Verification
How to Verify an Expert Advisor's Track Record: What Myfxbook and MQL5 Signals Do and Do Not Prove
By Benjamin Lee · Updated · 8 min read
Every Expert Advisor seller says their results are real. A public track record is the best evidence you can check, but a link is not proof, and a green badge is not a guarantee. A verified record can still hide a tiny starting deposit, a demo account, a cherry-picked time window or a different EA from the one you are buying. This guide shows what the main verification systems prove, the numbers to look at, the tricks that make a record look better than it is, and a checklist you can use on any account page.
Key takeaways
- A verified record proves the history matches the broker. It does not prove the strategy is good or that it is the EA you are buying.
- Check whether the account is real or demo, how old it is, how many trades it has, and what the drawdown was.
- Percentage gain can be inflated by a very small first deposit. Look at money earned and at deposits.
- Compare balance with equity. A smooth balance curve can hide a large floating loss.
- Your own forward test on demo is the final check, because it uses your broker and your settings.
What verification actually proves
Track-record sites do not trade for anyone. They connect to a trading account with a read-only investor password and copy the history they see. Myfxbook's help pages explain two separate checks. One compares the history shown on the site with the history held by the broker, so that it cannot be edited after the fact. The other confirms that the person who added the account controls it. Read Myfxbook's own help page for the current wording and the exact badges.
| It can show | It cannot show |
|---|---|
| The trade history matches what the broker reports | That the strategy will work in the future |
| When the account started and how many trades it has made | That the account is run by the EA you are buying |
| Drawdown, gain, win rate and average trade size | That your broker, spread and execution will give the same result |
| Whether the account is demo or real, if labelled | Whether the owner runs other accounts and shows only the best one |
On MQL5, the Signals service lists accounts that providers have chosen to publish, with statistics such as growth, drawdown, number of weeks, trades, profit factor, leverage and subscribers. The page states that historical statistics cannot guarantee any future profitability. It also shows a reliability score without defining it, so do not lean on a number you cannot interpret.
The numbers to check on any account page
| Item | What to check | Why it matters |
|---|---|---|
| Account type | Real or demo, and the broker and server name | A demo account has no real spread or slippage pressure |
| Age of the account | Months or years of history, not weeks | A short window may have been a calm patch for the strategy |
| Number of trades | Hundreds or more | A few dozen trades prove little about luck versus skill |
| Initial deposit and deposits later | The first deposit, and any top-ups | A very small first deposit can make the percentage gain huge |
| Gain versus profit in money | Both the percentage and the money earned | Percentages can mislead; money earned relative to deposits does not |
| Maximum drawdown | Balance and equity drawdown, and the date | Shows the worst experience, not the average |
| Balance versus equity curve | A gap between the two | A big gap means open losing trades being carried |
| Lot sizes over time | Constant, or growing after losses | Growing lots after losses are a martingale signature |
| Average trade length and symbols | Matches the EA's described style | A scalper with long holds is a different system |
| Leverage | The account's leverage setting | Leverage changes margin and the maximum position size |
| Stops used | Whether trades close at a fixed stop | No stops means open-ended risk |
Example: how a small deposit inflates the gain
Take two accounts. Account A starts with 100 euros and earns 200 euros, a gain of 200 percent. Account B starts with 10,000 euros and earns 2,000 euros, a gain of 20 percent. The headline favours A, yet B made ten times the money with a far larger position. An account that begins with a tiny deposit and is topped up later can show an impressive percentage while the money earned is small. Always compare the percentage with the amount earned and the deposit history, and be cautious of percentage figures without those.
Drawdown works the other way. A 50 percent drawdown needs a 100 percent gain to recover. If a record shows a high gain and a drawdown above 30 percent, ask whether you could sit through that loss with your own money. The drawdown calculator and our drawdown guide show how recovery scales.
Tricks that make a record look better than it is
- Many accounts, one shown. A seller can run ten accounts with different settings and publish only the one that did well.
- A good window. A record that starts after a profitable stretch, or ends before a loss, hides the bad period.
- Demo presented as live. Demo fills are friendlier than real ones, so check the account type.
- Deposits that look like profit. Balance jumps from deposits can be mistaken for gains. Check the deposit and withdrawal history.
- Smooth balance, bad equity. A system that keeps losing trades open shows a rising balance while the equity curve sinks.
- A different EA. The record may belong to a manually traded account, or to an older version of the EA. The seller should show which EA and setfile ran it.
- Cosmetic settings. A private or hidden account, or a profile with the stats page but no trade list, hides what matters.
Manipulated backtests are a separate problem with the same aim. See can EA backtests be manipulated for how to spot them, and how to spot forex EA scams for the wider pattern.
