Best Time to Trade Forex: Session Hours and How to Find Yours

The best time to trade forex, if you want one answer, is the London and New York overlap: 08:00 to 12:00 New York time. That is 13:00 to 17:00 in London. Both of the largest FX centers are open, so liquidity peaks and spreads tighten.

The busiest window is not automatically the most profitable one for your strategy. That part you measure from your own trades, and this guide shows how.

Forex market hours: the four sessions

The forex market runs 24 hours on weekdays, handed off between four financial centers. Each session is a business day in its own city. This is a retail convention, not an exchange rule, and sources vary by an hour.

Session	Local hours	UTC in January	UTC in July
Sydney	08:00 to 17:00 Sydney time	21:00 to 06:00	22:00 to 07:00
Tokyo	09:00 to 18:00 Tokyo time	00:00 to 09:00	00:00 to 09:00
London	08:00 to 17:00 London time	08:00 to 17:00	07:00 to 16:00
New York	08:00 to 17:00 New York time	13:00 to 22:00	12:00 to 21:00

BacktestFlow files trades into sessions with these same hours. Two things about them trip people up.

These hours move. Sessions follow local business hours, not fixed UTC offsets. Sydney, London and New York change their clocks on different dates, and Tokyo never changes. Sydney’s summer runs the opposite way: in 2026 its clocks went back on April 5 and forward on October 4.

In 2026 the US moved on March 8 and the UK on March 29. For those three weeks, and again from October 25 until November 1, the overlap runs 08:00 to 13:00 New York time.

The overlap is the real event. UK and US dealing desks handled about 38% and 19% of global FX turnover in April 2025, according to the BIS Triennial Survey. Together that is 57% of the market. The overlap is the only stretch of the day when both centers are open at once.

Why everyone recommends the overlap

The standard advice isn’t wrong. During the London and New York overlap:

• Liquidity is highest, so orders fill closer to the price you wanted
• Spreads are tightest, particularly on EUR/USD and GBP/USD
• Moves are larger, so a trade with a sensible stop has room to reach a target
• News lands here: US jobs and inflation data come out at 08:30 New York time, half an hour into the overlap

If you trade breakouts, momentum or news, this is a reasonable place to start.

Where the standard advice breaks down

Many guides on this question end with the same hedge. The best session, they say, depends on your strategy, your pairs and the hours you can be at the screen.

That hedge is the true part of those guides. It’s also where the advice stops being useful. It hands the question back to you without a way to answer it.

Consider what the “just trade the overlap” recommendation assumes:

• You trade breakouts. If you trade mean reversion with tight stops, the overlap’s volatility is what stops you out. The quiet Tokyo session may suit you because it moves less.
• You trade EUR/USD or GBP/USD. If you trade AUD, NZD or JPY pairs, Asian hours are when their home markets are open. The overlap is not automatically your window.
• You can be at your desk at 08:00 New York time. A strategy you can only trade in theory has an expectancy of zero. Testing hours you won’t trade live is one reason backtest results don’t match live trading.
• Volatility helps you. For some setups, more movement simply means more false signals.

The general answer describes the average trader. You don’t have to guess: you have your own trade history.

How to find your best trading session

The method takes five steps:

1. Record the entry time of every trade in one consistent timezone. If you backtest a strategy by hand, log it as you go.
2. Tag each trade with the session it opened in: Sydney, Tokyo, London, New York or the overlap.
3. Compare expectancy per session, not win rate alone. Expectancy is your average result per trade in R, wins and losses included. With 1R losses, a 40% win rate at 3R earns +0.6R per trade. A 70% win rate at 0.5R earns +0.05R, so win rate alone points you the wrong way.
4. Check your sample size before believing any of it (see below).
5. Cut or reduce the sessions that lose, then check whether the improvement holds on new trades, not only in the historical slice.

“What time should I trade?” has no single answer. “Which session pays for my strategy?” does, and your own trades hold it.

The part most traders skip: sample size

Here’s the trap. You break your trades down by session and see a 60% win rate in London against 45% in New York. You conclude London is your session.

With 20 trades in that London bucket, 12 wins give a 95% confidence interval of roughly 39% to 78% (Wilson score interval). That range includes 45%. At a 1:1 payoff, anything under 50% loses money. You haven’t discovered anything yet: you’re looking at noise with a decimal point on it. You can check the range for your own numbers.

This is a common way traders draw confident conclusions from data that can’t support them. Splitting your history into four or five session buckets divides your sample by four or five. Small buckets produce dramatic differences from pure chance. The full math is in how many trades you need to backtest a strategy.

Before you act on a session breakdown:

• Look at the number of trades per bucket, not just the percentages
• Prefer expectancy or average R over win rate
• Be suspicious of any bucket with fewer than a few dozen trades
• Ask whether the difference would survive if you removed your two best trades

How BacktestFlow measures your sessions

BacktestFlow is built around this problem: turning your own trades into statistics you can rely on.

• On the Pro and Max plans, performance breaks down by market session, plus hour, weekday, pair and direction. A trade opened in the overlap counts in both London and New York.
• Each session is set in its own city’s time, so the buckets follow every clock change without conversion. Multi hour candles sit on the DST aware 17:00 New York grid, so H4 matches tradingview.com.
• The open markets readout shows which of Sydney, Tokyo, London and New York are open at the chart’s moment. Each trade is filed by its entry time on the replay clock.
• Every trade is recorded before the replay moves on. Each close opens a write up, and the replay waits until it is saved. No session quietly loses the trades you’d rather forget.
• Statistics in R, %, money or pips, so you compare sessions in the unit you use for risk.
• Flow Coach names a session only when the numbers allow it. The session needs at least 8 trades. Its expectancy must differ from your overall expectancy by more than chance allows, at 90% confidence. The first report unlocks at 35 backtest trades.

See performance by trading session alongside the other metrics BacktestFlow computes, more than 40 in five sections.

The short version

The London and New York overlap, 08:00 to 12:00 New York time, is the highest liquidity window in forex. It’s a sensible default if you trade breakouts or momentum. But “best session” is a question about your strategy, and it has a measurable answer. Tag your trades by session, compare expectancy rather than win rate, and check your sample size before you trust the result.

---

Nothing here is financial advice. Backtested performance does not guarantee future results.