How to Trade Gold (XAU/USD): A Founder's Honest Guide

I blew up my first real account on gold. Not a demo. Real money, money I'd saved, gone in about three weeks. The bot I was running had a 95% win rate. Sounds incredible until you understand that the other 5% was a martingale stack that doubled into a single move and erased eight months of green in one Tuesday afternoon.

So when I tell you how to trade gold, I'm not reading it off a course. I learned it the expensive way. This is the guide I wish someone had handed me before I clicked buy the first time.

Let me start with the thing most articles bury: if you can't say your worst-case loss out loud before you enter, you're not trading. You're gambling with extra steps.

What XAU/USD really is (and why gold trades like its own animal)

XAU/USD is the price of one ounce of gold quoted in US dollars. "XAU" is the market ticker for gold, "USD" is what you're paying with. When you see 2,400, that means one ounce costs $2,400. Simple enough.

What's not simple is how it moves. Gold is not a stock. A stock has earnings, a CEO, a product. Gold has none of that. It's a 5,000-year-old store of value that the whole planet watches at once, which makes it react to fear, interest rates, the dollar, and a hundred macro headlines that have nothing to do with any company.

That's why gold trades like its own animal. It can sit dead flat for hours, then rip 30 dollars in fifteen minutes when a US data release hits. It loves the London-New York overlap and gets thin and jumpy in the dead Asian hours. The spread (the gap between buy and sell price) on XAU/USD is wider than EUR/USD, and it punishes overtrading.

One more thing nobody tells beginners: gold moves in dollars and cents, but your profit is measured in pips and lot size. A "pip" on most brokers is a 10-cent move. We'll get to the math, because the math is the whole game.

The anatomy of one gold trade: entry, hard stop, position size, exit

Here's how a single trade is actually built on your own MetaTrader 5 account. Four parts. Skip any one and you're flying blind.

Entry. The price where you get in. Buy if you expect gold to rise, sell if you expect it to fall. You can short gold as easily as you buy it. That's the point of a CFD or spot account.

Hard stop. The price where you're wrong and you get out. Automatically. This is non-negotiable. You set a stop-loss order at the same moment you enter, not after, not "I'll watch it." The stop is the only thing standing between a bad trade and a blown account.

Position size. How many lots you trade. This is where most people destroy themselves. On gold, one standard lot is 100 ounces, so a $1 move is roughly $100 of profit or loss. A micro lot (0.01) is about $1 per dollar move. Your size should be calculated backwards from your stop, never picked because it "feels right."

Exit. Where you take profit, or where your stop catches you. A trade isn't a position, it's a complete plan with both ends defined before you commit.

Want to see the size math worked properly? I lay it out step by step in how it works. The short version: pick your risk in dollars first, then let the stop distance decide your lot size. Most people do it backwards and wonder why one trade hurts so much.

Risk note: gold can gap through a stop on news or weekend opens, so your actual loss can exceed the planned amount. Hard stops cap risk, they don't guarantee the exact fill.

Why I read price, the numbers, and time instead of indicators

Here's where I'll lose some people, and I'm fine with that.

No moving average ever saved me. No RSI, no MACD, no Bollinger band, no ATR. I tried all of them. I stacked them on charts until the price was barely visible under the spaghetti. And every single one of them did the same thing: confirmed what had already happened.

That's the problem with public indicators. They're math done on past prices. A 50-period moving average is, by definition, an average of the last 50 candles. It's a rear-view mirror with extra steps. By the time it "crosses," the move it's describing is already old. Lagging tools tell you where price was, dressed up to look like where it's going.

So what do I actually read? Three things: price, the numbers underneath it, and time.

Price is honest. It's the only thing on the chart that isn't a calculation of something else. The numbers underneath it are the structure most traders never bother to see, the levels where the market keeps making decisions. And time is the part almost everyone ignores entirely. Gold behaves differently at 3am than it does at the New York open. The same setup at the wrong hour is a different trade.

I'm not going to hand you my exact method. No real fund publishes its alpha, and I won't pretend otherwise. But I'll tell you the direction to look: stop adding indicators and start subtracting them. Watch how price reacts at the same levels, at the same times, over and over. The patterns are there. Maybe one trader in a hundred ever bothers to work them out, because it's slower and harder than slapping an oscillator on a chart and calling it a system.

That's the whole edge behind Axiom FX, our gold-only MT5 Expert Advisor. No indicators. Just price, the numbers, and time, run with discipline I could never hold by hand at 2am.

The risk rules I never break

This is the part I'd tattoo on a beginner if I could.

Cap risk per trade. I never put more than a small, fixed slice of the account on one trade. The number matters less than the fact that it's fixed. When risk is fixed, no single loss can hurt you, and losses come in clusters whether you like it or not.

Set a drawdown ceiling. Drawdown is how far you've fallen from your peak. I have a hard line where I stop, full stop, no "just one more to win it back." Here's the math that should scare you straight, and it's exact:

To recover from a drawdown, you need: gain = DD / (1 − DD).

• Down 20%? You need +25% to get back to even.
• Down 50%? You need +100%. You have to double what's left.
• Down 90%? You need +900%. That account is basically gone.

Losses and gains are not symmetric. That asymmetry is why protecting capital beats chasing returns every time. A 50% loss doesn't need a 50% gain to fix. It needs a miracle.

Judge yourself in R, not win rate. R is your risk per trade, one unit. If you risk $100 and make $300, that's +3R. My old 95%-win-rate bot had a beautiful win rate and a catastrophic expectancy, because the rare loss was enormous. A system that wins 45% of the time but makes +2R on winners and loses 1R on losers will quietly get rich. Win rate is the number that flatters you. Expectancy in R is the number that pays you. You can see real R-based numbers, including the worst days, on the results page.

Demo first, micro lots second: how I'd start over from zero today

If I had to rebuild from nothing, here's the exact order.

Open a demo account and trade it for at least a month. Not a weekend. A month. Long enough to see gold do something you didn't expect, because it will. Treat the fake money like it's real or the demo teaches you nothing.

Then go to micro lots, not small lots. Micro. 0.01. Trade so small that a loss is almost boring. The goal at this stage is not profit, it's proving you can follow your own rules when actual money is on the line. Real money changes your psychology in ways demo never will, and you want that lesson to cost you ten dollars, not ten thousand.

Keep a log. Every trade: entry, stop, size, exit, and one honest sentence about why. Read it back monthly. The log is where you find out whether you have an edge or just a story you tell yourself.

Only after months of micro lots with a positive expectancy would I size up, and even then, slowly. There is no rush that's worth your account.

Honest expectations: what a realistic month of gold trading looks like

Let me kill the fantasy. A realistic good month is not "I doubled my account." A realistic good month is a handful of percent, with red days inside it that made you want to quit.

You'll have a week where nothing works and you do everything right and still lose. You'll have a day where one clean trade pays for the whole week. The equity curve is never a straight line going up, no matter what the screenshots on social media show you. Anyone promising smooth, steady gains is either lying or about to blow up, probably both.

The traders who last aren't the ones with the best entries. They're the ones who are still here after the bad months, because they never let a bad month become a bad year. That's it. That's the secret. Survival is the strategy.

When you're ready to let a disciplined system trade gold on your own account, with a hard stop on every trade and a 30-day profit-or-refund, that's exactly why I built Axiom FX. But whether you use it or trade by hand, please, lead with risk. Define your worst case before every click. The rest takes care of itself.

Risk note: trading gold carries real risk of loss. Past performance does not guarantee future results, and you should never trade money you can't afford to lose.

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.