Most traders read a candle only after it has closed. Order-flow traders watch it get built, order by order, on the depth-of-market ladder. The DOM and the tape show who is buying and selling at this exact second, which is the closest thing futures gives you to seeing the other side of your trade. Scalp ES or MES at the cash open and this is the resolution you are missing while you stare at a five-minute chart.
What order flow is
Order flow analysis means reading the live stream of buy and sell orders, both the resting liquidity sitting in the book and the trades actually executing against it, to infer intent and aggression. Compare that to a moving average, RSI or MACD. Those are computed from candles that have already closed, so they lag by construction. Order flow looks at the touch right now, before the bar prints.
One framing keeps you out of trouble. Order flow is descriptive, not predictive. It tells you what is happening at the best bid and best ask at this instant, not what price will do next. You read pressure and reaction, then make a probabilistic call. You are not receiving a signal. Anyone who treats a delta reading as a mechanical buy button finds out the expensive way.
Order flow tells you what is happening at the touch right now. It does not tell you what happens next, and pretending otherwise is how most people misuse it.
The DOM, a price ladder
The DOM, also called the ladder or depth of market, is a vertical price ladder. Each price level shows the resting limit-order size at that price. Bids (buyers willing to buy lower) sit below. Asks, or offers (sellers willing to sell higher), sit above. The two prices that matter most are the best bid and the best ask, together called the inside market or the touch. Those are the only two prices at which an immediate trade can occur.
How deep the ladder shows depends on your platform and your data feed. Many retail platforms display five or ten levels. CME disseminates Market-By-Price depth to ten levels in its standard depth feed (verify current CME specs before you rely on a number). Full Market-By-Order data, where you see every individual order rather than aggregated size per price, requires a richer feed such as CME MBO, which not every retail platform surfaces. Do not assume a universal depth figure.
You trade directly off the ladder. Click a price cell and it sends an order at that price, the one-click workflow scalpers live in. That speed is the point, and it is also why your order types matter as much as your reading. If you want to build resting structure around the touch, brackets and OCO orders are part of the same toolkit, covered in futures order types explained.
Bid, ask and the spread
Best bid is the highest price a buyer will pay. Best ask, or offer, is the lowest price a seller will accept. The spread is simply the ask minus the bid. On a deeply liquid front-month future, the spread is usually locked at one tick, the minimum price increment for that contract.
Tick sizes and values are stable, but re-check them against current CME specs before you publish or size risk on them. As commonly specified:
| Contract | Tick | Tick value | Full point |
|---|---|---|---|
| ES (E-mini S&P 500) | 0.25 | $12.50 | $50 |
| MES (Micro E-mini S&P 500) | 0.25 | $1.25 | $5 |
| NQ (E-mini Nasdaq-100) | 0.25 | $5.00 | $20 |
| MNQ (Micro Nasdaq-100) | 0.25 | $0.50 | $2 |
| CL (Crude Oil) | 0.01 | $10.00 | n/a |
| GC (Gold) | 0.10 | $10.00 | n/a |
Two order types define your role in the book. A market order crosses the spread immediately, taking liquidity at the touch and walking deeper if your size exceeds the resting size. It is the aggressor, and it pays the spread. A limit order rests in the book at your chosen price and provides liquidity. It fills only if price comes to you and someone trades against it, so it does not pay the spread but is not guaranteed to fill. Whether the contract you trade even offers a one-tick spread reliably is a product choice, which is part of why people weigh ES vs NQ vs MES vs MNQ before committing screen time.
Reading the tape
Time and sales, the tape, is a chronological log of executed trades: price, size, timestamp, and usually a color tag for whether each trade executed at the bid or at the ask. That tag is the whole game. By convention, a print at the bid means someone hit the bid, classified as aggressive selling. A print at the ask means someone lifted the offer, classified as aggressive buying.
This bid/ask classification is the raw form of order flow, and it is the same logic that later powers delta and footprint charts. One caveat keeps you honest. The aggressor tag is an inference about which side initiated. It is generally reliable but not a perfect statement of intent. Exchange trade-classification rules and odd-lot or auction prints can muddy it, so treat the tape as strong evidence, not gospel.
The DOM lies; the tape doesn't. Displayed resting size is the most manipulated number on your screen, but executed prints actually transacted.
Footprint and volume profile
A footprint chart, also called a cluster or bid-ask chart, takes that bid/ask classification and maps it onto every price level inside every bar. For each price within a candle you see volume traded at the bid versus volume traded at the ask, written bid x ask. That shows you exactly where buying or selling aggression concentrated within the bar, not just the bar's net result.
Delta is the summary number: volume at the ask minus volume at the bid, the net of aggressive buying versus aggressive selling. Cumulative delta sums it across time so you can watch pressure build or fade. Volume profile is a different cut of the same data, a histogram of volume by price rather than by time over a chosen session or range. Three levels do most of the work:
- POC (Point of Control): the single price with the highest traded volume.
