WhitmanTrading

London Session: Participation, Not Magic

The London session covers the European trading hours, roughly 8am to 5pm UK time, and is the busiest stretch of the foreign exchange day. What a session changes is how many people are trading, which affects spreads, depth and range rather than direction.

How it works

A candlestick chart of the site's shared price history. The headline on the chart reads: The hours when Europe is at its desk.
The hours when Europe is at its desk. Illustrative chart - not real market data.

The London session is roughly 8am to 5pm UK time, when European banks, funds and brokers are at work. It is the largest of the three commonly named sessions by forex turnover.

A gently rising stretch of the long price series. The headline on the chart reads: The overlap with New York is the busiest stretch of the day.
The overlap with New York is the busiest stretch of the day. Illustrative chart - not real market data.

The overlap with New York — roughly four hours in the London afternoon — is the busiest window of the twenty-four. Two large populations are trading simultaneously, which is why spreads are tightest and depth is best then.

A candlestick chart with a volume histogram beneath it, with the volume histogram emphasised. The headline on the chart reads: Participation is the only thing a session really changes.
Participation is the only thing a session really changes. Illustrative chart - not real market data.

That is the whole mechanism, and it is worth being precise about. A session does not make price more predictable. It changes how many people are present, which changes volume, spread, depth and typical range. Direction is not on any timetable.

What follows from participation

A calmly advancing stretch of the long price series. The headline on the chart reads: It matters most on currencies and least on US stocks.
It matters most on currencies and least on US stocks. Illustrative chart - not real market data.

Sessions matter most where the market never closes. Forex trades continuously, so the only thing distinguishing 3am from 3pm is who is awake. US equities trade in one session only, so the concept barely applies — there is no “London session” for a Nasdaq stock, just premarket.

A flat, quiet stretch of the long price series. The headline on the chart reads: And the hour before it is the quietest of the twenty-four.
And the hour before it is the quietest of the twenty-four. Illustrative chart - not real market data.

The hour before London opens is typically the quietest of the day — Asia is winding down and Europe has not started. Ranges are narrow, spreads are wide, and a break in those conditions is the easiest kind to manufacture.

A strongly rising stretch of the long price series. The headline on the chart reads: The Asian range is the level every London method uses.
The Asian range is the level every London method uses. Illustrative chart - not real market data.

Which is exactly why the Asian session range is the level every London method is built around. A quiet market builds a narrow range; a busy one breaks it. The London open breakout page covers what that trade is worth once the base rates are attached.

In practice

A declining stretch of the long price series. The headline on the chart reads: European data lands early in the session, not randomly.
European data lands early in the session, not randomly. Illustrative chart - not real market data.

European economic releases are concentrated in the early part of the session. So a substantial share of London-session movement is a reaction to scheduled information, which is a confound for anyone attributing results to the hour rather than to the calendar.

Splitting your own results by release day is a straightforward test and it is the one that separates trading a session from trading a data schedule.

A flat but volatile stretch of the long price series. The headline on the chart reads: The clock moves twice a year and not on the same dates.
The clock moves twice a year and not on the same dates. Illustrative chart - not real market data.

Daylight saving changes the alignment twice a year, and the UK, the EU and the US do not change on the same dates. For a few weeks annually the overlap window is an hour different from the one every article describes.

A long-horizon candlestick view of the same price series. The headline on the chart reads: On a daily chart the session is inside one bar.
On a daily chart the session is inside one bar. Illustrative chart - not real market data.

On a daily chart the session is inside a bar. Everything on this page is an intraday concern, and for a position held over weeks it is noise.

A candlestick series containing several opening gaps, with the largest opening gap marked. The headline on the chart reads: The weekend gap lands at the Asian open, not this one.
The weekend gap lands at the Asian open, not this one. Illustrative chart - not real market data.

The weekend gap happens at the Asian open, hours before London, so by the time Europe arrives the repricing has already been traded.

A declining stretch of the long price series, with the entry price and the level at which a stop would trigger drawn as horizontal lines. The headline on the chart reads: And a stop set in one session sits through the others.
And a stop set in one session sits through the others. Illustrative chart - not real market data.

A stop placed during London rests through New York and Asia, in conditions that are different in every respect. A stop distance chosen for a busy market is a different risk in a thin one.

A candlestick chart of the site's shared price history, annotated with the round-trip cost. The headline on the chart reads: Spreads narrow here, and the round trip is still a share of a bar.
Spreads narrow here, and the round trip is still a share of a bar. Illustrative chart - not real market data.

Spreads are at their tightest in this session, and the round trip is still 2% of a median bar’s range on this site’s shared history.

A 72-bar candlestick section of the shared price history. The headline on the chart reads: The book is deepest when the most people are awake.
The book is deepest when the most people are awake. Illustrative chart - not real market data.

Book depth follows the same curve. The order book is thickest during the overlap, which is the practical reason to trade then if you trade in size.

What the London session is not

It is not a directional bias. Nothing about the hour says which way price goes.

It is not a fixed clock. Daylight saving moves it, and the exact boundaries differ between sources.

It is not relevant to every instrument. For a stock that trades one session a day, there is no alternative session to compare it to.

And it is not separable from the data calendar unless you separate it yourself.

When it fails

A sideways, range-bound candlestick series. The headline on the chart reads: And on a quiet day the session is as flat as any other.
And on a quiet day the session is as flat as any other. Illustrative chart - not real market data.

Averages contain quiet days. “London is busier” is a statement about the mean, and a rule that requires a trade during the session will produce one on the days that had nothing in them.

The second failure is the holiday. European bank holidays empty the session while the rest of the world trades, and the usual expectations do not hold.

A third is the news confound. Attributing results to the hour when they came from a data release means your rule will stop working the moment the release schedule changes.

A fourth is applying it to the wrong instrument. Session structure on a US small-cap is a description of premarket illiquidity, not of a trading session.

And a fifth is the daylight-saving drift. A rule tied to a local clock silently changes what it does twice a year, and the results from those weeks are not comparable with the rest.

The original data

On this site’s shared 576-bar history, bar ranges span 0.17 at the tenth percentile to 1.10 at the ninetieth — a ratio of 6.5 — against a median of 0.493, and the round-trip cost of 0.0098 price units is 2% of that median bar and 45% of the smallest bar in the series. The figures are in research/series-measurements.json, produced by site/measure_series.py.

A 72-bar window of the shared price history, cut short at the decision bar. The headline on the chart reads: London has opened and nothing has moved. Wait?
London has opened and nothing has moved. Wait? Illustrative chart - not real market data.

That 6.5-fold range spread is what a session filter is actually exploiting. Trading only the busy hours means trading the larger bars, where a fixed cost is a small share of the opportunity rather than a large one — which is a real, arithmetic benefit that has nothing to do with prediction. Compute the average bar range by hour on your own instrument, and the right session for you falls out of the numbers rather than out of a convention named after a city.

Trading sessions is the parent page covering the whole day. Asian session builds the range London trades against. And New York session is the other half of the busiest overlap.

What I actually do

Sessions changed my results by changing when I was allowed to trade, not by making anything more predictable. Restricting myself to the hours with actual participation removed the trades I took in dead markets, and that was the whole improvement.

— Michael Whitman

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