L2 order book liquidity

Top-of-book metrics capture only the frontier of the limit order book. The deeper levels — where the bulk of resting liquidity sits — tell you about market resilience: the ability to absorb large orders without significant price displacement.

Order book shape as a market state variable

Biais, Hillion & Spatt (1995) conducted one of the earliest empirical analyses of the full limit order book (on the Paris Bourse), demonstrating that the shape of the book — not just the top level — contains predictive information about future price dynamics. They showed that the distribution of depth across price levels reveals the strategic behavior of limit order traders, who place orders deeper in the book during periods of uncertainty and closer to the market when they want to trade.

Gould et al. (2013) provided a comprehensive review of the limit order book literature, establishing that book depth at multiple levels is a key state variable for models of price formation. Markets with deep, evenly distributed books are resilient — they can absorb large trades with minimal price impact. Markets with concentration at the top and sparse deeper levels are fragile.

What we measure

We track order book state at four cumulative depth tiers: 5, 10, 20, and 25 price levels per side.

• Aggregate bid/ask depth at each tier — total size available within the top N levels
• Depth distribution — is liquidity concentrated at the top or spread evenly? Concentration at the top with empty deeper levels signals a single dominant market maker — remove them and there's a gap.
• Cumulative depth imbalances — bid vs. ask size across all tracked levels. This extends the Cont, Kukanov & Stoikov (2014) order book imbalance signal into the deeper book.

Depth volatility: the resilience metric

How stable is the total depth over an interval? If depth is stable, market makers are committed — they're not pulling quotes defensively. High depth volatility means liquidity providers are repricing rapidly, which Foucault, Kadan & Kandel (2005) modeled as the optimal response to changing information asymmetry: when adverse selection risk rises, limit order traders either cancel or reprice their quotes.

Depth volatility spikes often precede or accompany large directional moves. The mechanism: market makers detect toxic flow (through their own monitoring of order flow patterns) and begin pulling quotes before the move becomes visible in prices. The L2 depth drain is the leading indicator; the price move is the lagging consequence.

The L1-L2 divergence signal

The most actionable signal from L2 data is its divergence from L1:

• Thin L1 bid, deep L2 bid support → the top-of-book thinness is a market maker inventory adjustment, not genuine weakness. Natural buyers are resting orders deeper in the book.
• Thick L1 bid, empty L2 behind it → a single concentrated quote creating the illusion of support. This is fragile — one cancellation and there's a gap.
• L2 depth draining while L1 looks normal → the most dangerous configuration. The next large trade will hit a much thinner book than the top-of-book suggests. This is the micro-level expression of what Brunnermeier & Pedersen (2009) described as liquidity evaporation.

References:

• Biais, B., Hillion, P. & Spatt, C. (1995). "An Empirical Analysis of the Limit Order Book and the Order Flow in the Paris Bourse." Journal of Finance, 50(5), 1655–1689.
• Gould, M.D., Porter, M.A., Williams, S., McDonald, M., Fenn, D.J. & Howison, S.D. (2013). "Limit Order Books." Quantitative Finance, 13(11), 1709–1748.
• Cont, R., Kukanov, A. & Stoikov, S. (2014). "The Price Impact of Order Book Events." Journal of Financial Econometrics, 12(1), 47–88.
• Foucault, T., Kadan, O. & Kandel, E. (2005). "Limit Order Book as a Market for Liquidity." Review of Financial Studies, 18(4), 1171–1217.
• Brunnermeier, M.K. & Pedersen, L.H. (2009). "Market Liquidity and Funding Liquidity." Review of Financial Studies, 22(6), 2201–2238.