This article covers the live interaction layer of order-book trading: DOM, tape, heat maps, and what they can and cannot tell you in real time. It is not the home for broad market structure or for recorded footprint analysis. Those live in market-structure and order-flow-and-footprint. This page focuses on the practical question of whether visible liquidity is real, whether it trades, and how live activity changes around key levels.



Part I: What the DOM Actually Shows

The DOM Shows Intent, Not Just Outcome (Flow Horse)

The DOM and heat map show displayed liquidity and how it changes. That matters because live interaction can tell you whether the market is about to speed up, stall, or test a level more seriously.

Useful live information:

  • where visible liquidity is concentrated
  • where the book is thin and likely to move faster
  • where price may slow down at a sticky point
  • whether displayed size remains firm or gets pulled

This is why Flow Horse talks about trading from one sticky point to the next. Sticky points are simply the places where liquidity is thicker and price is more likely to pause or transact.

OrderFlowLabs adds a useful clarification here: in centralized futures markets, the DOM is the live view of the central limit order book. What it shows is resting limit liquidity waiting to interact, not the trades themselves.

That distinction matters because:

  • bids and offers on the DOM are unfilled interest
  • the tape shows what actually crossed and traded
  • reading the book without the tape can confuse advertised interest with completed business

Visible Size Is Not the Same as Filled Size

One of Flow Horse’s most practical warnings is that visible size alone is not enough. A large resting bid or offer may look important, then disappear before it trades.

So the live question is:

  • did the level actually get hit?
  • did the size stay there and transact?
  • did price continue through it or respond to it?

Without that, the trader may be reacting to intention that was never real.

Pulling, Stacking, and the Pull-Stack Flip (OrderFlowLabs)

OrderFlowLabs uses a simple live-book vocabulary that is worth keeping:

  • stacking means more resting size is appearing at a price
  • pulling means that visible size is being removed
  • a pull-stack flip is when one side of the book stops leaning in its own direction and the opposite side begins to firm up

This is not a standalone signal. It becomes useful when price is already interacting with a key structural area and the trader wants to know whether displayed intent is changing hands there.


Part II: Tape and Heat Map Use

Tape Shows What Really Traded (Flow Horse)

The tape is useful because it shows actual prints, not just displayed interest. A heat map or DOM can advertise liquidity; the tape confirms whether trading actually happened there.

Best use:

  • pair the tape with the DOM or heat map at key levels
  • check whether the visible size really traded
  • look for shifts in aggressiveness when price reaches an important area

This is also why footprint and tape are closely related. The footprint is effectively recorded tape organized by price and time.

OrderFlowLabs makes the mechanic explicit:

  • trades printed at bid show aggressive selling into resting bids
  • trades printed at ask show aggressive buying into resting offers
  • prints that move through multiple price levels are sweeps, which show urgency rather than simple one-price interaction

Heat Maps Are Best as Context, Not as a Standalone Signal

Flow Horse’s use of heat maps is practical rather than mystical. They help map where liquidity firms up, thins out, or repeatedly appears.

Good uses:

  • identify likely sticky points
  • see whether a key level is being advertised in advance
  • prepare for where price may speed up if the book is thin

Bad uses:

  • assuming every bright level will hold
  • assuming spoof-like behavior is actionable by itself
  • treating the heat map as a self-sufficient trading system

The heat map gives context for interaction. It does not remove the need to see what actually trades there.


Part III: Reading Real-Time Interaction

The Key Read Is Change in Behavior (Flow Horse)

The highest-signal live read is often not one static observation. It is the change in behavior as price reaches a level.

Useful shifts:

  • visible liquidity remains firm and absorbs repeated hits
  • visible liquidity gets pulled and price accelerates through
  • tape aggressiveness increases sharply into the level
  • tape aggressiveness dries up as the level is reached

This is more useful than staring at a static DOM snapshot.

Progress Matters More Than Raw Print Size (OrderFlowLabs)

A strong OFL addition is that aggression only means so much unless price actually progresses. Heavy selling that produces little downside movement can be absorption. Heavy buying that cannot lift price meaningfully can be buyer entrapment or exhaustion.

So the real question is not only “how much traded?” It is:

  • how much traded
  • on which side it traded
  • how far price actually moved as a result

Wait for the Book to Actually Change Hands

The OFL replay examples improve the timing rule here. Traders often see one large sweep or one burst of apparent absorption and rush in too early. The better sequence is:

  • pressure into the level
  • poor progress at the level
  • visible weakening from the pressing side
  • opposite-side initiative finally appears

That final step matters. Absorption without initiative can still become continuation. The better live read is when the book actually starts changing hands instead of merely slowing down.

Use the DOM Around Key Levels, Not Everywhere

Flow Horse is explicit that the human brain is not built to process highly erratic BTC DOM activity continuously and compete with algorithms. The better use of the DOM is selective:

  • around important levels
  • when a setup is already defined
  • when price reaches a known sticky point
  • when you want to know whether the level is firming up or clearing out

That turns DOM reading into a focused decision aid instead of a constant noise source.

Pair Live Interaction with Simpler Synthesized Tools

Another practical takeaway is that DOM and tape often become more useful when combined with simpler synthesized feedback such as:

  • CVD
  • volume spikes
  • spot versus perp comparison
  • a nearby footprint panel

That is usually more realistic than expecting a trader to derive every edge from raw tape speed alone.


Part IV: Limits of DOM Trading

Crypto DOM Is Extremely Noisy (Flow Horse)

Flow Horse’s stance is pragmatic: crypto DOM can be useful, but it is easy to overestimate what a human can extract from it in real time.

Core limits:

  • visible size can vanish
  • activity is fast and fragmented
  • multiple venues dilute the meaning of one local book
  • your reaction time is far slower than the fastest participants

This is why DOM should usually refine a trade, not originate the whole idea.

The Best Outcome Is Often Better Exclusion

Like footprint, DOM often adds the most value by helping the trader stay out:

  • the level did not really firm up
  • the book stayed too thin for a clean response
  • the tape never showed meaningful defense
  • the market was too random to justify interpreting microstructure

That is a valid use. Good tape reading often protects capital more than it generates trades.

Sources

  • Flow Horse, Workshop Presentation #2 - Footprint, TPO, Tape, Nuance
  • Flow Horse, First Workshop Q&A session - 4/16/2025
  • Flow Horse, Workshop Q&A 5/22/2025
  • Flow Horse, Workshop Q&A 6/5/2025
  • OrderFlowLabs, EDU Depth of Market 101
  • OrderFlowLabs, OFL 101 Time and Sales