Does the record belong to the EA you are buying?
This is the question most buyers skip. A record shows an account. To connect it to a product, ask for the EA name and version, the setfile, the symbol and the timeframe, and check that the trade pattern fits: magic numbers are visible in the terminal but not usually on public pages, so look at trade times, lot sizes and the number of positions open at once. If the seller cannot show which settings ran the account, treat the record as marketing and rely on your own test. The MQL5 Signals list has a column for the Expert Advisors an account uses, which helps where the provider fills it in.
Run your own check
Step 1: Run the same settings in the Strategy Tester
Use real ticks and your broker's costs. If the seller's period and settings are published, test those exactly, as in how to backtest an EA in MT5.
Step 2: Compare the tester result with the track record
A big gap between your test and their live record tells you something has been left out, or that execution matters more than the tester shows.
Step 3: Forward test on demo
Run the EA for several weeks, through at least one news week. See forward testing on demo.
Step 4: Go live with the minimum lot
Compare fills and spread with the demo before you size up.
Step 5: Publish your own record
Connect your own account to a track-record site with the investor password. Your history then builds a record you can trust.
A five-minute walk-through of one record
Step 1: Open the account page and read the header
Note the broker, server, account type, leverage and start date. Real or demo is the first filter, and the age tells you how many market conditions it has seen.
Step 2: Look at the equity curve, not just the balance
A balance line that climbs while the equity line drops below it is carrying open losers. Zoom in on the worst point.
Step 3: Open the trade list
Sort by profit and loss. If one winning trade is most of the profit, the record depends on luck. If the lot size rises after losing trades, suspect martingale.
Step 4: Check deposits and withdrawals
Match balance jumps to deposits so you know what is profit. A late large deposit makes drawdown percentages look smaller than the earlier history.
Step 5: Compare with the EA's claims
Check symbol, holding time and trade count per week against the manual. Differences mean the account may not be running that EA.
Sources and further reading
- Myfxbook Help: VerificationHow a trading account's track record is verified on Myfxbook.
- MQL5 Signals: signal list and statisticsThe statistics MQL5 shows for each signal provider, including growth, drawdown, weeks and subscribers.
- MetaTrader 5 Help: Testing ReportWhat each figure in the tester's report means, including drawdown.
- CFTC Customer Advisory: AI Won't Turn Trading Bots into Money MachinesA US regulator's warning about trading-bot claims of unrealistically high returns.
Frequently asked questions
What does a verified Myfxbook account mean?
It means the history shown matches the history held by the broker, and, with a separate check, that the person who added the account controls it. It does not mean the strategy is good or will repeat. Read Myfxbook's help pages for the current definitions.
Can a Myfxbook or MQL5 record be faked?
Verification limits editing of a history, but a seller can still show only the best of several accounts, a demo account, or a favourable window. Look at the age, trades, deposits, drawdown and whether the EA used is named.
How long should a track record be?
Longer is better, and a record should cover both calm and volatile conditions. A few weeks tells you very little, and several months is the lowest range worth looking at, with years being much stronger.
What is the difference between gain and absolute gain?
Gain is a percentage that depends on how the platform compounds returns over time, so a small starting deposit can inflate it. Absolute gain relates profit to deposits. Compare both with the money earned, and read the platform's definition.
Is a demo track record worthless?
No, but it is weaker. Demo shows the logic works under test conditions, without real slippage and spread pressure. Treat it as a first step, then forward test on your own broker.
Should I trust a record without a link?
No. A screenshot can be edited or taken from the best of several accounts. Ask for a link to an independent track-record page that shows the broker, the start date and the full trade list.
Expert Advisors in these styles
Apply the checks above to any of these before you buy: read the manual, run your own backtest, and start on demo.
Keep reading
- Buying guideEA Buying Checklist: 15 Questions to Ask Before You Pay15 questions to ask before buying an MT5 Expert Advisor: who built it, licence terms, backtest proof, risk logic, support and refunds, with a printable checklist.
- BacktestingHow to Read an MT5 Backtest ReportLearn to read an MT5 Strategy Tester report line by line: drawdown, profit factor, Z-score, trade quality, robustness checks and the red flags that expose a weak EA.
- Choosing a methodEA vs Copy Trading vs Signals: Which Is Right for You?Compare automated trading with an EA, copy trading and manual signals: control, cost, delay, risk and who each suits, with a decision table and checks.
- Pillar guideMT5 Expert Advisors: The Complete Guide for BeginnersEverything about MetaTrader 5 Expert Advisors in one place: how they work, strategy types, risk, backtesting, buying safely, gold, EURUSD, prop firms and Europe.