Orderbook Pressure

Orderbook pressure measures the imbalance between resting bid and ask size — typically the ratio of total bids to total asks within a band around the mid price — quantifying short-term supply/demand skew.

Orderbook pressure is a classic high-frequency, low-persistence signal: its predictive power decays fast with horizon, showing statistical significance from seconds to minutes and essentially vanishing beyond an hour. Always fix the depth band (say ±0.5%), since different bands can yield opposite conclusions. As with large-order statistics, the cancellability of resting orders caps reliability — a sudden shift in the ratio is often just market makers repricing rather than a real change in supply and demand.

Volume and market-structure metrics describe the size and distribution of trading activity. Methodologies differ sharply between venues (wash-trading filters, USD conversion), so confirm the accounting basis before comparing platforms side by side.

Two normalisations are standard: the plain bid-to-ask ratio, and the bounded imbalance, bids minus asks over bids plus asks, which lives between minus one and one and is easier to chart and compare. Both are computed within a fixed band around mid, and the common choices of half a percent and two percent answer different questions, with the tight band describing immediate microstructure and the wide one describing positioning further out. Sampling at one-second to one-minute intervals and then smoothing is typical, because the raw series is dominated by quote flicker. Empirically the signal is mean-reverting and horizon-limited: imbalance predicts the next minutes' drift slightly better than chance, and little beyond that. Its more durable use is anomaly detection, since a persistent one-sided imbalance that survives several refresh cycles at meaningful depth flags either genuine directional demand or a spoofing pattern, and cross-checking with actual fills, meaning whether the pressure converts into volume, separates the two.

The live chart for this metric is still being wired up. Global volume, venue share and perp-vs-spot pages already carry real data.