Hosting next to the broker's server rather than next to the session is the right principle, and it answers my question better than any speed test would.
One correction on the figure: London to New York is roughly 35 ms each way, so an order plus its confirmation costs around 70 ms round trip. Still irrelevant for most discretionary M1 tickets, but it matters to anyone cancelling resting orders close to a print.
Before assuming LD4, I'd also check where the account's server really sits. Several retail brokers keep some servers in NY4 and others in London, depending on platform and entity. A cheap VPS in each location, pinging for a day, settles it for a few pounds.
In my own journal, the NY-open-only weeks showed no measurable slippage difference between London hosting and home fibre. The leak was the pre-overlap routine, as I suspected.
Is a London VPS worth it if you only trade NY open?
-
LondonScalper
- Posts: 965
- Joined: Sat Sep 05, 2026 7:54 am
-
FXS Prop Desk
- Posts: 1689
- Joined: Wed Oct 07, 2026 9:52 am
Re: Is a London VPS worth it if you only trade NY open?
That's the key question. An easy way to find out is to ping the broker's trade server from a few cloud regions and compare. If latency is lowest from London, the broker's server is probably in LD4 regardless of which session you trade. For discretionary M1 trading, stability and uptime of the VPS matter more than shaving a few milliseconds.PTScalper wrote:Only if your broker's matching engine is actually in New York.
For traders on a tight budget, a home connection with a reliable fibre line and a wired connection (not Wi-Fi) is often enough for discretionary M1 trading.
FXS Official Team | The Idea Lab & Strategy