The London session is a useful local-time window for organising FX observations. Compare its quote records and execution costs with other windows on the same account. A large morning move does not establish a trading edge or explain a previous trader’s performance.

Separate market geography from session hours

FX providers have their own weekly schedules, holiday exceptions and breaks. London is one important sales desk location in BIS turnover surveys. Geography and time are separate: a country share cannot establish the amount traded in one hourly window.

BIS classifies geographic turnover by sales desk location. That differs from an intraday London session or a city-only total. Use the survey period, measure and double-counting adjustment when quoting shares, and do not convert country statistics into an hourly liquidity ranking.

London’s time zone connects Asian and American working windows. That gives useful context for observing order flow, but it does not prove that every account has its deepest book or lowest costs throughout London hours.

When clocks go back, one local time occurs twice

This is a weekend time-export test, not an open-market session example. Two fictional events are recorded on 25 October 2026 at 00:30 and 01:30 UTC. Both have local time 01:30 in Europe/London.

Scroll sideways to see the full diagram

Autumn 01:30 occurs twiceClock exercise for Sunday 25 October 2026. The instant 00:30 UTC is 01:30 in London with offset +01:00. The instant 01:30 UTC is also 01:30 in London, with offset +00:00. One real hour elapses. Local date and time without an offset cannot distinguish the records.SAME TIME, DIFFERENT INSTANTSEurope/London · Sunday 25.10.2026UTC00:3001:30+1 hourLondon01:3001:30UTC +01:00UTC +00:00Two distinct instants.Local time alone cannot tell them apart.Full record: date + time + offset.

The events are fictional; the clock-change rule is real. In the UK on 25 October 2026 clocks go back from 02:00 BST to 01:00 GMT. Offsets +01:00 and +00:00 apply to the instants shown, not a fixed London offset for the whole year. The document GOV.UK — When do the clocks change? confirms that UK clocks go back one hour on 25 October in 2026 and the winter time is GMT.

The UTC line preserves order and the one-hour interval. The two local labels 01:30 need different offsets to identify their instants.

The label "2026-10-25 01:30" alone does not identify which instant was recorded. The first full timestamp is 2026-10-25T01:30:00+01:00; the second is 2026-10-25T01:30:00+00:00. Keep an unambiguous UTC instant or a timestamp with its offset, plus the time-zone name and event ID. If only local labels remain, sorting the clock times cannot recover their order. This problem can occur in a test export or journal even when you are not watching the market during the clock change.

Session hours and why they drift

This article uses 08:00-17:00 Europe/London as an illustrative desk-activity window. London uses GMT in winter and British Summer Time in summer. Convert its boundaries for the actual date and your own zone; this convention is not the trading schedule of every broker or exchange.

For 2026, the UK and Poland change clocks on 29 March and 25 October at the same UTC instant, leaving Warsaw one hour ahead of London. The chosen London window is therefore 09:00–18:00 Warsaw, but 08:00–17:00 UTC in winter and 07:00–16:00 UTC in summer. New York’s 8 March/1 November changes create the separate March/October mismatch. Dates, holidays and actual provider hours still need individual checks.

An export can have the right record count but the wrong members. Assume four fictional records at 07:30, 08:30, 15:30 and 16:30 UTC on both 15 January and 15 July 2026, and an end-exclusive [08:00, 17:00) London window. January includes the last three; July includes the first three after the BST conversion. A fixed [08:00, 17:00) UTC filter returns three on both dates yet wrongly rejects July’s 07:30 and includes its 16:30. Check boundary records and their dated conversions, not just row counts.

Compare pair costs on the actual account

EUR/USD, GBP/USD and EUR/GBP are examples of pairs to monitor around relevant European and UK events. Choose instruments according to your tested method, available quotes and execution conditions; the session label does not establish a ranking.

Compare your own instruments

For EUR/USD, GBP/USD or EUR/GBP, record the same account, size, dated window and sampling rule. Compare spread, commission and actual execution separately. This is a measurement checklist, not an observed depth ranking.

A hypothetical EUR/USD cost check uses pip = 0.0001, EUR 10,000 on a USD account and an assumed USD 1 round-turn commission. Buy at Ask and sell immediately at unchanged Bid, with no slippage, financing or conversion. At Bid/Ask 1.10000/1.10012, the 1.2-pip spread gives USD −1.20 from the execution prices and USD −2.20 after commission. At 1.10000/1.10030, the three-pip spread gives USD −3 and USD −4 net. The cost difference is USD 1.80. These invented snapshots are not session averages or a trading recommendation; spread is already in the prices and is not deducted again.

Why the open can accelerate so sharply

Asian activity continues during the European night. A narrow overnight range is one possible observed pattern, not a rule that the largest participants are asleep. Record the actual range and costs for the dates used in your comparison.

European releases and order flow can coincide with a move beyond an earlier range. A breakout can also reverse. Distinguish an observed pattern from a causal claim, and test the setup with costs before treating it as a rule.

A fast move can create both opportunity and execution risk. Its timing alone does not show how much capital entered the market or predict the next direction.

The overlap with New York and the 16:00 fix

With London 08:00–17:00 and New York 08:00–17:00 local definitions, their intersection is 13:00–17:00 London time for most of the year and 12:00–17:00 during US–UK daylight-saving mismatch periods. These boundaries do not guarantee minimum spreads or the most decisive moves.

The WMR Closing Spot Rates include a benchmark at 16:00 London time. It is a reference rate with its own methodology, rather than a guaranteed executable price for your order. Benchmark-related activity does not ensure a predictable direction or spread.

Using Tokyo 09:00–18:00 Asia/Tokyo and London 08:00–17:00 Europe/London, their chosen windows overlap for one hour in UK winter and two in UK summer. These conventions organise time; they do not establish the busiest or cheapest trading period.

What to do tomorrow

  1. Keep named clocks for Europe/London and America/New_York alongside your own zone. Use the date to convert both window boundaries and retain UTC or offsets in exported records.
  2. Compare morning and evening price ranges across matched dates using the same timeframe. Preserve extreme observations and missing data. A candle chart illustrates range but does not, by itself, establish spread costs, execution quality or profitability.
  3. Collect Bid and Ask observations in selected windows on the same account and size. Record timestamps and sampling rules. Add commission, financing and actual fill comparisons before drawing conclusions about transaction costs.
  4. Plan orders according to your availability and supervision rules. Pending orders require accepted instructions and appropriate trigger conditions; execution price and strategy success are not guaranteed while you are away.