- Value Area: the price range holding a set percentage of total volume. The standard is 70 percent, one standard deviation, the convention from J. Peter Steidlmayer's Market Profile work popularized via the CME. It is a default, not a law, and it is configurable.
- VAH and VAL: Value Area High and Value Area Low, the top and bottom of that 70 percent range.
Absorption and imbalance
Absorption rewards patience. It happens when a large resting limit order soaks up repeated aggressive market orders without price moving through it. If sellers keep hitting a bid and price refuses to drop, a buyer is absorbing the selling, which is potential support and a possible reversal. The same logic runs in reverse when a large offer absorbs aggressive buyers.
Imbalance is the momentum read. It is a diagonal comparison of ask volume at one price against bid volume at the price below (or the reverse), where the ratio exceeds a chosen threshold. A common default is 3:1, meaning 300 percent, but the threshold is user-set, not universal. When several imbalances stack in a row in the same direction, that is sustained one-sided aggression, a momentum signal rather than a reversal one.
Now the contrarian view. For the retail prop trader, absorption tends to beat imbalance. Stacked-imbalance momentum is crowded and fast. Absorption at a known level gives you a defined structure to lean against, so you are either right quickly or out cheap. That is exactly the risk shape a funded account needs.
Order flow on a prop account
The classic use case is scalping the cash open, the 09:30 ET equity open, when volume and volatility spike and the flow is at its most readable. That readability is tied to liquidity and time of day, which is why the same tools feel useless overnight. The RTH vs ETH timing breakdown matters here.
Tooling, spelled exactly. NinjaTrader "Order Flow+" is NinjaTrader's premium suite, which includes Volumetric (footprint) bars, Order Flow VWAP, a Trade Detector and more. Quantower "DOM Surface" is Quantower's heatmap-style depth panel. Bookmap is a standalone platform built around a liquidity heatmap plus a DOM that shows resting orders over time. Each carries a real subscription cost and a steep learning curve, and they differ in feel, worth weighing in something like NinjaTrader vs Quantower for charting.
A worked example on MES. Tick is 0.25 index points, tick value is $1.25, and one full point is $5.00 (4 ticks). On the ladder the inside market is best bid 5000.00 and best ask 5000.25, so the spread is 5000.25 minus 5000.00, which equals 0.25, one tick, fully liquid. You see a large resting bid of 800 contracts at 5000.00. On the tape, aggressive sellers hit that bid repeatedly: prints of 50, 120, 90, 140 and 100 at the bid. Total sold into the bid is 50 plus 120 plus 90 plus 140 plus 100, which equals 500 contracts. Price does not trade below 5000.00, and the 800-lot bid keeps refilling. That is absorption: 500 contracts of selling aggression, zero ticks of downside.
The decision, illustrative and not advice. A buyer is absorbing at 5000.00, so the bias is long. You enter long 4 contracts at the ask, 5000.25. Stop is 1.0 point below entry, 5000.25 minus 1.00, which is 4999.25, a 4-tick risk. Target is 2.0 points, 5002.25, an 8-tick gain. Now the arithmetic. Risk per contract is 4 ticks times $1.25, which is $5.00, and across 4 contracts that is 4 times $5.00, which is $20.00. Reward per contract is 8 ticks times $1.25, which is $10.00, and across 4 contracts that is 4 times $10.00, which is $40.00. Reward to risk is $40 divided by $20, which is 2.0 to 1. Round-turn commission is illustrative (often roughly $0.50 to $1.50 per micro round turn, so verify your firm's rate), say $1.00 times 4, which is $4.00. Net winner is about $40 minus $4, which is $36. Net loser is about minus $20 minus $4, which is minus $24.
Notice why you waited for 500 contracts of real executed selling rather than just trusting the 800 displayed. The 800 could have been a spoof. Executed prints cannot be spoofed, because they happened. That is the discipline: trust the tape (fills) more than the book (displayed size).
The absorption-reversal checklist. All of these should be true before you click:
- Location: price is at a pre-marked level (POC, VAH/VAL, prior day high or low, overnight high or low, or a clean swing). No level, no trade.
- Aggression: the tape shows sustained one-sided market orders, for example heavy selling hitting the bid.
- Failure to progress: despite that aggression, price holds and does not break the level by more than about one tick.
- Resting size holds or refills: the absorbing bid or offer stays present. Allow for it being partly iceberg, so do not trust the raw number alone.
- Confirmation: delta or footprint shows the aggression is being absorbed, ideally a delta-versus-price divergence.
- Risk defined first: the stop is a fixed small number of ticks beyond the level, sized so the stop equals your fixed dollar risk and respects the firm's daily-loss and max-drawdown limits.
- Exit plan: a predefined target of at least 1.5 to 2R, plus a thesis-broken exit. If the level prints through on real volume, you are out immediately, no waiting for the stop.
Kill switches matter as much as entries. Take two consecutive losers (or your firm's max), or come near the daily-loss limit, and you stop for the session.
