The Depths of the Market: Understanding Liquidity and the Order Book

In this series so far, the invisible has kept turning out to carry more weight than the visible. The Long Shadow of Markets traced how the same value priced in two places disagrees with itself, and how arbitrage drags the two back together. The Invisible Balance showed why a crowded edge decays once enough people act on it, and why the crowd ends up as fuel for whoever is positioned against it.

This part goes into the engine room, where every tick, every candle and every apparently random fluctuation is made: liquidity, and the place it shows itself, the order book.

Liquidity usually gets the classroom definition, "the ease of buying or selling without moving the price." That's true, and far too thin. Liquidity is the bloodstream of a market, the medium everything else circulates through, and the condition for trade to happen at all. Like a river, it doesn't flow evenly. It pools in some places and dries up in others, surges when storms arrive, and now and then floods and drowns the people standing in it. Understanding it takes you underneath the candle to the why beneath the what: why price accelerates without warning, why reversals keep appearing at the same kinds of levels, and why so much of the noise turns out to carry fingerprints.

This page is meant to be come back to. It starts with the anatomy of the book and the orders that fill it, then the ways the book misleads, stop runs, liquidity pockets and fair value gaps, price discovery and reversion to the mean, a case study in spoofing on the public record, the places these readings fail, and a checklist to run before trusting a level.

#The Order Book: Anatomy

An order book is the list of resting buy and sell orders on one venue for one instrument, arranged by price. Buy orders are bids, sell orders are asks (or offers). The highest bid and the lowest ask are the best bid and best ask, together the top of the book, and the gap between them is the spread. Everything resting at prices further away is depth: how much you could trade, and at what prices, before you'd run out of counterparties.

Two roles follow from the book. A maker posts an order that rests and waits, adding liquidity. A taker sends an order that executes against a resting one, removing it. Most venues charge takers more than makers, and some have paid makers to rest orders, which is the edge I wrote about in A Small Edge, Carefully Kept. Every trade has one of each: the maker who waited and the taker who crossed the spread to reach them.

How much of the book you can see depends on the feed. Level 1 is the top of the book: best bid, best ask, last trade. Level 2 is depth by price, the total resting at each level. Level 3, where a venue publishes it, is the book order by order. A chart shows none of this; it shows the trades the book produced, after the fact.

#Order Types

The stop deserves a second look, because most of this page turns on it. A stop-loss is usually a stop-market order: an instruction to sell (or buy) at whatever price is available once a level breaks. It sits off the visible book, so nobody sees it, and it becomes a market order at exactly the moment the market is already moving through its level. Thousands of them at the same obvious place make a pool of guaranteed, price-insensitive orders. Keep that in mind for the sections below.

#Theatre of Intention

At first glance the book looks like a neutral display of supply and demand. Beginners take it at face value, as if what's shown must be what exists. The book is closer to a stage. Intentions are signalled on it, disguised and withdrawn in quick succession, and a lot of what's displayed is there to shape what others believe rather than to be filled.

Take the iceberg. An institution that needs to buy thousands of contracts can't put the whole order into the book, because showing it would move the price against them. So it shows a small tip, and as each tip fills, the next replaces it. To a casual observer the level looks ordinary; underneath it, someone is feeding, slowly and deliberately. You only find an iceberg by watching a level absorb more than it ever displayed.

Then there's the opposite: size that's displayed and never meant to trade. Spoofing is placing orders with the intent to cancel them before they execute, to create a false impression of supply or demand. The Dodd-Frank Act of 2010 made it explicitly illegal in US futures markets from July 2011, and it's prosecuted (the case study below has two examples on the record). The tactic shows how performative the book can be: a wall of size four levels from the touch tells you something was placed there, and tells you nothing about whether it will still be there when price arrives.

So read the book as a living text that changes while you read it. Some orders are genuine, some are decoys, and all of them sit inside a game of exposure and concealment.

#Stop Runs: Harvesting Liquidity

If the book is the stage, stop runs are where the plot turns. Large participants can't enter or exit size without counterparties, and a pool of stops supplies them on demand. Price is drawn towards areas dense with resting orders, because that's where size can be traded at once. That much is structure. Whether a particular run was engineered, to trade size without revealing intent too early, is a reading, and I'll mark it as one when I make it.

