Best Hours to Trade EURUSD for US Traders: Measured in New York Time

If you are trading EURUSD from the United States, the window that matters is 8:00 to 11:00 am Eastern. Those three hours carry the widest ranges of the whole 24-hour cycle, because the London session is still open while New York opens on top of it. Everything after 12:00 pm ET is progressively thinner, and the evening hours are the worst payoff for your attention on the entire chart.

Most session articles give you a coloured graphic of four overlapping bars and no numbers. Here is the same question answered by counting every bar of the last three months.

Average 5-minute range by hour

Real EURUSD 5-minute bars, 16,926 of them, June 3 to August 26, 2026. For every hour I took the average high minus low of its bars, in pips.

The 5 pm spike is a trap, not an opportunity

One bar breaks the pattern: 5:00 pm ET shows 2.69 pips, as wide as the morning. That is the daily rollover, when liquidity providers step back and spreads widen. The range is real but you cannot trade it at a normal cost, and a stop sitting in that window can be taken out by a wick that no one traded through. I flatten before it.

Day of week matters as much as hour

Thursday gave 40% more room than Tuesday for exactly the same screen time. If you can only trade three days a week, the data says Wednesday, Thursday and Friday.

How to turn this into a schedule

• Prime window: 8:00 to 11:00 am ET. Full position sizing, your best setup only.
• Second window: 3:00 to 5:00 am ET if you are up for the London open. Wider than the US afternoon.
• Do not trade: 12:00 pm to 8:00 pm ET on EURUSD. Half the range, same spread.
• Flat by 4:55 pm ET to stay out of the rollover.
• Pair this with a session filter in your rules rather than willpower. My RSI session test showed the same indicator earning twenty times more per trade inside the right window.

Frequently asked questions

Do these hours change with daylight saving?

The UTC hours shift, the New York hours do not. This table is built in New York time for that reason: the US session anchors to the equity open, not to the clock in London.

Is a 2.9 pip average range enough to scalp?

That is the average range of a single 5-minute bar. Over a 30-minute trade it compounds into something workable. What it tells you is the ratio between movement and spread, and that ratio is three times better in the morning than in the evening.

Does this hold for gold and the indices?

Gold overlaps but peaks later, and the indices are anchored to the cash session. Same measurement, different instrument, and it deserves its own article rather than a guess here.

Why is Sunday missing?

The Sunday session is a few hours long and averaged 6 pips, which is not a trading day, so I left it out of the comparison.

Three months, is that enough?

For the shape of the day, yes, the pattern is the same in every month of the sample. For the exact pip figures, treat them as this summer’s numbers, not as constants.

Method and data. EURUSD 5-minute OHLC, 16,926 bars over 73 calendar days, June 3 to August 26, 2026, retrieved from Yahoo Finance on August 26, 2026. Hourly figure is the mean of high minus low across all 5-minute bars in that hour, converted to pips at 0.0001. Daily range is the high to low of each calendar day. New York time is UTC minus 4 for this period. The script is kept with the article.