Quant Memo
Core

How The Queue Rebuilds After A Sweep

What happens to an order book's depth in the seconds after a large aggressive order clears out a price level, and why the speed and shape of the refill matters for anyone trading right after.

Prerequisites: Order Book Mechanics

A large market order sweeps through several price levels of the book, clearing out the resting limit orders at each one. For a moment right after, the book at those prices is empty or near-empty. What happens next isn't instantaneous — the book doesn't teleport back to its pre-sweep state. It rebuilds gradually as new limit orders are placed, and the speed and pattern of that rebuild is itself informative: a book that refills fast and deep suggests the sweep wasn't driven by new information, while a book that stays thin suggests liquidity providers are cautious about the price level, possibly because they suspect the sweep was informed.

What determines the rebuild

Market makers who had orders cleared out by the sweep face a decision: was that sweep just a large but uninformed order (a fund rebalancing, an index tracker trading), or was it driven by someone who knows something about where the price is headed? If they believe the former, they replenish their quotes quickly, often near the pre-sweep price, since nothing fundamental has changed. If they suspect the latter — that the sweep reflects real information — they widen their quotes or pull back from posting at the old price altogether, worried about being picked off again by the same informed flow. This is why the same-size sweep can produce very different rebuild speeds depending on context: a sweep during a scheduled index rebalance rebuilds fast, while a sweep right before unexpected news breaks (or immediately after) tends to rebuild slowly and at a shifted price, because liquidity providers are repricing their belief about where fair value now sits.

Worked example

Before a sweep, a stock shows 2,000 shares at each of five price levels above the best ask, spanning $50.00 to $50.08. A large buy order sweeps through all 2,000 shares at $50.00 and 1,500 of the 2,000 at $50.02, moving the best ask to $50.02 with 500 shares remaining there. If the sweep is judged uninformed, within a few seconds market makers typically re-post near $50.00–$50.02 with depth close to the pre-sweep 2,000-share level, and the book looks largely as it did before, just with the last trade printed higher. If instead the sweep is judged informed — say it coincided with unusual options activity — market makers might re-post at $50.02 with only 400–500 shares, well below the pre-sweep depth, and wait to see where the price settles before committing more size. The visible difference: fast, deep refill at roughly the old price versus slow, thin refill at a shifted price — both start from the identical post-sweep snapshot but diverge based on what liquidity providers infer about the sweep's motive.

0 depth uninformed: fast, full rebuild informed: slow, thin rebuild
Both paths start at the same swept-out depth, but a sweep judged uninformed refills quickly to its old level while a sweep judged informed leaves the book thinner for longer.

What this means in practice

Traders working an order right after seeing a sweep should treat the immediate post-sweep book with caution: apparent depth at the new price might be thin and unreliable if liquidity providers haven't yet decided the sweep was harmless. This is directly relevant to execution algorithms, which often build in a short pause after detecting a large sweep before resuming aggressive participation, precisely to let the rebuild pattern reveal whether the level is safe to trade at again. Measuring how fast depth typically returns after a sweep, for a given stock, is a genuine microstructure signal — a "liquidity resilience" characteristic that varies by name and by market condition.

An order book doesn't reset instantly after a sweep — it rebuilds at a speed that reflects what liquidity providers infer about the sweep's cause. Fast, deep refill signals the sweep was seen as uninformed; slow, thin refill signals providers suspect it carried information and are pricing that risk in.

Related concepts

Practice in interviews

Further reading

  • Biais, Hillion, Spatt, An Empirical Analysis of the Limit Order Book and the Order Flow in the Paris Bourse, Journal of Finance (1995)
ShareTwitterLinkedIn