Most of the world's liquid markets — equities, futures and nearly every digital asset venue — match buyers and sellers through a continuous limit order book. Understanding that mechanism is the starting point of microstructure research: before asking where a price is going, it helps to understand how a price is actually made.
§01The anatomy of a limit order book
A limit order states a side, a price and a quantity, and rests in the book until it is matched or cancelled. The highest resting buy price is the best bid; the lowest resting sell price is the best ask. The difference between them is the spread, and the quantity resting at each price level is the depth.
A market order — or any marketable limit order — consumes liquidity by trading immediately against resting orders. Most venues match orders by price-time priority: better prices first, and among orders at the same price, earlier orders first. Queue position therefore has economic value; an order at the front of a queue is more likely to be filled before the price moves away.
§02Mid-price and micro-price
The mid-price — the average of best bid and best ask — is the usual reference price, but it ignores how much quantity sits on each side. If the bid queue is much larger than the ask queue, the next price move is more likely to be up. A simple refinement weights each side by the opposite side's queue:
Stoikov (2018) formalised this idea as the micro-price: an estimator of the expected mid-price in the near future, conditioned on the current spread and queue imbalance, that behaves better than either the mid or the simple weighted mid at high frequency.
§03Order flow imbalance
Cont, Kukanov and Stoikov (2014) studied how events at the best quotes — limit orders, cancellations and trades — relate to price changes over short intervals. They defined order flow imbalance (OFI) as the net quantity added to the bid side minus that added to the ask side, accounting for quote moves:
Their central empirical finding was that, over short intervals, price changes are well described as approximately linear in OFI, with a slope inversely related to market depth. Thin books move more for the same imbalance. This is a mechanical relationship rather than a trading signal in itself — but it is the foundation on which short-horizon models are built.
§04Price impact and Kyle's lambda
The idea that order flow moves prices has a long theoretical history. In Kyle's (1985) model, a market maker who cannot distinguish informed from uninformed flow sets prices as a linear function of net order flow. The slope, often called Kyle's lambda, measures how much the price moves per unit of net demand — an inverse measure of market liquidity that remains a standard empirical tool.
§05Why this matters
- Prediction: queue imbalance and order flow carry information about short-horizon price moves, which is the raw material of microstructure research.
- Cost: the same mechanics determine what a trade really costs. Crossing the spread and consuming depth moves the price against the trader.
- Robustness: microstructure differs between venues and changes over time with fee schedules, tick sizes and participant mix. Models must be validated per venue and monitored for decay.
Research note: relationships documented in academic studies were estimated on specific markets and periods. They are starting hypotheses to test, not facts to assume.
