TradeLocker Copy Trading: Five Things That Break First

TradeLocker looks like the easy one. There is no terminal to install, no EA to compile and no DLL to whitelist. You get an email, a password and a web page. Then you point a copier at it and discover that almost everything you learned from MetaTrader is wrong here, usually in the middle of a trade. This is what breaks, in the order most people hit it.

Why is copying into TradeLocker different from copying into MT5?

Because TradeLocker is not a venue. It is a platform that brokers and prop firms license and run under their own name, so HeroFX, FunderPro, Blue Guardian and dozens of others are all running their own instance of the same engine. Two accounts can both be "TradeLocker" and agree on almost nothing.

With MetaTrader you can lean on a lot of shared assumptions. Lot steps are usually 0.01. Symbol names are broadly familiar. A position is a position. None of that is safe here, because the broker configures it and the broker is different every time.

Each row below is one of those, with what it looks like when it goes wrong.

Break 1: the login fails and the password was never the problem

The server name is a third credential, not a setting. It identifies which broker or prop firm instance you are logging in to, and TradeLocker will refuse a perfectly correct email and password without it. The failure reads as a credentials problem, which sends most people off resetting a password that worked all along.

TradeLocker support is unusually direct about this. Its own documentation states that without the correct server, login will fail even if your email and password are correct, and that the server name is always provided by your broker or prop firm. TradeLocker does not create or manage server names and cannot tell you yours.

Three things to check before you blame the software:

• It is not your firm website address. It is an identifier like HEROFX, sent in the welcome email. Search your inbox for the firm name or for "TradeLocker account", and check spam.
• Copy it exactly. Treat it as case sensitive and watch for a trailing space picked up from an email client.
• Demo and live are different clusters. An account that exists on one does not exist on the other. Picking the wrong one produces the same unhelpful rejection.

If the server field in your welcome email is blank, only the firm can fix it. That is worth knowing early, because it is the one failure on this list that no software can work around.

Break 2: the order was rejected and the API said everything was fine

TradeLocker answers a rejected order with HTTP 200 and an error inside the response body, in the shape {"s":"error","errmsg":"..."}. Any software that trusts the status code will treat a margin rejection, a closed market or a forbidden order type as a completed copy, and then believe it holds a position that was never opened.

This is the quietest failure of the five and the most expensive. The copier reports a success. The dashboard shows a position. The account is flat. Nothing reconciles until the source exits and the exit has nothing to close.

Order validity is enforced on the server too, and gets refused the same way. A market order must be immediate-or-cancel and a limit or stop must be good-till-cancelled. Send the wrong combination and you get "TIF/Order type combination is forbidden for the route", again as a 200.

If you are writing against the API yourself, check the s field before anything else and never branch on the status code. If you are choosing software, this is a fair question to ask the vendor, because the answer tells you whether anyone tested against a rejection rather than only against a happy path.

Break 3: do TradeLocker accounts hedge or net?

They hedge. Measured on a live account on 19 September 2026, two same-side market orders on the same instrument produced two separate positions rather than one combined position. There is no single net position per symbol to read or to close.

That one behaviour changes how every exit has to work. On a netting account, closing a symbol is one instruction. On a hedged account, closing a symbol means enumerating the open positions on it and closing each one, and a partial exit means deciding which of them to take size from and in what order.

It also creates an identity problem. TradeLocker gives an order and the position it creates two different identifiers, and its own documentation confirms that when an order fills, a new position is created with its own positionId, different from the original orderId. So the response to your entry does not tell you what to close later.

Picking "the newest position on this symbol" is the obvious shortcut and it is wrong. When two orders land within the same second, which happens constantly with a copier feeding several accounts, newest-wins picks the other trade. Tagging each entry with a unique identifier and finding the position by that tag is the only version that stays exact.

Break 4: why a partial close takes off more than you asked for

A close request smaller than the instrument minimum lot is rounded up by the broker rather than refused. Measured on 19 September 2026, a request to close 0.005 of a 0.02 position closed 0.01, and the same request against a 0.01 position flattened it outright while answering with a success response.

Read that second case again, because it is the dangerous one. You asked to take a quarter off. The account went flat. The response said it worked.

On a scaled exit this is compounding. A source that takes profit in three pieces sends three partial closes. If the first one over-closes, the account is already out of the trade, and the second and third find nothing left to close. The copied account has exited early on a trade the source is still running, and every reconciliation afterwards is working from a position that no longer exists.

The safe behaviour is asymmetric, and worth understanding whichever software you use. Under-closing is self-correcting, because the source's final exit closes whatever remains. Over-closing is not, because there is no instruction later that puts the trade back. So a partial that would land below the minimum lot should be declined out loud rather than sent and rounded up.

