The Method The Hub FAQ Join the Room
Forex

Forex Sessions Explained: London, New York, Tokyo and the Overlap

Three city skylines lit in sequence along the curve of a dark globe with a band of brighter light where two of them overlap

The forex market runs around the clock by passing between regional banking centres — Sydney, Tokyo, London and New York. Liquidity is not spread evenly across those hours. It concentrates in the London session, and peaks in the four-hour window where London and New York are both open.

This is the part of forex that beginners take as a feature and experienced traders take as a constraint. "Open 24 hours" sounds like freedom to trade whenever it suits you. In practice it means the same chart can be a fast, liquid, tradable market at one hour and an expensive, directionless one four hours later, with nothing on the chart itself to warn you which you are looking at.

The four sessions and when they run

Each session is really the working day of a group of financial centres. The hours below are the conventional core windows, and they are approximate by nature — see the daylight saving section further down before you write any of them into a plan.

SessionLocal core hoursUTC (approx.)ET (approx., US summer)
Sydney8:00 a.m. – 5:00 p.m.22:00 – 07:006:00 p.m. – 3:00 a.m.
Tokyo9:00 a.m. – 6:00 p.m.00:00 – 09:008:00 p.m. – 5:00 a.m.
London8:00 a.m. – 5:00 p.m.07:00 – 16:003:00 a.m. – 12:00 p.m.
New York8:00 a.m. – 5:00 p.m.12:00 – 21:008:00 a.m. – 5:00 p.m.
London–NY overlap12:00 – 16:008:00 a.m. – 12:00 p.m.

Two structural points follow from the table. First, there is no moment in the trading week when every centre is closed — the handover from New York to Sydney is nearly seamless, which is why forex does not gap intraday the way a single-exchange market does. Second, the sessions are wildly unequal, and the next section is the evidence for that.

Where forex trading actually happens

Session strength is not folklore. It is a measurable fact about where the sales desks sit, and the Bank for International Settlements counts them every three years.

In the April 2025 Triennial Central Bank Survey, desks in four locations — the United Kingdom, the United States, Singapore and Hong Kong — intermediated three quarters of all FX trading. The United Kingdom alone accounted for about 38%, the United States about 19%, Singapore 11.8% and Hong Kong 7.0% (BIS, OTC foreign exchange turnover in April 2025).

Read that as a clock rather than a map and the trading day explains itself. Roughly two fifths of the world's FX flow is transacted in a city whose working day is the London session. Add New York and you have well over half of global turnover concentrated into a stretch of about ten hours — with a four-hour window in the middle where both are at their desks simultaneously.

Why this matters more than a volume indicator. Retail platforms show you the volume of your own broker's flow, not the market's — forex has no central tape. The BIS survey is one of the few genuinely market-wide measurements available, and it is telling you where the counterparties are. When you trade outside those hours, you are not trading a quieter version of the same market; you are trading against a materially smaller set of participants.

The London–New York overlap

The four hours from 12:00 to 16:00 UTC are the busiest of the forex day. Both of the two largest centres are open, the US economic releases land inside the window, and European desks are still positioned and active into their close.

Practically, that produces three things at once:

The third point is the one that gets skipped. A busier market is not a safer market; it is a market where being wrong resolves faster. That is exactly why entry confirmation matters more here, not less — see break and hold: why confirmation beats anticipation.

Daylight saving: why every session table is approximate

Session times are quoted in local working hours, and local clocks move. The United Kingdom, the United States, the European Union and Australia all change their clocks on different dates, and Japan does not change at all.

Two consequences worth knowing:

  1. For a few weeks each spring and autumn, the London–New York overlap is an hour longer or shorter than usual, because the two regions have not yet both switched.
  2. Any session table you find on the internet — including the one above — is correct only for part of the year. UTC is the only stable reference.

The fix is unglamorous and takes thirty seconds: once a quarter, work out what 12:00 UTC is on your own clock and write it on a sticky note. That is your session anchor, and it is worth more than memorising four rows of a table.

What each session is actually good for

SessionMost active pairsTypical character
SydneyAUD and NZD pairsThin. Australasian data lands here; ranges are usually narrow.
TokyoJPY pairs, AUD/JPYModerate. Often sets the range that London later breaks.
LondonEUR, GBP and CHF pairsLargest. Trends frequently begin at or shortly after the open.
New YorkUSD pairs, USD/CADActive into the overlap; the afternoon commonly drifts.

The pattern many traders build around is straightforward: the Asian session sets a range, the London open breaks it, and the overlap decides whether the break holds. That is a description of a tendency, not a rule — it fails often enough that trading it without confirmation and a written invalidation is just a habit dressed up as an edge. Which pairs suit which session is covered further in the best forex pairs for beginners.

