How to Test Your VPS Latency to Your Broker (2026 Guide)

The latency number shown in MetaTrader 4 or MetaTrader 5 is not your real broker RTT. MT4 and MT5 are trading platforms, not network measurement tools. The figure displayed in the bottom-right corner of the terminal is an application-layer estimate, and it regularly underreports or misrepresents actual round-trip time to the broker’s execution server.

To make accurate infrastructure decisions, you need to measure latency at the network layer, from the VPS itself, using the correct port. A ping from your home PC to a broker hostname tells you nothing about what your Expert Advisor is actually experiencing.

This guide covers four methods:

• ICMP ping
• TCP port testing with Psping
• Hop-by-hop path analysis with tracert and pathping
• Sustained monitoring with WinMTR

Each method serves a different diagnostic purpose. Use all four before concluding your VPS location or broker server selection.

What Do You Need Before Running Any Latency Test?

Before running any test, you need three pieces of information. Without them, you will measure the wrong host, the wrong port, or the wrong path.

1. Your Broker’s Trading Server Hostname or IP

This is not the broker’s marketing website. It is the actual execution server your trading platform instance connects to. You can find it in one of two ways:

• In MT4/MT5: Go to File > Open an Account > View the server address displayed in the login window. Example: ICMarketsEU-MT5-Live01.gbm.network
• In cTrader: Navigate to Settings > Connection > Server address field

Do not use the broker’s general website domain for latency testing. Trading platforms connect to dedicated execution servers that may be in different data centers than the marketing site.

2. The TCP Port Your Broker Uses

Most retail forex brokers use one of three port configurations:

• Port 443 (HTTPS/TLS): the most common default for MT4 and MT5 in 2025–2026
• Port 1950: an older MetaQuotes standard used by some MT4 brokers
• Port 443: frequently used by cTrader and NinjaTrader as well

If you do not know your broker’s port, check their technical documentation or ask support. The port matters because broker firewalls commonly block ICMP (standard ping) while permitting TCP on the trading port.

According to Cloudflare’s documentation on ICMP, ICMP Echo Requests are frequently blocked by financial infrastructure firewalls as a standard security posture, which means TCP-based testing is required for accurate latency measurement on broker servers.

3. A Remote Desktop Session Open to the VPS

All tests must be run from the VPS via Remote Desktop Protocol (RDP), not from your local computer. Your Expert Advisor runs on the VPS. Only the VPS-to-broker path determines execution quality.

What are the Four Methods With Which to Test Your VPS Latency to Your Broker?

Method 1: Use ICMP Ping to Measure Baseline RTT to Your Broker

ICMP ping is the fastest way to establish a baseline round-trip time and identify obvious connectivity problems. Use this as your first diagnostic step. If ping times out, it does not mean the broker is unreachable. It means ICMP is blocked, which is common on financial server firewalls. Proceed to Method 2 (Psping) immediately.

Use this when: You want a quick first measurement and to confirm basic connectivity to the broker’s hostname.

How to Run ICMP Ping from a Windows VPS

Open Command Prompt on the VPS (Win + R, type cmd, press Enter).

Basic continuous ping (Ctrl+C to stop):

ping -t BROKER_HOSTNAME_OR_IP

Example: ping -t ICMarketsEU-MT5-Live01.gbm.network

Extended count ping (200 samples for statistical reliability):

ping -n 200 BROKER_HOSTNAME_OR_IP

How to Interpret ICMP Ping Output

Limitations of ICMP Ping for Broker Latency Testing

ICMP ping measures a different network path than the one your trading platform uses. Your trading platform connects via TCP on a specific port. Broker infrastructure is commonly configured to deprioritize or block ICMP responses while maintaining normal TCP trading connections. A ping timeout does not mean your execution latency is high. It means the firewall blocks this diagnostic protocol. Use Method 2 for accurate production-equivalent latency measurement.

Method 2: Use Psping to Test TCP Port Latency to Your Broker

Psping is the correct primary latency measurement tool for forex VPS testing. Psping measures TCP round-trip time on the actual port your trading platform uses, bypassing ICMP blocking and generating a statistically meaningful sample. Download Psping from Microsoft Sysinternals at learn.microsoft.com/en-us/sysinternals/downloads/psping. Psping is a free, signed Microsoft utility.

Use this when: You want accurate, statistically robust TCP latency measurement on the production trading port. This is the primary method for evaluating VPS location and broker server selection.

Installation and Setup

1. Download the Sysinternals Suite (or Psping individually) from the Microsoft Sysinternals page at learn.microsoft.com/en-us/sysinternals/downloads/psping
2. Extract the psping.exe file to a permanent directory such as C:\Tools\psping.exe
3. Accept the license agreement on the first run: psping.exe -accepteula

Core Psping Commands for Broker Latency Testing

Basic TCP latency test (100 samples, IPv4 forced):

psping.exe -4 -n 100 BROKERIP:PORT

Example: psping.exe -4 -n 100 198.51.100.45:443

Continuous TCP latency test with interval (runs until Ctrl+C):

psping.exe -4 -t BROKERIP:PORT

Extended 500-sample test for statistical reliability:

psping.exe -4 -n 500 -i 0.1 BROKERIP:PORT

Saving output to a file for later analysis:

psping.exe -4 -n 200 BROKERIP:PORT > C:\latency_test_results.txt

Parameters explained:

• -4 — Force IPv4 to prevent IPv6 routing which may follow a different path
• -n 100 — Send 100 test packets
• -t — Run continuously until stopped with Ctrl+C
• -i 0.1 — Send one packet every 100 milliseconds (10 per second)

Key Output Metrics from Psping

Method 3: How Do You Use Tracert and Pathping for Hop-by-Hop Broker Latency Analysis?

