How to Spot a Scam Gold EA Before It Drains You

Let me tell you something against my own interest.

I sell a gold EA. One product, $999, trades only XAU/USD on your own account. So when I tell you how to spot a scam forex EA, you should read it as the most dangerous kind of advice there is: the seller handing you the exact checklist to catch sellers like me lying.

I'm fine with that. Because I got wiped once. A martingale bot with a 95% win rate and a screenshot so clean it looked notarized. It made money for four months. Then one Tuesday on a gold spike it gave back fourteen months of "profit" in nine hours and kept going. I learned how this stuff works the expensive way. So here's the whole trick, no charge.

The MyFXBook magic trick: how "verified" results are manufactured

That "verified" badge does not mean what you think it means. It verifies that an account exists and that the trades on it are real. It does not verify that the account belongs to the person showing it to you, that it's the only one they ran, or that real money was ever at stake.

Here's the move. A vendor opens ten demo accounts. Same EA, same settings, slightly different start dates or a tiny tweak each. Markets are noisy, so a few of those ten will get lucky and ride a clean run. Nine of them blow up or chop sideways. The vendor deletes those nine and screenshots the survivor. Verified. Real trades. Real account. Total fiction — because you're seeing the winner of a lottery and never the graveyard behind it.

So when you look at a track record, ask the questions the screenshot can't answer. How long is it? Three months of a survivor proves nothing; eighteen months across a real drawdown proves something. Is it a live account with actual deposits, or a demo? Does the equity curve have ugly stretches, or is it a suspiciously smooth ramp? Real trading is jagged. A curve that climbs in a straight line is either curve-fit, cherry-picked, or about to detonate. Sometimes all three.

Fake myfxbook results aren't usually fake numbers. They're real numbers, ripped out of context, with the losers cropped off. That's the part nobody tells you.

Returns that should make you run

Let's talk about what's actually possible, because the number a vendor promises tells you almost everything before you read another word.

Realistic, on a strategy with real risk control, is 2-5% a month. Some months better, some months red. That's it. That's the honest band. It sounds boring because it is boring, and boring is the entire point — boring compounds, boring survives.

Anyone promising 30% a month with no losing weeks is showing you the survivor of that graveyard you'll never see. Run the math on the promise itself. 30% a month compounded is roughly 2,200% a year. If that were real and repeatable, the seller would not be emailing you a $999 PDF. They'd be running other people's billions and taking 20% of the upside. The fact that they're selling the map instead of walking it is the tell.

The same logic kills the "no losing weeks" claim. Markets don't hand you a strategy that never loses. They hand you a strategy whose losses you can survive. A vendor bragging about an unbroken win streak is either too new to have hit their bad week yet, or they're hiding it. Both should make you close the tab.

I tell my own buyers to expect a range, to expect red months, and to judge me over a quarter, not a Tuesday. If a number sounds like a dream, it's bait. (Past performance never guarantees future results — gold can gap, and any month can be a losing one.)

The real red flag isn't the upside they show — it's the drawdown they hide

This is the one I'd tattoo on people if I could.

Everybody stares at the return. Wrong muscle. The return is the marketing. The risk is the product. And the single biggest tell of a scam gold expert advisor isn't the size of the upside they advertise — it's the silence on the downside.

If a vendor won't lead with their worst drawdown, their hard stop per trade, and exactly what happens when they're wrong, walk. Not "ask twice." Walk. Because a real operator knows their worst day by heart. It's burned in. I can tell you mine without checking a file, the way you remember the day you crashed a car.

Here's why drawdown is the number that actually matters, and the math nobody on a sales page wants you to do. Recovering from a loss isn't symmetric. The formula is:

recovery needed = drawdown / (1 − drawdown)

• Lose 20%, you need +25% just to get back to even.
• Lose 50%, you need +100%. You have to double.
• Lose 90%, you need +900%. You're effectively dead.

A martingale bot looks unstoppable right up until the drawdown goes vertical, and then the recovery math becomes impossible in a single afternoon. That's what got me. The win rate was a magic trick to keep my eyes off the one number that mattered. A 95% win rate with a catastrophic 5% is a losing strategy wearing a winner's costume.

This is why I'd take expectancy in R over win rate every single time. Expectancy asks: across a hundred trades, what do I make on average per unit of risk? A strategy that wins 45% of the time but wins 2R when it's right and loses 1R when it's wrong is a money machine. A strategy that wins 95% of the time and loses everything on the 5% is a slot machine with extra steps. Win rate is the number scammers wave around precisely because it's the most misleading one. You can read more about how I think about this in how it works and see the real numbers on results.

