order book data · level 2 market data · limit order book data · market depth data
Order book data
Order book data shows market depth: the bids and asks available at different price levels. It is used for liquidity analysis, market microstructure, execution modelling and high-frequency research. Compared with trades or OHLCV bars, order book data is closer to the mechanics of how markets actually clear.
What this means
An order book is the set of resting bids and asks for an instrument. Order book data records the state of that book or the events that update it. It is often described as level 2 data or market depth data.
The key distinction is between a snapshot of visible depth and a stream of incremental updates. Reconstructing a book from updates is more powerful, but operationally harder.
Main data and source types
Order book datasets can include best bid and offer, multiple price levels, full depth, snapshots, incremental updates, sequence numbers and venue fields. Crypto venues, equities exchanges and futures exchanges may expose different structures.
- Level 1 for best bid, best ask and last trade.
- Level 2 for multiple depth levels.
- Full-depth feeds for detailed market microstructure work.
- Snapshot files for lower-frequency liquidity studies.
Free or public sources
Some crypto exchanges provide public order book streams and snapshots. They can be useful for learning book reconstruction, but availability, schema stability and historical completeness vary by venue.
For regulated markets, free public order book history is usually limited. Exchange-derived or vendor-supplied data is more common for serious work.
Paid or vendor sources
Paid order book data may provide exchange feeds, historical archives, normalisation, replay tooling and support. Licensing can be restrictive because the data is close to exchange infrastructure.
Check how dropped messages, sequence gaps, auctions, halts, hidden liquidity and venue-specific order types are represented.
API and infrastructure considerations
Order book infrastructure needs careful handling of sequence numbers, snapshots, incremental updates, dropped packets, compression, time ordering and replay. Small reconstruction errors can create large research errors.
Store enough raw information to rebuild or audit the book. A derived depth table may be convenient, but it can hide reconstruction assumptions.
Common use cases
Common uses include liquidity measurement, spread analysis, queue dynamics, slippage modelling, execution algorithms, market impact research, HFT research and crypto market structure analysis.
Limitations and risks
Order book data is large, fast and sensitive to implementation details. Visible depth is not the same as executable liquidity, and book state can change faster than a backtest can realistically react.
Selection checklist
Define the venues, depth, update format, timestamp requirements, reconstruction method, storage budget, replay needs and licence rights before selecting a source.
FAQ
What is order book data?
Order book data shows outstanding bids and asks at different price levels, giving a view of market depth and liquidity.
What is Level 2 market data?
Level 2 market data usually refers to market depth beyond the best bid and ask, often including multiple price levels.
Is order book data needed for backtesting?
It is needed for some execution-sensitive or high-frequency strategies, but not for every backtest.
Why is order book data difficult to store?
It changes frequently, can be very large and often requires sequence-aware reconstruction from snapshots and incremental updates.
Ledgerstone is an independent financial-data research guide. It does not provide investment advice, trading advice, brokerage services, data vendor services or financial promotion.