What Execution Speed Is Normal in MT5? Measured Latency Benchmarks
Updated 13-07-2026 · 6 min read · by the TradeStats team
"Fast execution" is on every broker's homepage, rarely with a number attached — and when there is one, it's a marketing benchmark, not your fills. Here is how execution speed actually breaks down, and what measured numbers from real accounts look like.
The three legs of latency
An order's life has three timed legs: your terminal → broker server (your internet plus distance to the server — tens to low hundreds of ms, a VPS near the broker shrinks it), broker processing (order received → routed → filled), and liquidity-provider responsefor brokers that pass orders through. Only the middle leg is the broker's own performance, and it's the one MT5's millisecond timestamps let you measure honestly: the server records both when it received your order and when it filled.
What measured numbers look like
Measured on real fills across connected accounts, well-run brokers fill liquid forex symbols in double-digit milliseconds median, and a few hundred milliseconds is nothing to worry about. The distribution matters more than the headline: a broker with a 40ms median and a 2,000ms p95 is slower where it counts than one with a 90ms median and a 300ms p95, because the slow tail concentrates exactly when markets move. The scoreboardshows both per broker, and each broker's execution-speed deep-dive breaks latency down by session and symbol.
Why the tail is where the money is
The p95 latency is the fill you get during the news candle — when your stop triggers, when the copy signal lands, when everyone else's orders are queued ahead of yours. A broker's median is measured in boredom; its p95 is measured under load. If you trade events, weight the p95 and the by-session table far more than the median.
Reading your own numbers
Your MT5 order history already contains the timestamps; a journal that parses them turns anecdote ("fills feel slow lately") into a chart. Connect an account read-only and your fills feed both your own analytics and the public per-broker measurements — the more accounts measure a broker, the harder its numbers are to dispute.
FAQ
What is a normal order execution time in MT5?
Broker-side, from order received to filled: double-digit milliseconds is common on liquid forex symbols at well-run brokers; a few hundred milliseconds is unremarkable. Consistent multi-second fills, or a p95 far above the median, are the red flags — check measured numbers per broker rather than a universal cutoff.
Does my internet connection affect measured execution speed?
Not the broker-side measurement. MT5's order history carries the server's own timestamps — order received vs order filled — so the metric isolates the broker's processing. Your network adds latency on top of that before the order reaches the server, which a VPS near the broker can reduce.
When does execution speed actually matter?
Scalping, news trading and copy trading — where prices move within hundreds of milliseconds. For multi-day swing trades the median barely matters, but a long p95 tail still hurts: stop-losses fill in fast markets, which is exactly when slow brokers are slowest.