Two data sources sit underneath every order-flow tool, and they answer different questions. Treating them as interchangeable is why so much order-flow analysis arrives at confident conclusions that do not survive the session.
The book is intent. The tape is fact.
What each one is
The limit order book is the set of orders resting at each price, waiting. It is a statement of what participants say they are willing to do. It exists only in the present tense — the book as it stood ten minutes ago is gone unless someone recorded it.
The tape is the record of trades that actually executed: price, size, and which side crossed the spread to make it happen. It is history. It persists, it can be replayed, and it cannot be withdrawn.
That difference — intent versus fact — has a consequence that is easy to gloss over.
Why book imbalance is a live-only measurement
Book imbalance — more resting size on one side than the other — is genuinely useful information about the current moment. It cannot be reconstructed historically without recorded depth data. Some platforms record depth and exchanges sell depth history, but most chart indicators never see it.
So when a tool shows you a book-imbalance reading on a chart from last Tuesday, one of three things is true: it recorded the depth itself and has been running continuously since then, you are paying for a depth-history feed, or the number is an inference — usually "depth" estimated from executed volume, which is a different thing, and worth knowing you are looking at.
This is why our own book pressure reads No L2 rather than showing a plausible
number when no depth is available. A blank is honest. A fabricated reading is worse than nothing,
because you will trade on it.
Spoofing, and why the book lies more than the tape
Resting orders are free to place and free to cancel. That asymmetry is exploitable, and it is exploited: size is displayed to create an impression, and withdrawn before it can be filled.
You do not need to prove intent to handle this correctly. You only need to treat displayed size as a weaker class of evidence than executed size — and to weight a wall that got hit and held far more heavily than a wall that merely sat there looking large.
A wall that absorbs aggression and remains is real by demonstration. It has already spent money proving itself. That is a tape event, not a book event, and it is the reason our block qualification is anchored to executed flow rather than displayed depth.
Using them together
They are complements, not substitutes, and the useful reads come from their relationship.
- Aggression into a stacked opposite side. Buyers lifting into heavy resting offers. The conflict is the read: someone is paying up into size, and one of them is wrong.
- Aggression into a thin opposite side. The same buying against nothing. This is where price moves quickly and where liquidity vacuums matter.
- Balanced tape against a lopsided book. Displayed intent with no follow through — the weakest configuration, and the one most likely to be withdrawn.
The pairing carries information that neither source has alone. Book depth tells you what price should take to move. The tape tells you what it actually took. When those disagree, the disagreement is the signal.
The practical rule
Build your structural decisions on the tape, because it is durable, auditable and replayable. Use the book for what only the book can tell you — the shape of resting liquidity right now — and never let a book-derived reading silently become part of a historical record it cannot honestly be part of.
Any tool claiming otherwise is asking you to trust a number about the past that nobody recorded.