2013issue C1242-48
Liquidity pockets and order routing after the speed race
Short-horizon prices can leave fundamentals behind, so execution rests on tape, book, and flow. After visible size became easy to hide, unused liquidity pockets grew smaller and shorter-lived. Choose an order that can complete, judge routing from the public book and prints, and screen cost before sending size.
- Day-to-day and minute-to-minute prices can diverge from fundamentals, so short-horizon decisions rest on tape, book, and flow rather than the news headline.
- A liquidity filter is a go or no-go screen that asks whether a pocket still holds enough executable size before any order is sent.
- When a tactic is easy to see, participants hide intent and the edge decays, so order routing and fill quality have to be judged from the public book and prints.
- Thin, segmented volume produces a thunder-and-tumbleweed tape that leaves unused liquidity pockets smaller and shorter-lived.
Tape, book, and flow come first
Day-to-day and minute-to-minute price movement can diverge from fundamentals. For a short-horizon order, the useful picture is the tape, the book, and the flow, not the news headline.
Order-flow analysis is the habit of reading those three together over the life of the order. The point is to see who is absorbing size and who is pressing it, including the ax: the participant whose prints and book activity are currently leading the price in the name.
Visible size taught participants to hide
Visible Level II size and time-and-sales once let clustered retail orders move a name when a leading market maker held the bid or pressed the offer. That playbook weakened after decimalization and the fade of that momentum regime.
When a tactic becomes too easy to see, participants hide intent and the edge decays. The same transparency cycle later repeated from early electronic daytrading into the speed race among high-frequency programs. What used to sit in plain sight on the book now has to be inferred from prints, spread, and how the ax behaves while the order is live.
Oscillations after speed stopped being scarce
After speed ceased to be scarce, participation algorithms such as VWAP were described as a large source of the oscillations that short-horizon traders still have to trade through. Those swings are part of the tape the order has to live inside. They are not a reason to skip the fill check or the cost of getting in and out.
Smaller pockets on a thunder-and-tumbleweed tape
A liquidity pocket is a brief window of executable size that can magnify a move, then disappear. Thin, segmented volume leaves fewer of those windows unused, and the ones that remain last for less of the order life cycle.
The matching tape is thunder-and-tumbleweed: a market state that switches between sudden, forced bursts of activity and long stretches of thin, segmented volume. The bursts bring false breakouts and sudden reversals. That is the setting in which unused liquidity pockets get smaller and shorter-lived.
Route from the public book and prints
Order routing is the choice of how and where an order is sent so it can fill without leaking intent or paying avoidable implementation cost. Dark-pool access and institutional flow information remain an execution advantage that a retail order cannot assume it shares.
Routing and fill quality then have to be judged from the public book and prints. If the book, the spread, and the tape do not show size that can complete, the order is leaking intent into a pocket that is already gone.
Screen size before the order is sent
A liquidity filter is the go or no-go screen that asks whether enough executable size exists in a pocket before a trade is sent. A practical version of that screen is to watch price action together with book, spread, volume, and cost metrics before sending size, so the chosen order can actually complete across its life cycle.
All readings on this track · 12 readings
- 1988Constructing a value area from TPO, volume, and order flow
- 1988Constructing Value Area trends from half-hour auction profiles
- 1988Constructing auction structure with volume as a clock
- 1999Direct access routing and the visible book
- 1999A floor-to-screen case for the trading decision
- 2001Round-number clusters behind support and pattern signals
- 2001Displayed-size filters in multi-venue index-share execution
- 2006A high-volume wide bar as an auction-intent check
- 2007Market Profile as auction context for inventory imbalance
- 2008Decimalization, liquidity, and why price-only charts miss the tape
- 2013Liquidity pockets and order routing after the speed race
- 2015Bid-ask queues and the instant-adjustment critique