Order Book and Market Depth

At a glance

What it shows: Resting limit orders by price level
What it hides: Hidden orders, intentions, and anything not yet submitted
Most reliable use: Measuring available liquidity for your own order
Least reliable use: Predicting direction from displayed imbalance

Key takeaways

The order book displays resting limit orders, which can be cancelled instantly and frequently are.
Displayed depth understates true liquidity because of hidden and iceberg orders, and overstates it because much of it will be pulled.
Order book imbalance has weak predictive value at retail speeds, because the participants who can act on it are far faster.
The most practical use is sizing: the book tells you how much you can trade without moving the price materially.
Time and sales, showing actual executions, is more informative than the book because trades cannot be cancelled.

What the book contains

        BIDS                       ASKS
  Size    Price              Price    Size
  1,200   50.12  <- best     50.14    900   <- best
  3,400   50.11              50.15  2,600
    800   50.10              50.16  1,100
  5,000   50.09              50.18  4,300
  2,100   50.08              50.20  8,700

Best bid 50.12, best ask 50.14, spread 0.02
Midpoint 50.13

Reading this correctly:
  - 900 shares can be bought at 50.14
  - Buying 2,000 consumes 900 at 50.14 and
    1,100 at 50.15, average 50.1455
  - The 8,700 at 50.20 may be real, may be
    iceberg, and may be cancelled before you reach it

A depth view, and what each element means.

The book is a snapshot of intentions that can be withdrawn at any moment. In fast markets, a large displayed size can vanish in milliseconds as price approaches it, which is why depth is a much better guide to current available liquidity than to future liquidity.

What the book does not show

Hidden and iceberg orders. A displayed 200 shares may be the visible tip of a 20,000 share order, which is why price sometimes absorbs far more than the book suggested.
Orders held at brokers. Stop orders and conditional orders generally do not appear until triggered.
Off-exchange liquidity. Dark pools and internalisers hold substantial volume that never appears on any displayed book.
Intentions not yet submitted. The largest orders are frequently worked by algorithms that submit small slices, so the parent order is never visible.
Orders that will be cancelled. A large share of displayed depth is withdrawn before it can be executed against.
Other venues. In fragmented markets, the book you see is one venue among many.

What the book is genuinely useful for

Use	Reliability	Notes
Sizing your own order	High	Shows how much you can trade without moving price
Estimating slippage	High	Walk the book to compute your expected average fill
Measuring current liquidity conditions	Moderate to high	Thin books indicate a risky environment
Identifying the spread and tick structure	High	Directly observable
Predicting short-term direction from imbalance	Low at retail speed	Faster participants act on this first
Identifying support and resistance from large orders	Low	Large orders are frequently pulled
Detecting absorption	Moderate, with tape	Requires watching executions, not just the book

The reliable uses are all about your own order rather than about prediction. Walking the book to estimate your fill before sending an order is genuinely valuable and available to anyone with depth data.

Time and sales: what actually happened

The tape shows completed transactions with price, size, and timestamp, and in many markets whether the trade occurred at the bid or the ask. Unlike the book, a trade cannot be cancelled, which makes the tape more informative about what participants actually did.

Trades at the ask indicate a buyer crossing the spread urgently; trades at the bid indicate an urgent seller. This is the closest available measure of directional pressure.
Large prints indicate institutional activity, though the direction and the reason are not observable.
Absorption appears as heavy volume at a price without the price moving, suggesting a large order is being filled passively.
Speed of the tape is itself information: acceleration indicates urgency and usually precedes volatility.
Trade size distribution distinguishes algorithmic slicing from block activity.
Off-exchange prints appear with a delay and at prices that may not reflect the current market.

A realistic assessment for individuals

Order flow analysis is a legitimate discipline practised by professionals with microsecond infrastructure and message-level data. The retail version, watching a depth ladder, faces a structural problem: by the time a human perceives an imbalance, faster participants have already acted on it.

Do use it for execution. Estimating your fill and sizing to available liquidity is valuable and does not require speed.
Do use it for context. Recognising that the book is unusually thin tells you to reduce size, which is actionable.
Do not use it for prediction at retail speed. Displayed imbalance is acted upon in microseconds by participants you cannot outpace.
Be sceptical of retail order flow tools that present the book as a predictive indicator. The data is real; the claimed edge usually is not.
Consider whether you need it at all. For strategies holding positions for hours or longer, depth data adds cost and distraction without adding much.

Frequently asked questions

Can I predict price direction from the order book?

Imbalance has measurable short-term predictive value, but it is exploited within microseconds by participants with far better infrastructure. By the time a human observes and acts, the information is priced. The book is more useful for estimating your own execution cost than for forecasting.

What is an iceberg order?

A large order that displays only a small portion at a time, replenishing as each visible slice fills. It reduces information leakage at the cost of queue priority on the hidden quantity. Their existence is why displayed depth can understate true available liquidity considerably.

Is Level 2 data worth paying for?

For strategies that explicitly model order book dynamics, yes. For anything holding positions for minutes or longer, generally not: it adds cost and screen time without changing decisions. It is among the most commonly purchased and least used datasets in retail trading.

What is spoofing?

Placing orders with no intention of executing them in order to create a misleading impression of supply or demand, then cancelling once others react. It is illegal in regulated markets and has resulted in prosecutions. Its existence is a reason to treat large displayed orders as weak evidence rather than as reliable levels.

Why does the price sometimes move through a large order?

Either the order was cancelled as price approached, which is common, or it was genuinely executed against by a larger opposing flow. Both happen frequently. Displayed size is an intention that can be withdrawn instantly, which is the fundamental limitation of reading the book as though it were a commitment.

Test this idea before you trade it

Describe the rules in plain language and AlgoTrader AI turns them into a structured strategy blueprint with a configurable historical backtest, cost assumptions, and exportable code.

Build a backtest

Keep reading

Referenced by

Educational use only. This guide explains how a strategy works. It is not investment advice, not a recommendation, and no result described here is a forecast. Test any approach on historical and out-of-sample data, size positions conservatively, and never risk money you cannot afford to lose.