 'sellers are absorbing' explanation is wrong." That is testable against data; "the pin bar failed" is vaguer.

One example proves nothing

Any book can show a chart where its method worked. That is cherry-picking, and one chosen example proves nothing about average behaviour. The message here is "more information and better questions," not "my method beats yours." Many experienced traders combine price levels with order flow rather than choosing one.

Common mistake: "Volume traders are always ahead of pattern traders." This is a claim without evidence.

Common mistake: "Long wick plus high volume means a certain reversal." This is a popular claim, not proven.

Try it: Take any gold candle with a long wick at a prior day's high or low. Before reading the outcome, write down the four volume trader questions and answer them from the data. Only then look at what happened next, and remind yourself that it is one example.

---

Order Flow Myths: The Holy Grail and "Big Volume = Big Buyers"

Order-flow education is often sold with big promises. This section is a vaccination against them. For each myth, ask three questions: Where is the evidence? Does it hold after costs? On how large a sample?

Myth 1: "Order flow is the holy grail"

No data set is a holy grail. Well-known educators in this field openly describe order flow as neutral information that cannot rescue a method that does not work in the first place. In our own testing, order-flow signals on their own did not predict direction after costs. Order flow is information, not a system.

Myth 2: "Big volume means big players are buying"

Volume is always both buying and selling. High volume means a lot of trading happened; it says nothing about direction. A very high-volume candle might be big participants exiting, a cluster of stops being triggered, or a battle between two sides.

Myth 3: "I can see the institutions"

Exchange data has no identities: you see price, size, time and (with good data) the aggressive side. Large orders are usually split by algorithms into small, often passive pieces, while one large print may be several small orders matching at once. "Who" in order flow means size, side and timing, never names.

Myth 4: "Delta divergence means a reversal is coming"

A popular claim, not proven. When price and cumulative delta disagree (Chapter 3), the disagreement can last hours or days while price keeps going; some platform educators warn it may reflect large passive flows rather than weakness.

Myth 5: "Absorption always leads to a bounce"

A popular claim, not proven. Heavy aggressive volume that fails to move price can still break through; an apparent wall may be a temporary iceberg order that simply finishes.

Myth 6: "Tick volume in forex is the same thing"

Related, but not the same; the next section explains why.

"A popular claim, not proven" does not mean false. It means the idea has not been shown to work in a fair test: on enough data, out of sample, after costs. Treat such ideas as hypotheses, not facts.

Key idea: Volume is direction-free. High volume tells you "a lot happened here," never "this way next."

Try it: Find three high-volume 5-minute candles on a gold chart from the past week. For each, note what happened in the following hour. You will very likely find that the same kind of candle was followed by continuation in one case and reversal in another.

---

Real Volume vs Tick Volume: Why Futures, Not Spot or CFD

Many traders first meet gold through a forex broker, trading XAUUSD on MetaTrader or a similar platform. Those charts also have a "Volume" panel. It is important to understand that this is usually not the same kind of volume as on a futures chart.

Real volume

Real volume is the actual number of contracts traded, as reported by the exchange. In GC, every trade is published with its price, its size (say, 3 contracts) and its time. CME's market-data feed even identifies which side was the aggressor (Chapter 3 explains how). Because all GC trades pass through one central order book on CME Globex, the exchange's count is the complete count for that contract.

Tick volume

Tick volume is the number of times the quoted price changed during a candle, regardless of how much was traded. In MetaTrader and most forex or CFD charts, this count is what is labelled "Volume".

A small example shows the difference. Suppose in one minute of real trading:

• a 1-contract trade prints at 4000.1;
• a 1-contract trade prints at 4000.2;
• a 500-contract trade prints at 4000.2.

Real volume for that minute is 502 contracts. A tick-volume counter that counts price changes sees only two updates (to 4000.1 and to 4000.2). The 500-contract trade, the most important event of the minute, did not change the price, so it may not register at all. In tick volume, a 1-lot trade and a 1,000-lot trade weigh the same, or nothing.

Why spot and CFD markets have no central volume

Spot gold and CFDs are decentralised (over-the-counter) markets. A CFD (contract for difference) is a private contract between you and your broker, not an exchange-traded product. Each broker and liquidity provider streams its own prices, and no single place records every trade in the market. The "volume" on your broker's XAUUSD chart reflects only the price updates in that broker's feed.

