The last two hours of New York on EURUSD
After London closes, EURUSD changes character. Roughly from 16:00 London time through to the New York close, volume drops off, and the pair tends to drift.
Some people love this window. It's calm, there's less noise from news, and if you like slow mean reversion it can suit you. Others, me included most days, find it hard to trade well. The moves are small compared with the spread on a lot of accounts, especially closer to the 17:00 New York rollover when spreads start widening.
Things I've noticed. The London close itself can produce a last push, sometimes a move toward the day's high or low, then a fade. After that, levels from the day act more like soft magnets than hard walls. Breakouts in this window rarely get much follow through unless there's US news or a Fed headline.
So if I trade it at all, it's fades back toward the day's midpoint or VWAP style area, with small targets and a hard cutoff well before rollover. And I'm fine going flat instead. Late session trades are where I've noticed I get lazy with rules, because it feels low stakes.
If your best hours are London, this is a reasonable time to just stop.
The last two hours of New York on EURUSD
The last two hours of New York on EURUSD
It’s Fairman 
Re: The last two hours of New York on EURUSD
Fridays are different here. The last couple of hours on a Friday can have position squaring into the weekend, and if there's been a big move during the week, you sometimes get a sharp partial retrace late in the day. It isn't predictable enough to trade on its own, in my view, but it's a reason not to assume the late session will be quiet just because it usually is. Month end Fridays can be even livelier with rebalancing flows. So on a Friday I'm more likely to just be flat by the London close and not bother with the last stretch.
It’s Fairman 
Re: The last two hours of New York on EURUSD
For anyone who trades this window from Asia or Australia, it overlaps with your early morning, which might be the only time you can trade. That's a fair reason to use it. In that case I'd look at USDJPY or AUDUSD rather than EURUSD, since they pick up a bit as Asia comes online. EURUSD has the least going for it then. Either way I'd keep an eye on the spread right through until Tokyo is properly open. Late NY spread on EURUSD can widen a lot near 17:00 New York, and that's exactly when Asia traders are starting their morning.
It’s Fairman 
Re: The last two hours of New York on EURUSD
Month end can make this window unusual. On the last trading day of the month, fixing related flows around 16:00 London are done, but some rebalancing still trickles through into the New York afternoon. It's not predictable in direction, and you sometimes see a sudden move with no news at all. I'd treat month end and quarter end afternoons as different from normal days, and maybe step back, even if the rest of the month this window has been quiet and tradable. The same goes for the days around US holidays, when the afternoon can go flat early and then jump on a single order.
It’s Fairman 
-
FXS Prop Desk
- Posts: 1689
- Joined: Wed Oct 07, 2026 9:52 am
Re: The last two hours of New York on EURUSD
The late New York drift can suit mean reversion well, as you say. The one risk to watch is rollover around 23:00 Prague time, when spreads widen sharply. Any trade still open at that point can be stopped out by the spread alone.
Volume is also very low after about 21:00, so a single large order can move the price more than usual. Smaller targets and a hard stop before rollover make this window safer.
Volume is also very low after about 21:00, so a single large order can move the price more than usual. Smaller targets and a hard stop before rollover make this window safer.
FXS Official Team | The Idea Lab & Strategy
-
Shadow Trader
- Posts: 5594
- Joined: Wed Sep 09, 2026 5:19 pm
Re: The last two hours of New York on EURUSD
After 4pm New York, the euro and sterling pairs lose steam. AUD and NZD pairs keep a little more life, since Sydney is getting ready and the antipodean currencies have their own buyers and sellers.
The result is a pocket of activity around 5 to 8pm New York where AUDUSD and NZDUSD move in cleaner, if smaller, ranges than the rest. Spreads are wider than at peak, but not extreme.
The behavior is different from the European session. There are fewer big banks, so moves are driven by smaller orders and by anything coming from Asia. Chinese equities, commodity prices, and early data releases play a bigger role.
There's a trap. The calm can end abruptly. A single headline from Asia or a move in the Japanese yen can cause a sudden jump, and the thin book makes it overshoot.
A scalper working this window needs to accept smaller targets and wider stops, or skip it. Treating it as a miniature London session is the common mistake.
Is there anything repeatable in the 5pm to 8pm New York window on AUD and NZD pairs?
The result is a pocket of activity around 5 to 8pm New York where AUDUSD and NZDUSD move in cleaner, if smaller, ranges than the rest. Spreads are wider than at peak, but not extreme.
The behavior is different from the European session. There are fewer big banks, so moves are driven by smaller orders and by anything coming from Asia. Chinese equities, commodity prices, and early data releases play a bigger role.
There's a trap. The calm can end abruptly. A single headline from Asia or a move in the Japanese yen can cause a sudden jump, and the thin book makes it overshoot.
A scalper working this window needs to accept smaller targets and wider stops, or skip it. Treating it as a miniature London session is the common mistake.
Is there anything repeatable in the 5pm to 8pm New York window on AUD and NZD pairs?
Re: The last two hours of New York on EURUSD
Something that catches people here is the spread widening that starts before the daily rollover. On many brokers the spread on EURUSD begins to widen in the last half hour before 17:00 New York, and if you're in a trade with a tight stop, it can be hit on the spread. If you trade the last two hours, closing positions a bit before the rollover, or at least moving stops away from the spread risk, avoids a lot of these. The rollover itself rarely moves price, but the spread does.
It’s Fairman 