Tracert (traceroute) and pathping reveal which specific network hop is contributing the most latency to the VPS-to-broker path. This is a diagnostic tool, not a production measurement. Use tracert and pathping when Psping shows unexpectedly high RTT, and you need to identify where in the routing path the latency originates. Microsoft’s Windows command documentation covers both tools; tracert and pathping.

Use this when: You have unexpectedly high Psping RTT and want to identify the specific routing hop causing the problem. Also useful when comparing VPS providers to understand the physical path to the broker’s data center.

Tracert Commands

Standard tracert (30 hops maximum):

tracert BROKER_HOSTNAME_OR_IP

Extended tracert with higher hop limit and forced IPv4:

tracert -4 -h 50 BROKER_HOSTNAME_OR_IP

Parameters:

• -h 50 — Allow up to 50 hops (default is 30)
• -4 — Force IPv4

Pathping for Combined Path and Packet Loss Data

Pathping combines the path display of tracert with per-hop packet loss statistics over time. Pathping sends 100 packets to each hop, making it more diagnostically accurate than tracert for identifying the source of packet loss:

pathping -4 BROKER_HOSTNAME_OR_IP

Pathping runs in two phases: first, it discovers the path (like tracert). Then it sends 100 test packets to each hop and calculates the loss percentage. The entire test takes several minutes to complete.

How to Interpret Tracert Output

Method 4: How Do You Use WinMTR for Sustained VPS-to-Broker Latency Monitoring?

WinMTR provides a real-time, continuously updating view of per-hop latency and packet loss over time. WinMTR is the most comprehensive single tool for sustained network monitoring from a Windows VPS. Download WinMTR from the official GitHub repository at github.com/White-Tiger/WinMTR. It requires no installation and runs as a standalone executable. WinMTR is the Windows equivalent of the mtr command on Linux systems.

Use this when: You want to monitor latency and packet loss over an extended period (30+ minutes), particularly during live market conditions. WinMTR is the best tool for detecting intermittent packet loss that a short Psping test would miss.

Setup and Configuration

1. Download WinMTR from github.com/White-Tiger/WinMTR
2. Extract and run WinMTR.exe — no installation required
3. In the Host field, enter the broker’s hostname or IP address
4. Set the interval to 100ms (10 packets per second) for market-hour testing
5. Click Start and let WinMTR run for a minimum of 30 minutes during active trading hours

Key WinMTR Metrics

Exporting WinMTR Data

Click Export to TEXT to save a copy of the full per-hop data table. When reporting issues to your VPS provider, always include the WinMTR export plus the specific time window of the test, the broker server hostname tested, and the trading session during which the issue was observed.

What Is an Acceptable VPS-to-Broker Latency for Different Trading Strategies?

Latency thresholds vary significantly by trading strategy.

The IDCInfoBrief commissioned by Equinix confirms that physical proximity to execution venues is the dominant variable in financial network latency. A VPS co-located in Equinix NY4 with a broker whose matching engine is on the same campus achieves sub-1ms consistently. A VPS in a generic US-East data center 50 miles from the broker may deliver 20–80ms regardless of hardware quality.

How to Interpret Your Latency Test Results

After running all four methods, use this benchmark table to evaluate your VPS-to-broker latency position:

For ForexVPS.net customers, use the broker latency tool to identify which of the 22 global locations delivers the lowest RTT to your specific broker before running a test from the provisioned VPS.

Frequently Asked Questions About How to Test Your VPS Latency to Your Broker

Here are answers to sme frequently asked questions about testing your VPS latency to your broker:

Why Does My Ping Time Out Even Though MetaTrader Is Connected?

Broker firewalls frequently block ICMP Echo Requests (ping) as a security measure, even when the trading platform connects normally. The trading platform uses TCP on a specific port (typically 443 or 1950), not ICMP. ICMP blocking is a standard financial infrastructure security posture, as noted in Cloudflare’s ICMP documentation. Use Psping in TCP mode (psping.exe -4 -t BROKERIP:PORT) to measure latency when ICMP is blocked.

Should I Test Latency from My Home PC or from the VPS?

Always from the VPS. Testing from your home PC measures the path from your ISP to the broker, which is irrelevant to trading performance. Your Expert Advisor runs on the VPS, so the only path that matters is from the VPS datacenter to the broker’s execution server. A test from your home PC can show 5ms while the VPS is experiencing 80ms, or vice versa. Per MetaQuotes’ VPS documentation, VPS-to-broker latency is the only metric that affects order execution.

How Often Should I Test My VPS-to-Broker Latency?

Run a full test (Psping plus WinMTR) at least once per month and after any change to your VPS plan, broker server, or trading platform configuration. Also, run tests immediately if you notice unusual slippage, order rejections, or disconnections during live trading. Latency can degrade due to routing changes at transit carriers, which are outside both the VPS provider’s and broker’s direct control.

Does Packet Loss Matter More Than High Latency?

For EA trading, yes. Consistent low latency with 0% packet loss is preferable to low average latency with any packet loss. According to Cloudflare’s packet loss analysis, TCP retransmissions triggered by packet loss introduce delays equal to multiples of the base RTT. A consistent 15ms RTT with 0% loss outperforms a 5ms average with 0.1% packet loss for EA strategies executing at high frequency.