Understanding Order Flow and DOM

Introduction

This lesson assumes you have completed the Foundations order book lesson.

Order flow is the live stream of aggressor trades and resting limit orders that build every candle. It combines DOM (limit orders waiting), time & sales (executed trades), and footprint (volume per price split bid/ask). Together they show who is pushing and who is absorbing — before the candle prints.

Price action tells you what happened. Order flow tells you how and why it happened — in real time.

For first-principles microstructure see Larry Harris, Trading and Exchanges: Market Microstructure for Practitioners (Oxford, 2003); for the Wyckoff effort/result frame see chapters on absorption and exhaustion in David H. Weis, Trades About to Happen (Wiley, 2013).

Bitcoin is traded heavily on centralized exchanges like Binance, Bybit, and Coinbase. These platforms offer depth, tape, and even footprint-like visualizations of real-time market activity.

In this post, we’ll explore:

Order flow concepts through BTC/USDT examples
DOM (Depth of Market) snapshots and what they reveal
Reading the tape (time & sales) for aggressive intent
How to use footprint-style data to confirm trades

What Is Order Flow?

Order flow tracks aggressors — any taker order (market or marketable limit) that crosses the spread. On Binance and Bybit feeds the aggressor side comes from the trade-tape m flag, not the order type itself. Lifting the ask = buy aggressor; hitting the bid = sell aggressor.

Buyers lift the ask (aggressive buying)
Sellers hit the bid (aggressive selling)

When we track this in BTC, we’re watching how big players interact around key levels like $60,000, $63,200, or $65,000.

DOM in Bitcoin (Depth of Market)

DOM shows limit orders sitting at each price level — waiting to be filled.

Example from BTC/USDT:

Bids (Buy Limit Orders)	Price	Asks (Sell Limit Orders)
6.2 BTC	63,180.00	4.7 BTC
10.1 BTC	63,175.00	3.5 BTC
12.8 BTC (stacked)	63,170.00	2.2 BTC
		0.5 BTC (iceberg refresh slice)

Order Book Depth

Imbalance: 0% (balanced)

Bid/Ask Ratio: 50% / 50%Neutral

What This Tells Us:

There's strong buy interest building at $63,170
The sell wall at $63,180 might cap upside unless aggressively lifted
The 0.5 BTC ask is an iceberg — only a small refresh slice is visible at any moment, auto-replenishing from a hidden parent order after each fill. You detect it by repeated prints at the same level eating into apparent size that never shrinks (see Larry Harris, Trading and Exchanges, ch. on hidden orders).

Spoofing and Layering

Large limit orders that appear, sit, and vanish before being touched — especially when price approaches them — are spoofs or layering. They distort the DOM but never trade. The signature: size that pulls on approach is a spoof; size that survives the test is real. CFTC enforcement actions (e.g. CFTC v. Avraham Eisenberg, 2023) document this pattern in crypto specifically.

DOM vs Tape vs Footprint vs CVD

Lens	What it shows	Time horizon	Best for	Failure mode
DOM	Resting limit orders (intent)	Snapshot, microseconds	Spotting walls, icebergs, queue position	Spoofs and layering distort the picture
Time & Sales	Executed trades (reality)	Tick-by-tick	Sequencing aggressors, sweeps	Loses size context - no levels shown
Footprint	Volume per price split bid/ask	Per-bar (e.g. 1m, 5m)	Absorption, exhaustion, imbalance	Aggregation interval can hide intra-bar flow
CVD	Cumulative buy-vs-sell delta	Continuous, multi-bar	Divergence vs price, regime shifts	Per-venue CVD diverges from cross-venue truth

Tape Reading (Time & Sales) in BTC

The tape shows executed trades in real time.

Time	Price	Volume	Type
12:05:12	63,175.00	0.5 BTC	Buy MKT
12:05:13	63,175.50	0.9 BTC	Buy MKT
12:05:15	63,176.00	1.2 BTC	Sell MKT
12:05:17	63,175.50	3.4 BTC	Sell MKT
12:05:18	63,175.00	0.8 BTC	Buy MKT

What You’re Seeing:

BTC buyers tried to push up → lifted ask to $63,176
Then a wave of sellers hit the bid → 3.4 BTC sold aggressively
Reversal likely as supply absorbed the push

Tape reveals aggressive participants trying to move price — and whether they succeed.