Spoofing (placing large orders with no intent to fill, then pulling them) is illegal under the Dodd-Frank Act (2010) and prosecuted by the CFTC, yet it still occurs. Iceberg orders, which display only a small visible slice and refill as they fill, are legal and routine.
Where this is not the answer. Order flow is high-skill, high-screen-time and low-timeframe. It rewards fast reading and execution, so it suits part-time, set-and-forget, swing and systematic traders poorly. It is least reliable in thin books: low-volume contracts, overnight and holiday sessions, far-month contracts, and the moments around major data releases when the book empties and reloads chaotically. It also does not fix a bad strategy or poor risk management. It is an execution overlay, not an edge on its own.
One more honest limit, since this site is published by a trade-copier company. Discretionary order-flow scalping is the opposite of copy-trading. A copier replicates a master's executed orders on a delay and cannot replicate split-second tape reading. Latency and queue position make the same DOM signal stale by the time it copies, so mirroring a tape-level scalp across accounts degrades exactly the timing edge it depends on. Phoenix Technologies builds Thor, a server-based futures and CFD copier (about 17ms copy latency, flat $39 per month, 14-day free trial), and the honest fit is slower, rules-based strategies, not live-ladder scalping. Match the tool to the job and see the pricing only if a copier fits your actual plan.
The same simulated DOM caveat applies on a prop challenge. A sim evaluation ladder does not behave like the live book. Sim fills are usually modeled and optimistic, you do not experience real queue position, partial fills or slippage the same way, and your own orders are not interacting with real liquidity. Skills built purely on sim can over-fit to sim fill behavior, so validate your reading where it counts before you trust it with real drawdown.
Frequently asked questions
What is order flow in futures trading?
Order flow analysis is reading the real-time stream of buy and sell orders, both the resting limit orders in the book and the trades executing against them, to infer aggression and intent. It differs from indicators like RSI or MACD, which are computed from already-closed candles and therefore lag. Order flow is descriptive rather than predictive: it shows what is happening at the best bid and best ask right now, not what price will do next.
What is the DOM (depth of market)?
The DOM, also called the ladder, is a vertical price ladder where each price level displays the resting limit-order size, with bids below and asks above. The best bid and best ask, together called the inside market or touch, are the only two prices where an immediate trade can occur. Display depth varies by platform and feed, commonly five or ten levels, and CME disseminates Market-By-Price depth to ten levels in its standard feed, so verify current specs rather than assume a universal number.
How do you read the tape (time and sales)?
The tape is a chronological log of executed trades showing price, size, timestamp, and usually a color tag for whether each trade hit the bid or lifted the ask. By convention a print at the bid is classified as aggressive selling and a print at the ask as aggressive buying, which is the raw logic behind delta and footprint charts. That aggressor tag is an inference about which side initiated, generally reliable but not infallible, since exchange classification rules and odd-lot or auction prints can blur it.
What is absorption in order flow?
Absorption is when a large resting limit order soaks up repeated aggressive market orders without price moving through it. For example, if sellers keep hitting a bid and price refuses to fall, a buyer is absorbing the selling, which can signal support or a reversal. Many experienced prop traders favor absorption at a known level over stacked-imbalance momentum because it offers a defined, low-risk-against structure where you are right quickly or out cheaply.
What is delta and the value area in volume profile?
Delta is volume traded at the ask minus volume traded at the bid, representing net aggressive buying versus selling, and cumulative delta sums it over time. Volume profile is a histogram of volume by price, where the Point of Control (POC) is the single highest-volume price and the Value Area is the range holding a set percentage of total volume. The standard Value Area is 70 percent, a Market Profile convention, but it is a configurable default, not a fixed rule.
Is spoofing on the DOM illegal?
Yes. Spoofing, which means placing large orders with no intent to fill and then pulling them, is illegal under the Dodd-Frank Act of 2010 and is prosecuted by the CFTC. It still occurs despite being illegal, which is why experienced traders weight executed prints on the tape over displayed resting size. Iceberg orders, which display only a small slice of a larger order and refill as they fill, are legal and routine.
Why does a simulated prop-challenge DOM differ from the live market?
A simulated evaluation ladder does not behave like the live book because sim fills are usually modeled and optimistic. On sim you do not experience real queue position, partial fills, or slippage the same way, and your orders are not actually interacting with real liquidity. As a result, skills built purely on sim can over-fit to sim fill behavior, so it is wise to validate your reading on real conditions before trusting it with funded drawdown.
Can a trade copier replicate order-flow scalping?
No, not effectively. A trade copier replicates a master account's executed orders on a delay and cannot reproduce split-second tape-reading decisions. Latency and queue position make the same DOM signal stale by the time it copies, so mirroring a tape-level scalp across accounts degrades the very timing edge it relies on. Copiers are better suited to slower, rules-based strategies than to discretionary live-ladder scalping.