Is tick volume worthless?

Not completely. Studies have found a fairly strong correlation between tick volume and real volume on higher timeframes and during busy hours. It is a reasonable rough proxy for how active the market is. But the relationship is unstable, and it tends to break down exactly at important moments, such as when a few large trades go through with very few price changes.

More importantly, tick volume contains no information about the aggressive side, trade sizes, or volume at each price, so a genuine footprint, delta or cumulative delta cannot be built from it. Any "delta" on CFD data is an estimate from price movements.

Key idea: For footprint, delta and cumulative delta you need trade-by-trade data with sizes and an aggressive side. Only a centralised exchange market such as CME gold futures provides that.

What this book uses

The reference data throughout this book is GC (and MGC) from CME, even if a reader ultimately trades something else. If you trade XAUUSD, you can still read GC's order flow as a guide to what is happening in the central gold market, translating levels with the approximate futures-to-spot gap.

Gold note: When you transfer a GC level to XAUUSD, subtract the current gap (recently roughly $27–30), and remember that the gap moves. A GC level of 4035.0 would sit at approximately 4,005–4,008 on XAUUSD. Check the live gap on the day rather than using a fixed number.

Real futures data has a cost: CME charges a monthly market-data fee (lower for non-professional users), and the quality of what you see depends on your data feed, which is the subject of the next section.

Common mistake: "The Volume in MT4/MT5 is real gold volume, and the delta on my XAUUSD chart is real." The first is almost always your broker's tick volume; the second is an estimate, not exchange aggressor data.

---

The Tool Map: Platforms and Data Feeds

To read order flow you need three separate things, which are often chosen independently:

1. A platform: the software that displays and analyses the data.
2. A data feed: the service that delivers trade-by-trade data and order-book depth from the exchange to your platform.
3. A broker (in futures, an FCM, futures commission merchant): the firm that holds your account and sends your orders to the exchange.

The data flows like this: CME Globex → data feed → platform → you. Orders flow the other way: you → platform → broker/FCM → CME Globex.

The main platforms

The descriptions below are neutral summaries, not recommendations. Features, prices and versions change often, so always check the current documentation.

• NinjaTrader 8. A general trading platform whose Order Flow+ add-on provides Volumetric Bars, its name for footprint charts. Delta can be calculated as BidAsk (each trade compared with the bid and ask) or UpDownTick (each trade compared with the previous trade's price). BidAsk bars on historical data need Tick Replay and tick history that includes bid and ask.
• ATAS. Focused on volume analysis: many cluster (footprint) styles, cumulative delta with an optional session reset, and cluster statistics.
• Bookmap. Built around a heatmap of order-book history (time across, price up, colour = resting size), with trades drawn as bubbles. Suited to watching intent and changing depth (Chapter 10).
• Sierra Chart. Fast and highly configurable; its footprint is called Numbers Bars, and it stores full tick-by-tick history. Steeper learning curve.

Data feeds

For CME futures, the most common feeds among retail order-flow traders are Rithmic and CQG. Sierra Chart also offers its own feed, Denali. When comparing feeds, ask:

• Is it unfiltered, tick-by-tick, or are trades aggregated or sampled? Aggregation ruins footprint accuracy.
• How many levels of depth does it deliver?
• Does it provide historical tick data with bid and ask, so your platform can rebuild past footprints?
• What does the CME market-data subscription cost for your user category? This is a separate exchange fee.

Same data, different lenses

All of these platforms display the same CME data. None of them has access to secret information. The differences are in how they display it and how they calculate things from it.

This explains a confusing experience many beginners have: two platforms showing slightly different delta for the same gold candle. The usual reasons are a different classification method (BidAsk vs UpDownTick, for instance), different feed filtering, different handling of trades that are hard to classify, and different reset or session settings. Chapter 3 explains each of these in detail.

Key idea: Platform, data feed and broker are three different layers. The platform is a lens; the data underneath is the exchange's.

Common mistake: "Platform X shows hidden market data." All of them use public exchange data.