Bitcoin Footprint Chart Example

Platforms like ExoCharts or TradingLite visualize Bitcoin volume per price level (bid/ask split).

Imagine a footprint for a 5-minute BTC candle:

Price	Bid Vol	Ask Vol
63,190.00	50	120 (imbalance, buy pressure)
63,185.00	85	60
63,180.00	210 (absorption)	90
63,175.00	300	45

Insights:

In standard footprint convention, bid volume = trades executed at the bid (aggressive sellers hitting bid), and ask volume = trades executed at the ask (aggressive buyers lifting ask).

At $63,190 ask volume (120) >> bid volume (50): aggressive buyers attacked
At $63,180 bid volume of 210 means 210 BTC of aggressive selling hit a passive bid that did not break — a passive bid limit absorbed the market sells without yielding price. That is absorption: the passive side eats aggressors and the level holds.
Suggests strong support → bullish reaction likely (Wyckoff effort/result: large effort, no result down)

How This Helps Bitcoin Traders

Example Setup:

You’re watching BTC test support at $63,000 — your planned long entry zone (anchor this back to your written trade plan).

DOM shows stacked buy orders at $63,000 and $62,980
Tape shows market buyers stepping in with 1–2 BTC chunks
Footprint shows delta shift to positive — buyers dominate

You enter long at $63,000 with tight stop below $62,950

Live Reading Checklist (5 Steps)

Mark the structural level off your HTF chart before the session starts.
At the level, watch for ≥3 stacked aggressor prints on the same side within ~2 seconds.
Check DOM imbalance — at least 3:1 in your direction across the nearest 3–5 ticks.
Confirm cumulative delta sign aligns with your direction over the last 1–5 minutes.
Enter only if all three confirm within 30 seconds of price arriving at the level — else skip the trade.

What you actually get:

A tighter stop — the level either holds the bid or it doesn't, fast
An objective invalidation — delta flips negative
The option to scratch within seconds rather than minutes

When Order Flow Fails

OF degrades during low-liquidity sessions (Sun 23:00–02:00 UTC), macro-event spikes, and across fragmented venues. A Binance-only CVD can diverge from a Coinbase-led move. Treat OF as a confluence input, not a standalone signal — empirical hit rates on raw delta divergence in BTC perp sit well below 55% without structural context. For news-driven gaps and dislocations specifically, see trading around news.

Tools for BTC Order Flow

Type	Platforms
DOM + footprint + CVD (3D)	Trading Glass, Bookmap
DOM	Binance Futures DOM, Bybit, Bookmap
Tape	TensorCharts, TradingLite, Trading Glass
Footprints	ExoCharts, ATAS (for crypto), TradingLite, Trading Glass
Aggregated OI/Delta	Coinalyze, Hyblock, CryptoQuant

FAQ

What is the difference between DOM and time & sales?

DOM shows resting limit orders waiting to be filled — the intent side of the market. Time & sales shows executed trades only — the reality side. DOM tells you who is willing to trade; T&S tells you who actually did.

What is an iceberg order?

An iceberg is a hidden parent order that shows only a small refresh slice on the DOM at a time. After each fill, the slice auto-replenishes from the hidden remainder. You spot one when repeated prints at the same level keep eating apparent size that never shrinks.

What does absorption mean in order flow?

Absorption is when the passive side of the book eats aggressive market orders without yielding price. On a footprint, it appears as outsized volume on one side at a level that holds — large effort with no result, in Wyckoff terms.

Does order flow work for Bitcoin?

Yes per-venue, with caveats. BTC perp DOM and tape carry real signal at structural levels, but they are noisy across fragmented venues, degraded during low-liquidity hours and macro spikes, and exposed to spoofing. Use OF as confluence on top of structure, not as a standalone signal.

Final Thought

In Bitcoin, where volatility and manipulation are common, order flow gives you a tactical edge.

Caveat: "edge" here means a small probabilistic shift, not certainty. In BTC perp, OF reads carry high false-positive rates outside structural levels, and exchange fees plus slippage will eat most of the gross edge unless you maker-rebate or hold trades beyond the noise window.

It shows:

Whether breakouts have real participation
When sweeps appear to take out resting stops and immediately face absorption from passive size — readable post-event more reliably than live
When real buyers/sellers step in behind the scenes

Read structure first. Use order flow to time the attack.

Next: Algorithmic Thinking — how systematic traders encode the patterns above into rules a machine can execute.