What Is Liquidity in Trading? Order Books, the Busiest Hours and Where Stops Pool

Michael Whitman · Trading since 2017, full-time since 2020 · YouTube creator, 137K subscribers · Published 2026-09-05 · Updated 2026-11-02 · Reviewed 2026-11-01

▶ Watch: Ultimate Smart Money Concepts Tutorial (LuxAlgo)

Liquidity in trading is how much you can buy or sell near the current price without moving it, and it comes from the orders already waiting in the book. On a chart, traders also use liquidity for the stop losses resting above obvious highs and below obvious lows.

The two meanings are the same idea seen from two places. A stop loss is an order waiting at a price, and an order book is a list of orders waiting at prices.

This page measures both: a real bitcoin order book walked with four order sizes, a stock fund’s volume through the trading day, and whether the most obvious stop pools on a chart actually got reached more often. The wider question of failing to sell at all is on the liquidity risk page.

How it works

Limit orders supply it. Every resting limit order is an offer to trade a fixed size at a fixed price. Stack them up and you get the order book: bids below the price, asks above it.

Market orders use it up. A market order takes the best price first, then the next, until it is filled. If the first level is small, the order climbs to worse prices. That climb is slippage, and it is why a quoted price only holds for a small trade.

So liquidity has three faces. The bid-ask spread is the cost of the first level. Depth is how much sits within a small distance of the price. Volume is how much actually traded, which shows where depth was, after the fact.

Stops turn the two meanings into one. A stop above a high is a buy order that fires when price gets there, so a crowd of stops above an obvious high is a crowd of buyers at a known price. A large seller looking for buyers has a reason to want price to go there.

A worked example

One live book. At 05:41:42 UTC on 2 November 2026 (9:41 p.m. Sunday in California), Coinbase’s bitcoin book showed a best bid of $78,244.58 and a best ask of $78,244.59: a one-cent spread, with the mid price at $78,244.585. Here is what market buys of four sizes would have paid, walking up the asks.

Read it like this. A basis point is 0.01%. The $10,000 order barely noticed the book. The $5,000,000 order cleared 346 price levels and its last coins cost $79,000.00, 96.55 basis points over the mid, even though the screen said the spread was one cent.

Depth near the price is the number that mattered. At that moment the asks within 0.1% of the mid added up to about $1,677,541. An order smaller than that stayed cheap; one larger had to reach further. The spread said nothing about it.

The original data

Two venues, 12 snapshots each. The same walk ran on Coinbase and on Kraken, 12 times over five and a half minutes (05:41 to 05:47 UTC), with bitcoin’s mid price between $78,244.58 and $78,340.35. The figures below are the median of the 12.

The cheaper small trade and the cheaper big trade were on different venues. Coinbase’s median spread was $0.01 and Kraken’s $0.10, and a $10,000 buy cost a median 0.04 basis points on Coinbase against 0.20 on Kraken. At $100,000 they were level, 0.56 against 0.53. At $1,000,000 it was 3.73 on Coinbase and 0.86 on Kraken, and at $5,000,000 27.58 against 4.29. Kraken simply had more resting near the price: a median $7,369,041 of asks within 0.1% of the mid, against $1,551,465 on Coinbase.

What this does not show. Twelve snapshots on a quiet Sunday night in the US are a small sample, so these numbers describe that moment, not either exchange in general. Kraken’s public book shows 500 levels a side, enough here to fill $5,000,000. Every snapshot is in the order book file.

When stock liquidity shows up. For a stock fund the public record is volume, which shows where the orders were after the fact. Yahoo Finance’s 5-minute bars for SPY covered 60 regular sessions, 7 August to 30 October 2026. Taking the median of each half hour across those sessions, the first half hour carried 12.78% of the day’s volume and the last half hour 18.86%. The quietest was 1:30 to 2:00 p.m., at 3.41%, so the close drew 5.52 times as much trading. Through the day the medians ran 12.78, 7.02, 6.19, 5.76, 4.94, 3.85, 3.96, 3.85 and 3.41%, then climbed 4.32, 4.91, 6.09 and 18.86% into the close.

Liquidity and volatility did not trade places. A common line says the two move in opposite directions. In SPY’s first half hour the median 5-minute bar spanned 0.106% of its price, the widest of the day, in the second-busiest half hour. The busiest, the last, had a median 0.063%, below each of the first five half hours. Volume and movement bunched together at the open and parted at the close.

