Position Sizing for Gold: The Math That Decides If You Survive

I've watched more accounts die from size than from being wrong. Let that sink in. The trader wasn't a bad reader of the market. He just bet too much on the trade that went against him, and one bad day ate six good weeks.

That's what this post is about. Position sizing for gold trading is the single most boring, least sexy, most account-saving skill there is. Nobody posts their lot-size formula on Instagram. They post the green P&L screenshot. But the formula is the only reason the screenshot exists. Get this wrong and it doesn't matter how good your read is — gold will find the one bad sequence and clean you out.

I learned this the expensive way. My first real account got wiped by a bot that won 95% of its trades. Ninety-five percent. Sounds incredible until you understand it had no fixed risk and no real stop — it just kept adding size to losers until the 5% showed up and took everything. Size with no math is a countdown timer. So let me show you the math I actually use, the same logic baked into how Axiom FX AI sizes every gold trade.

Gold Moves Different — Know Your Number First

Before any sizing math, you need one number burned into your head: on a standard lot of XAU/USD, gold moves roughly $100 for every $1 the price moves.

One standard lot is 100 ounces. Gold quoted in dollars per ounce. So if gold goes from 2400.00 to 2401.00 — a single dollar — and you're holding one standard lot, that's about $100 in your pocket or out of it. Move ten dollars against you on a full lot? That's a thousand dollars gone. On gold, ten dollars is nothing. It can happen in a slow afternoon.

Here's the scaling so it's clean:

• 1.00 lot (100 oz) ≈ $100 per $1 move ≈ $1 per cent
• 0.10 lot (10 oz) ≈ $10 per $1 move ≈ $0.10 per cent
• 0.01 lot (1 oz) ≈ $1 per $1 move ≈ $0.01 per cent

That 0.01 micro lot is the smallest most brokers let you trade, and people treat it like it's harmless. It isn't harmless if your stop is 30 dollars wide and you've stacked ten of them. Always do the multiplication. Gold doesn't care that the number felt small.

One thing that trips people up: brokers quote gold to two decimals (2400.00), and some platforms count a "pip" as a 10-cent move, others as 1 cent. Forget the word "pip" on gold. Think in dollars per ounce and dollars of account. That's the only language that won't lie to you.

Size From the Stop, Never From a Feeling

Here's the rule that took me years to fully respect: you do not pick your lot size. Your stop picks it for you.

Most people do it backwards. They decide "I'll trade one lot" and then place a stop wherever looks nice on the chart. That means their risk changes every single trade depending on how far the stop happens to be. Some trades they risk $200, some they risk $1,800, and they have no idea which is which until it's over. That's not trading. That's rolling dice with a different number of sides each time.

The right order is the reverse:

1. Decide your risk per trade as a fixed percent of the account. Mine lives around 1%. Never more than 2%. This is non-negotiable and it never moves based on how confident I feel.
2. Find where your stop goes — the price level where your reason for the trade is dead. Not a round number. The level where you were simply wrong.
3. Measure the distance from entry to stop, in dollars per ounce.
4. Solve for lot size so that distance equals your fixed dollar risk.

The formula is dead simple:

Lot size = (Account × Risk %) ÷ (Stop distance in $ × $100 per lot)

Let's run it with real numbers. Say you've got a $10,000 account and you risk 1%. That's $100 of risk on this trade — the most you're willing to lose, full stop.

Your setup needs a stop $5 away from entry (gold from 2400 to 2395). One full lot loses $100 per $1, so a $5 move on a full lot = $500. Way too much. So:

Lot = $100 ÷ ($5 × $100) = $100 ÷ $500 = 0.20 lots

You trade 0.20 lots. If gold hits your stop, you lose $100. Exactly $100. Not $97, not $340. The number you decided before you ever clicked.

Now change one thing. Same account, same 1%, but this setup needs a wider stop — $10 away. Watch what happens:

Lot = $100 ÷ ($10 × $100) = $100 ÷ $1,000 = 0.10 lots

Wider stop, smaller position. The risk in dollars stayed identical — $100 — but the lot size got cut in half because the stop got twice as wide. That's the whole point. The stop and the size move together so that your dollar risk never does. A wide stop isn't more dangerous than a tight one if you size for it. A wide stop on a fixed lot size is what kills you.

This is exactly how the engine behind Axiom FX handles it — risk is capped per trade as a percent, the hard stop sets the distance, and the lot size is solved from those two. There's no "this one feels strong, let's go bigger." The math doesn't have feelings, and that's a feature.

A quick risk note before we go further: every number here is about controlling loss, not predicting gains. Sizing keeps you in the game. It doesn't promise the game goes your way on any given trade.

Why One Trade Should Never Be Able to Hurt You

The reason I obsess over 1% isn't superstition. It's the recovery math, and the recovery math is brutal and exact.

When you lose, you don't need to make back what you lost — you need to make back more, because you're earning it on a smaller account. The formula is:

Gain needed to recover = Drawdown ÷ (1 − Drawdown)

• Lose 10% → you need +11.1% to get back to even
• Lose 25% → you need +33.3%
• Lose 50% → you need +100%. You have to double what's left just to break even.
• Lose 90% → you need +900%. You'd have to make ten times your remaining account.

That last one is why the 95%-win martingale bot was a death machine. The day it lost big, the hole was mathematically un-climbable. There's no read good enough to out-trade a 900% recovery requirement.

Now stack it against 1% sizing. Risk 1% per trade and even a genuinely awful run — five straight losers — costs you about 5%. Recovering 5% needs roughly +5.3%. Annoying. Survivable. You're still in your seat. That's the entire trade-off: small, fixed risk per position means no single trade and no short bad streak can put you somewhere the math won't let you return from.

This also reframes how you should judge a strategy. Forget win rate. A 40%-win system that makes 2.5R when it wins and loses 1R when it's wrong is a money machine. Win rate is the number that sells courses. Expectancy in R — what you make on average per trade, measured in units of your risk — is the number that fills accounts. And expectancy only means anything when your R is constant, which only happens when you size every trade off the stop. The whole structure depends on the sizing being disciplined first. You can read more about how we think about results and drawdown ceilings on the results page.

The Checklist I Run Before Every Gold Trade

Strip away everything else and this is the loop:

• Pick your %. 1% of the account. Write down the dollar figure. That's your max loss on this trade.
• Place the stop where you're wrong. Chart logic, not a comfort level.
• Measure entry-to-stop in dollars per ounce.
• Solve for lots: risk dollars ÷ (stop distance × $100).
• Round down, never up. If the math says 0.27, trade 0.25 or 0.20. Smaller never hurt anyone.
• Check the worst case out loud. "If this stops, I lose $___." If that number makes your stomach drop, you're too big. Cut it.

Do that on every trade and you've removed the one thing that actually kills accounts. You'll still have losers — everyone does. But your losers will be a paper cut instead of a chest wound, and you'll be there next week to take the trade that pays.

That's the unglamorous truth nobody screenshots. Survival isn't a vibe. It's risk ÷ (stop × $100), run every single time, no exceptions. If you want the sizing done for you with a hard stop and capped risk on every gold trade, that's the whole job of Axiom FX AI — and it's backed by 30-day profit-or-refund, $999 back in USDT if it doesn't deliver.

Risk note: trading gold carries real risk of loss. Position sizing limits how much a single trade can cost you; it does not guarantee profits.

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.