Market Sessions and When Liquidity Appears

The short answer

The same order, in the same market, costs different amounts depending on when you send it. That is the entire practical content of "trading sessions".

The clock is not doing anything. The number of participants willing to take the other side is, and it rises and falls on a daily cycle because the people and desks doing the trading live in particular time zones and keep particular hours.

Why the hour changes the price you get

Two numbers determine what an order costs beyond any fee: the spread you cross, and how far into the book your size reaches.

Both depend on how many participants are actively quoting. When many are, the best bid and best offer sit close together and there is size behind each level. When few are, the spread widens — because a market maker quoting into an uncertain, thin market demands more compensation for the risk of being picked off — and each price level holds less.

So the mechanism is not that "the night session is expensive". It is that fewer counterparties are present, and the cost of transacting is a function of how many are present. The hour is just a reliable proxy.

FINRA's guidance on extended-hours trading in US equities states the general case directly: extended-hours trading is less liquid, involves fewer counterparties, orders may be executed partially or not at all, and prices are often more volatile because fewer trades occur. That is a regulator describing, in plain language, the thing sessions are a shorthand for.

Overlaps, and why they matter

A session is a convention — a rough block of hours during which a major financial centre is at work. Overlaps are the hours when two of those blocks coincide.

The overlap matters because participation stacks. During an overlap you have two populations of active participants quoting into the same book at the same time. More quoting interest on both sides typically means a tighter spread and more depth at each level, which means your order moves price less.

That has one direct consequence and one indirect one:

• Direct: identical size transacts more cheaply. This is a cost fact, and you can put numbers on your share of it with the fee calculator alongside the material in costs of a round trip.
• Indirect: most of the day's volume, and therefore most of the day's range, tends to occur when most participants are present. Every indicator you compute — average range, volume averages, volatility measures — is shaped by that distribution.

The BIS Triennial Central Bank Survey is the standard reference for how much activity global FX markets actually carry; its April 2025 survey put OTC FX turnover at $9.6 trillion per day. The point for a retail trader is not the number. It is that turnover of that scale is not spread evenly across twenty-four hours, and neither is the depth you transact against.

Thin markets are not calm markets

This is the part that costs people money.

"The market is quiet" describes cost and depth. It does not describe risk. A thin book is exactly the condition in which a modest order can move price a long way, because there is very little resting on the other side to absorb it.

The Markets Committee study of the sterling flash event of 7 October 2016 is instructive precisely because of when it happened: during early Asian trading hours, one of the thinnest parts of the FX day. A thin session did not make the market safe. It made the market fragile — the same flow that would have been absorbed at a busier hour instead produced a violent move.

So the correct translation of "quiet" is:

The failure mode is specific and common: a position sized for normal conditions, left running into a thin session, with a stop resting just beyond a level. The spread widens across the handover, the stop triggers on a quote that would not have existed an hour earlier, and the fill lands well past the trigger. Nothing unusual happened. The market simply had fewer people in it.

What 24/7 markets change

Crypto markets do not close. That removes some things and leaves others intact.

What it removes. There is no closing auction, no official daily close, and therefore no reliable overnight gap — price moves continuously rather than jumping between a close and an open. It also removes the "I will deal with it in the morning" option: an adverse move at 04:00 is a live adverse move, not something waiting for a bell. See how crypto markets differ for the broader set of differences.

What it does not remove. The daily rhythm of participation. Market makers, desks and treasuries still operate on human schedules, still have risk limits that reset, and are still concentrated in a handful of time zones. Weekend liquidity in most crypto pairs is thinner than weekday liquidity for the same reason equity extended hours are thinner: fewer of the people who provide depth are at their desks.

The practical upshot is that a continuously traded market has session effects without session boundaries. You cannot look up the hours; you have to observe them in the instrument you actually trade.

Finding the pattern in your own market

This is measurable, and measuring it yourself beats taking anyone's word for it.

1. Record the hour with every trade you take. Entry time, exit time, in one fixed time zone — pick UTC and never change it, because daylight saving shifts will otherwise corrupt a year of data.
2. Record the spread at the moment you dealt, not the spread on the chart. Over a few dozen trades a shape emerges.
3. Compare like with like. An average range computed over thin hours and one computed over busy hours are not the same measurement, and comparing them will mislead you. The same caution applies across timeframes — see timeframes and context.
4. Write your tradeable hours into the plan. Not because other hours are forbidden, but because "I trade this instrument between these hours" is a rule you can follow, and "I trade when it looks good" is not.

What to do with this

Decide two things in advance and write them down.

First, which hours you will deal in, based on the spread you actually observe rather than on a session map from a broker's marketing page. Second, what you will do about positions that run through a thin period — reduce, close, widen the stop and cut the size to match, or accept the cost explicitly. Any of those is a decision. Discovering it at 04:00 is not.

And retire the word "quiet" from your vocabulary. Say "thin" instead. It carries the correct implication.