How a stop run typically unfolds:

1. Mapping the pool. Below recent swing lows, or above swing highs, sits a cluster of stops. Nobody needs to see them to know roughly where they are; structure tells you, because traders are taught to put them there.
2. The push. Price moves towards the pool, sometimes gradually through persistent order flow, sometimes abruptly on a catalyst.
3. The flush. The level breaks, the stops turn into market orders, and liquidity spikes. A buyer waiting below (or a seller above) fills size against forced orders at favourable prices.
4. The snapback. With the pool consumed and the waiting order filled, the urgent need to trade at that level fades, and price often reverses sharply, stranding whoever chased the break.

Picture price compressing sideways in a tight range. Retail traders buy the range with their stops just below its low. A participant who wants long exposure would rather not chase price higher; they want sellers. Price probes below the range, the stops trigger, each one a sell, and the waiting bid absorbs them. With the pool exhausted, price springs back into the range and carries on. What you can see is the footprint: a break, a burst of volume, an immediate reclaim. It's what you'd expect from someone accumulating without chasing, and on my reading that's often what it is. Wyckoff called the same event a spring, and Part 4 places it in the cycle; Setup 01 turns it into a procedure.

Where stop pools often form:

• Just beyond obvious swing highs and lows.
• Around round numbers and widely watched moving averages.
• At the edges of consolidation ranges, and at well-tested support and resistance.
• Near earlier gaps, session opens and closes, and the previous day's high and low.

In derivatives markets the same pools form from liquidation levels, placed by an exchange's rules instead of traders' hands, which Part 5 covers.

The practice for an individual trader follows. Keep using stops, because they're essential. Place them where the crowd isn't, size so that a sweep doesn't end the campaign, and build entries around structure that expects the sweep instead of fearing it.

#Liquidity Pockets and Fair Value Gaps

Liquidity concentrates in particular zones, places where many traders rest orders and where larger players can trade without revealing much. These liquidity pockets pull price towards them:

• Round numbers: psychological levels such as 100, 1,000 or 10,000.
• Prior highs and lows: the reference points the crowd watches.
• Support and resistance: levels defended or attacked repeatedly.
• Session opens and closes: where institutional flows concentrate.

Price is drawn to these pockets because they offer what the market is always looking for: counterparties. That's why obvious levels get tested so often. Liquidity gathers there because they're obvious, and the liquidity draws price back. It's also why so many charting conventions work: when enough people stare at the same level, the orders resting there become real, whatever the level's original justification.

Sometimes the opposite happens and liquidity vanishes. In a fast, one-sided market, price can trade through a whole region with very little two-sided business done in it. On a chart, that leaves a fair value gap (FVG), defined across three consecutive candles: in a fall, the first candle's low sits above the third candle's high, and the band between them is a range of prices only the middle candle traded through. In a rise it's the mirror, the first candle's high below the third's low.

Bitcoin's crash on 12 March 2020 left a clean one on Coinbase's daily chart. The 11 March low was 7,583.27; the 13 March high was 5,995. The 12 March candle fell through everything between, so price crossed the whole band from 7,583.27 to 5,995 within a single daily candle. Price first traded back into it on 19 March and reached its top on 23 April, six weeks after the gap formed.

Fair value gaps matter because they're unfinished business. A market that skipped a range of prices in a hurry often returns to it, as if the auction needs to check those prices for willing buyers and sellers. The return isn't guaranteed, and the timing never is, but the gap is a map of where trading was thin, and thin regions are where price moves fastest when it comes back.

#Price Discovery: The Endless Auction

Every market is an auction. Buyers bid, sellers offer, and the last agreed trade becomes the price. That continuous negotiation is price discovery, and it's messy and reflexive. Sometimes buyers turn aggressive and lift every offer in sight, and price climbs in near-vertical surges. Sometimes sellers lean on the bid until buyers capitulate. Each tick records the struggle between urgency and patience.

Over longer horizons a pattern emerges: markets tend to drift back to where volume built up, where buyers and sellers both agreed to trade heavily. Those zones become anchors of perceived fair value, and when price strays far from them, it often swings back, as though tethered by consensus. The picture behind that observation is a volume profile, volume summed by price over a stated window, and how to read one is in Turn the Candles Off.

Price discovery, then, is an auction with no final bell. The market is always probing, and every surge plants the seed of the move that answers it.

#Reversion to the Mean

However far or fast a market runs, it rarely escapes gravity for long. Reversion to the mean is the tendency of price, once stretched far from equilibrium, to come back: inefficiencies build, liquidity thins, participants overextend, and the imbalance invites a correction.

The "mean" is a cluster of reference points rather than a single line: moving averages, volume-weighted prices, value areas, earlier zones of balance. The further price drifts from them, the stronger the pull back tends to become.