The real cost of trading a thin market

Between sessions, two things move against you simultaneously. Spreads widen because fewer market makers are quoting, and average ranges compress because fewer participants are transacting. Your cost per trade rises at precisely the moment the market becomes least likely to travel far enough to cover it.

That is the arithmetic behind a common and confusing experience: a strategy that tests well on London-session data quietly stops working when the same trader runs it at 2:00 a.m. Nothing about the setup changed. The cost structure did. If you take five trades a day, you pay the spread roughly 1,250 times a year, so a fraction of a pip of extra cost is not a rounding error — it is the difference between an edge and a slow bleed. There is more on that trade-off in how many trades a day a beginner should take and in the cross-market view at the best time of day to trade.

Choosing a session around a job

Most people reading this cannot be at a screen from 12:00 to 16:00 UTC, and pretending otherwise is how traders end up forcing scalps into a dead tape. There are three honest options:

  1. Trade the session your clock allows, and pick pairs that suit it. If your evening is the Tokyo session, trade yen and Australian dollar pairs rather than forcing EUR/USD.
  2. Lengthen your holding period. Swing positions entered at any hour ride the sessions rather than fighting them, and you pay the spread far less often. See swing trading with a day job.
  3. Trade a different market entirely. If your available hours line up with a stock or futures session instead, that is a better answer than a compromised forex plan. Forex vs futures sets out the trade-offs.

The option that is not on the list is scalping the quiet hours because that is when you are free. It is the most expensive way to trade this market, and the sessions data above is the reason.

The Sunday open and the weekend

Forex closes for the weekend and reopens as the Asia-Pacific week begins. Prices can reopen away from Friday's close if news broke while the market was shut, which is a genuine risk for anyone holding through — and it is the one moment when a stop order provides no protection, because there is no trading between the two prices. Anyone carrying positions over a weekend should read overnight and weekend gap risk first. The first hour or two of the new week is also thin, and spreads reflect that.

The Generational Wealth way. Sessions are why break & hold exists as a rule rather than a preference. A level tapped in a thin market and a level broken and held into the London–New York overlap look identical on a static chart and are not remotely the same event — which is why we wait for the candle to close beyond the level. Know your next means the callout names the entry, the targets and where price aims after that, so a move that starts in one session still has a plan when the next one takes over. Trail & protect moves the stop up behind each target as it prints. See the method →

Frequently Asked Questions

What are the four forex trading sessions?

Sydney, Tokyo, London and New York. They follow the working day around the world, so as one regional market closes another is already open, which is what makes forex a roughly 24-hour market from Sunday evening to Friday evening. The sessions are not equal in size. London is the largest by a wide margin, New York is second, and the Asian sessions are considerably thinner in the major European pairs.

What is the best forex session to trade?

For the major European pairs, the London session and the London–New York overlap carry the most volume and the tightest spreads, which is why most intraday strategies are built around them. Best is partly a matter of what you trade, though: yen pairs are most active in the Tokyo session, and Australian and New Zealand dollar pairs respond to news that lands in the Sydney session. Match the session to the pair rather than forcing one habit onto every market.

Why do forex session times change during the year?

Because daylight saving starts and ends on different dates in different regions, and some regions do not observe it at all. London and New York shift their clocks on different weekends, so for a few weeks each year the overlap between them is an hour longer or shorter than usual. Japan does not observe daylight saving. Any session table you read is therefore approximate, and the reliable method is to check the times in UTC against your own clock.

Is it worth trading forex at night?

It depends entirely on which night and which pair. The Tokyo session is a real session with real participation, and yen and Australian dollar pairs trade actively through it. The genuinely dead hours are the gaps between sessions, where spreads widen and ranges compress at the same time. That combination raises your cost per trade exactly when the market is least likely to travel far enough to cover it.

Bottom line

Forex is continuous but not uniform. Desks in the UK, the US, Singapore and Hong Kong handle three quarters of global turnover, and the UK alone handles about 38% — which is why the London session and its overlap with New York carry the tightest spreads and the largest ranges of the day. Anchor your session times in UTC because local clocks drift, match your pairs to the hours you can actually trade, and treat the gaps between sessions as a cost problem rather than an opportunity. The wider mechanics sit in the forex trading guide, and the unit those ranges are measured in is explained in what a pip is and how to calculate pip value.

Survive first. Compound second.

The Hub stays free. When you want levels, targets and invalidation called in real time, the room is one click away.

Join the Room