Private method vs hidden risk: a legit vendor keeps one secret, never the other

Now, the objection I get thrown at me constantly: "If you're so honest, tell me exactly how the EA decides to trade."

No. And here's the line that matters, because scammers blur it on purpose.

There's a difference between the privacy of the method and the transparency of the risk. They are not the same thing. No real fund publishes its alpha. If I handed you the precise logic, it'd be copied, crowded, and dead inside a month — and you'd have paid $999 for something that no longer works. Keeping the edge private is what keeps it an edge. Any honest operator does this.

My edge, for the record, is a particular way of reading price itself, the numbers underneath it, and time — no indicators, no moving averages, no RSI, none of the public tools that lag because they're just price history rearranged and handed to everyone at once. That reading is the part I keep. Maybe one trader in a hundred ever works it out. That's mine to protect.

But risk? Risk is the part I owe you in full, before you pay a cent. The worst drawdown. The hard stop on every trade. The cap on risk per position. What happens on a gold gap. The refund terms. A scammer hides both — method and risk — and calls the whole black box "proprietary." A legit vendor keeps exactly one secret and is loud about everything that can hurt you. If someone won't separate those two things for you, they're counting on you not noticing the difference.

Lot sizes that grow after a loss: how to read a backtest for a hidden martingale

If a vendor sends you a backtest report — and you should always ask — open it and do one thing before anything else. Look at the lot sizes. Not the profit. The lot sizes.

Scroll to a losing trade. Now look at the very next trade. Did the lot size go up?

If it did, you may be looking at a hidden martingale, and that's the most dangerous bot on earth dressed as a steady performer. A martingale doubles (or scales up) after every loss, betting that the market has to turn eventually. And it does turn — almost always — which is exactly why these things show months of beautiful gains and a 90%+ win rate. The losses get papered over by ever-bigger winning recoveries. Until the one time the market doesn't turn fast enough, and the position size has ballooned so large that a single move wipes the account to zero.

So in any backtest, watch for these specifically:

• Lot sizes that increase after losing trades. The signature move. Honest position sizing is steady or scales with account equity, not with how recently you lost.
• A floating drawdown that's enormous even when the realized P&L looks calm. Martingales hold underwater positions and pray. Check the lowest the equity ever dipped, not just where it closed.
• A handful of trades doing all the damage. One or two monster losses among hundreds of tiny wins is the graveyard showing through.
• No hard stop loss on individual trades. If trades close at a profit or "eventually," but losers stay open indefinitely, there's no real stop. There's just hope and a margin call.

A backtest can be honest or it can be a costume. The lot-size column tells you which.

The questions a real seller answers instantly

Here's how you end the whole investigation in about ninety seconds. Send the vendor four questions and watch how fast they answer. Speed and specificity are the signal. Stalling, deflecting, or "it's proprietary" across all four is your answer.

1. What's your worst drawdown, and over what period? A real one fires back a number. A scammer changes the subject to win rate.
2. What's the hard stop on each trade, and the max risk per position? "There's a stop on every trade and risk is capped" should be instant. If there's no per-trade stop, that's a martingale or a prayer.
3. What happens when you're wrong — on a gold gap, on a bad week? They should describe the worst case plainly, not promise it can't happen.
4. What are the refund terms? Real skin in the game looks like a real guarantee.

So here are mine, since I'm asking you to hold me to the same bar. Every trade has a hard stop — no position runs naked. Risk per trade is capped, and there's a max-drawdown ceiling the system won't trade through. And there's a 30-day profit-or-refund: if you're not in net profit after thirty days, you get your $999 back in USDT. I lead with the worst day because the worst day is the product. The good days sell themselves.

That's the whole con, taken apart. The screenshot is theater. The dream return is a graveyard. The win rate is misdirection. And the one number that actually protects you — the drawdown, the stop, the worst day — is the one a scammer will never volunteer.

Make them volunteer it. If they won't, you already have your answer. And if you want to see what answering all four looks like before you pay, the terms are right there on the checkout page and laid out in plain language across the rest of the tools.

Privacy of the method, transparency of the risk. Never let anyone blur the two.

This is one trader’s opinion and education, not financial advice. Trading gold carries real risk of loss; any figures are illustrative and not a promise of results.