Checked on a second fund. IWM, the small-cap fund, had the same shape: a median 17.47% of the session in the first half hour, 16.00% in the last, 4.11% from 1:30 p.m., and its widest bars also came first.

Two limits are stated plainly: Yahoo’s 5-minute volumes added up to a median 85.97% of its own daily SPY volume, and in 18 of the 60 SPY sessions a single 5-minute bar away from the close carried more than a fifth of the day’s volume (the largest, 46.11% on 7 August), which is why medians are used. Leaving those 18 sessions out gives 14.07% for the first half hour, 20.76% for the last and 3.54% from 1:30 p.m.: bigger shares, the same shape. The half-hour figures for both funds are in the volume by half hour file.

Are obvious stop pools reached more often? On IWM’s daily bars from 26 May 2000 to 23 October 2026, 6,642 sessions, a swing high is a high above the two sessions either side. Equal highs are a swing high within 0.10% of the one before, with every high between them lower; a level is taken when a high trades above it within the 20 sessions after it is known.

Obvious did not mean more often reached. Both gaps are well inside chance (p = 0.87 for highs, p = 0.79 for lows), and widening the match to 0.25% or narrowing it to 0.05% did not change that.

The equal groups are small, 42 highs and 23 lows, so the test rules out a large difference, not a small one, and nobody outside the exchanges can count the stops that sat at either kind of level. On Nasdaq’s IWM prices for the last ten years the same rules found the same 16 equal pairs. Every swing is in the IWM swing file and the counts in the summary table.

The video record. Across our study of 24,971 trading videos, 471 have liquidity in the title. The median one gets 12,185 views and 65% never pass 50,000.

Where liquidity sits on a chart

Past anything everyone can see. Stops cluster at swing highs and lows; QQQ printed 13.75 five-bar swing highs a year since 1999, and 54.61 on a one-bar setting. Equal highs and lows, round numbers and trendlines collect them too, because every trader who drew the same line put a stop on the far side of it. Those clusters are what the liquidity pool page maps.

A real case. IWM made a high of $230.23 on 24 January 2025 and $230.32 on 31 January, 9 cents higher, with nothing between them above $230.05.

On 6 February it traded to $230.70, through both, then closed at $228.65, $1.67 under the level. No close in the next ten sessions got back above $230.32, and on 21 February IWM closed at $217.80. In hindsight that reads as a liquidity sweep; on 6 February before the close it was a high 38 cents above two others, and the liquidity sweep vs break of structure comparison measures how often that first read got overturned.

Your stop is part of the pool. A stop just above an obvious high feels safe because it is obvious, which is exactly why it sits with everyone else’s. The usual fix is a stop beyond the likely sweep with a smaller position, so the money at risk stays the same.

When it fails

The book changes before you trade. Depth you saw a second ago can be pulled. On the Coinbase book above, the cost of a $5,000,000 buy ran from 21.81 to 38.85 basis points across 12 snapshots in under six minutes.

It was a break, not a sweep. The move that takes a level and keeps going is a breakout. Nothing tells the two apart at the moment the level is taken, which is why the textbook entry waits for a close back inside.

The level was taken and nothing happened. An obvious level does not prove anything was resting there; the IWM test found obvious levels reached no more often than ordinary ones.

Thin hours. Trading at 1:30 p.m. or overnight means a fraction of the usual orders near the price, so the same order moves it further.

Questions people ask

What is liquidity in trading?

How much can trade near the current price without moving it, set by the orders resting in the book. In smart money language it also means the stops above obvious highs and below obvious lows.

Is liquidity a good trading strategy?

Liquidity is a condition, not a strategy. Trading styles built on it, such as fading sweeps, still need a rule for when the level breaks instead, and on IWM’s daily chart obvious levels were reached no more often than ordinary ones.

What is a good example of liquidity?

On 2 Nov 2026 a $10,000 bitcoin buy on Coinbase cost about 23 cents over the mid price, and a $5,000,000 buy cost about $12,403, because the bigger order had to reach 346 price levels.

When is liquidity highest in the stock market?

At the open and the close. In SPY from August to October 2026 (median of 60 sessions), the last half hour carried 18.86% of the day’s volume and the first 12.78%.

Related

Liquidity explains why price travels to a level. Market structure tells you whether that trip matters, an order block is how you mark the level it may come back to, and the fair value gap left by a fast move is one more place traders expect it to return.

This page is educational, not financial advice. Test every idea on your own charts before risking money.