This is also an argument for sizing sensibly in the first place. If your copied size is close to the minimum lot, partial exits were never really available to you on that account, and a strategy built on scaling out belongs somewhere else. The same logic applies to browser-driven prop platforms that only support full exits.

Break 5: the symbol on your chart is not the symbol on their book

Instrument names on TradeLocker are whatever the broker called them, and brokers routinely list the same instrument several times for their standard, raw-spread and ECN books. A source sending EURUSD can arrive at a firm whose only listing is EURUSD.r, and an exact-match lookup returns nothing.

Four kinds of mismatch show up in practice:

• Case. eurusd against EURUSD.
• Separators. EURUSD against EUR/USD or eur_usd.
• Book suffixes. EURUSD against EURUSD.r, EURUSD_raw, EURUSD.pro.
• Index and metal aliases. NAS100 against USTEC, US100 or NDX, depending entirely on who named it.

Suffixes deserve a specific warning, because several books can match one source symbol at once. If software resolves that non-deterministically, the same alert can land on the standard book today and the raw-spread book tomorrow, which changes the spread you pay without anything visible changing. Whichever tool you use, the resolution needs to be stable across restarts, not merely correct once.

The opposite failure matters more though. Guessing wrong opens a trade on an instrument you never asked for. Refusing the trade and naming the symbol that failed is the better outcome every time, even though it feels worse in the moment.

The rate limit that decides how many accounts you can copy into

TradeLocker limits requests per route, and the position and order endpoints are the tight ones. Two back-to-back position reads on a single account, which is exactly what a close does, returned HTTP 429 when measured on 19 September 2026. The 429 body is a different shape from a normal response, so software that only parses the success envelope cannot even see what happened.

The practical consequence is that every account you add contends for the same budget. Copying to one account is instant. Copying to five means the fifth one waits, and on a fast-moving entry that wait is slippage you did not choose.

Two things help. Reading the limits from the platform rather than hardcoding them, since they are published per route through the TradeLocker public API config endpoint, and throttling ahead of the limit rather than reacting to a 429 after an order has already been delayed. TradeLocker also runs a developer programme that issues a key with less restrictive limits for software acting on behalf of multiple accounts.

When you are comparing copiers, ask how many destination accounts they have actually run against TradeLocker at once. It is a more revealing question than the feature list.

What changes when the account is a prop firm evaluation

Every failure above gets more expensive on a funded account, because the account has rules attached and a breach is permanent. An over-closed partial is an annoyance on a demo and a missed profit target on an evaluation.

Two rule families interact badly with copying, and neither is a TradeLocker behaviour. The first is copy trading policy itself, which varies widely: some firms permit copying between your own accounts, some permit only an external source into one account, and some prohibit it outright. We cover where the line usually falls in our guide to what prop firms actually allow when copying between accounts.

The second is consistency rules, which cap how much of your total profit can come from your best day. Copying the same source into several accounts produces correlated best days across all of them, so a rule you would pass on one account can block payouts on several at once. That interaction is worked through in our breakdown of 2026 consistency rules.

Neither is optional reading. A copier does exactly what you configured, and a firm reviewing your account sees the pattern rather than the intent.

The short version

TradeLocker is a good destination for copied trades. It is quick to connect, there is nothing to install, and it is the platform behind a large share of the retail prop industry. It is also a white-label engine where the broker sets the rules, so the useful question is never "does this work with TradeLocker" but "does this work with my firm's TradeLocker".

Five things to check before a funded account is involved:

• Does it ask for the server name, and does it tell you clearly when the login is rejected?
• Does it read the response body rather than the HTTP status, so a rejection is not reported as a fill?
• Does it handle hedged positions by id, rather than guessing at the newest one?
• Does it decline a partial below the minimum lot instead of letting the broker round it up?
• Does it resolve your broker's symbol names, suffixed books included, and refuse rather than guess?

If a tool cannot answer those, it has been tested against one broker on a good day. There are at least 56 prop firms running TradeLocker, and they are not configured the same way.

Trade Dispensary runs on your own machine, copies from Telegram, MT5, TradingView, NinjaTrader, cTrader and TopstepX, and sends to TradeLocker alongside MT5, NinjaTrader, cTrader, DXtrade, TopstepX, Bybit and Propr. It is a one-time purchase with no monthly fee. Check whether your firm is supported, see the full feature list or buy a licence.

Before you connect anything to a funded account: leveraged trading carries substantial risk and losses can exceed what you expect. A copier multiplies the source in both directions, so a poor source gets worse rather than better. Run it on a demo or evaluation account until you have watched it handle a full trade cycle including a partial exit, and read your own firm rules on copy trading rather than trusting a snapshot of anyone else's.