Reversion keeps no timetable. Extremes persist, and trends run far beyond what reason suggests; the line usually credited to Keynes, that markets can stay irrational longer than you can stay solvent, is about exactly this. For traders, mean reversion is a danger and an opportunity at once. It punishes late trend-followers who mistake momentum for permanence, and it rewards those patient enough to fade an excess with a stop they can afford.

#Reference

#Case Study: Spoofing on the Record

Most claims about manipulation in the book are inferences. Two cases are on the public record, and between them they show what spoofing looks like when a regulator has the order data.

Navinder Sarao and the E-mini. In April 2015 the CFTC charged Sarao, trading from his parents' house in west London, with manipulating the E-mini S&P 500 future on CME. According to the complaint, he ran a "layering algorithm" that placed large sell orders several price levels above the best ask, kept them there as price moved so they stayed visible and rarely filled, and cycled it on and off to create imbalances in the displayed book. On 6 May 2010, the day of the flash crash, it ran for over two hours before the fall. In November 2016 he pleaded guilty to spoofing and wire fraud and admitted contributing to the order-book imbalance on Flash Crash Day. How much his orders contributed to the crash itself is disputed: the official 2010 report traced the fall to a large automated sell programme, and his orders, by design, were cancelled.

JPMorgan's metals and Treasuries desks. In September 2020 JPMorgan agreed to pay $920 million, the largest monetary relief the CFTC had ever imposed, after its order found that traders on its precious-metals and Treasury desks, including the heads of both, had placed hundreds of thousands of orders in gold, silver, platinum, palladium and Treasury futures between 2008 and 2016 with the intent to cancel them before execution.

Neither case needs a theory about who runs the market. In both, displayed size was placed to be seen and pulled before it traded, and in both it took years and the regulators' own order data to prove it. That's the lesson for anyone reading a book from the outside: displayed size is a claim, and you can't verify it from the screen.

#Where It Fails

• The gap that never fills. A fair value gap is a tendency. Some stay open for years, and trading a fill as a certainty is how traders lose money waiting.
• The stop run that's a breakdown. A break below a range that keeps closing below it is a breakdown, whatever it was hoped to be. A stop run is only proved by the reclaim, which is why Setup 01 waits for one.
• Depth that vanishes. The book you see can be pulled in a second. Market makers withdraw quotes when they fear informed flow, which is Part 6's subject, and the depth that looked like support is gone when price arrives.
• One venue's book. Most assets trade on many venues. A thin book on one may sit beside a deep one on another, and arbitrage connects them.
• Hidden size. Icebergs and off-exchange trading, the subject of Part 7, mean the book you can see is never the whole market.
• Seeing intent where there's only footprint. A sweep and a reclaim fit accumulation, and they also fit a market that simply ran out of sellers. The footprint is fact; the intent is a reading.

#Checklist

1. Where are the obvious levels (swing highs and lows, round numbers, range edges), and which side of them would the crowd's stops sit?
2. What does the book show near those levels, and has the displayed size actually traded?
3. Is there a level absorbing more than it ever displays?
4. Where are the fair value gaps above and below price, and how old are they?
5. Where did volume build up over the window you care about?
6. If price runs a level, what would prove it a sweep (a close back beyond the level on the timeframe you're reading) and what would prove it a breakdown?
7. Is the book you're reading the one where the size trades?

Since this was published: Applied Market Mechanics: Tool 01 · Fibonacci Ranges explains why the ratios it uses gather resting orders, Setup 01 · Spring and Reclaim turns the stop run into conditions, a trigger and an invalidation, and Turn the Candles Off covers volume profiles.

#Conclusion: Reading the Current

Liquidity, stop runs, icebergs, gaps and reversion each look like noise on their own. Seen together, they make a coherent pattern: a market built to find counterparties, to balance itself and to keep the flow moving. Every spike, every grinding consolidation and every violent reversal is the system doing what it must: seeking counterparties, redistributing risk and restoring equilibrium.

Learning to read the current changes how the market looks. It stops resembling a hostile riddle and becomes legible, even if it never becomes predictable. Watch the surface, and learn to read the current running underneath it.

Series note: Part 3 of Understanding Market Mechanics. Previous: Part 2 · The Invisible Balance. Next: Part 4 · The Script and the Rhythm. The practical companion is Applied Market Mechanics.

---

Understanding Market Mechanics describes how markets work, published purely as education. Nothing in it is financial advice or a recommendation to buy or sell anything, and I'm not a financial adviser. Every case study is drawn from the public record and has already resolved, and nothing in a part should be read as a statement of what I hold.