Common mistake: "A free, delayed feed is good enough for order flow." Delayed or aggregated data distorts footprint and delta.

---

Chapter summary

• A chart price is the last trade; the chain is decisions → orders → trades (volume) → price.
• A one-tick move can be a push (aggressive buyers) or a pull (sellers cancelling). The candle looks the same; volume and delta differ.
• Explaining is not predicting. In our own testing, order-flow signals alone did not predict direction after costs.
• MA, RSI and MACD are built mostly from closes; smoothing means lag, and they ignore volume, path, aggressive side and location.
• Auction Market Theory: price rises to find sellers and falls to find buyers; high volume suggests acceptance, thin fast trading suggests rejection. A framework, not a formula.
• The order book holds bids and asks; the spread in GC is usually one tick ($0.10). Globex mostly uses price-time (FIFO) priority.
• The aggressor triggers a trade; the passive order was resting. Bought and sold contracts are always equal; "buying" means aggressive buying.
• Pattern and volume traders ask different questions; the volume trader can often state sharper invalidation. One example proves nothing.
• Popular myths (holy grail, big volume = big buyers, seeing institutions, divergence = reversal, absorption = bounce) are unproven or wrong.
• Real volume (exchange contracts) is not tick volume (price updates); genuine footprint and delta need exchange data such as CME GC/MGC.
• Order flow needs a platform, a data feed and a broker; all platforms show the same CME data through different lenses.

Checklist

• I can explain the chain decisions → orders → trades → price, and why a chart price is not "the price I can buy at."
• I can describe a "push" and a "pull" one-tick move and how volume tells them apart.
• I can explain why averaging creates lag and list four things MA, RSI and MACD ignore.
• I can explain value, balance, imbalance, acceptance and rejection in auction terms.
• I can read a DOM ladder: best bid, best ask, spread, depth, and the effect of a market order on it.
• I can define aggressor and passive and explain why bought and sold contracts are always equal.
• I know the difference between real volume and tick volume, and why genuine delta needs exchange data.
• I can name the three layers of an order-flow toolkit and explain why two platforms may show different delta.

Quiz

1. Gold ticks up from 4000.1 to 4000.2 while only a handful of contracts trade. What is the most likely explanation?
2. Gold has closed at 4000.0 for four candles and then closes at 4010.0. What is the value of a 5-period simple moving average after that fifth candle, and what does the result illustrate?
3. The GC ladder shows best bid 4000.0 and best ask 4000.1, with 12 contracts offered at 4000.1 and 18 at 4000.2. A trader sends a market order to buy 20 contracts. Where does the order fill, and what is the average price?
4. A resting sell limit order at 4000.1 is filled by an incoming market buy order. Which side is the aggressor, and how should a volume trader describe this trade?
5. A 5-minute bar on a forex broker's XAUUSD chart shows "Volume: 950". Why can you not build a genuine delta from this number?

Quiz answers

1. Most likely the sellers resting at 4000.1 cancelled or moved their orders (a "pull"), so the best ask moved up without much aggressive buying. If aggressive buyers had pushed the price up, you would expect to see substantial volume traded at 4000.1.
2. (4000.0 × 4 + 4010.0) ÷ 5 = 4002.0. Price is already $10 higher, but the average has moved only $2. This illustrates lag: averaging the past always delays the response to a change.
3. 12 contracts fill at 4000.1 and 8 at 4000.2. Average price = (12 × 4000.1 + 8 × 4000.2) ÷ 20 = 4000.14. The 0.04 above the expected 4000.1 is slippage, about $80 in total on 20 GC contracts. The best ask becomes 4000.2.
4. The buyer is the aggressor, because the market buy order triggered the trade on arrival; the seller was passive. A volume trader would count this as aggressive buying (a buy at the ask), while remembering that the trade, like every trade, had exactly one buyer and one seller.
5. Because that "volume" is almost certainly tick volume: a count of price updates in one broker's feed. It does not contain trade sizes, does not cover the whole market, and has no information about which side was aggressive. Genuine delta requires exchange trade-by-trade data with sizes and an aggressive side, such as CME gold futures data.

Education only. Not financial advice. Trading involves substantial risk of loss.