Most traders do not fail because their view of gold was wrong. They fail because the position built on that view was too large to survive being wrong for a week. Risk management is the part of the process that decides how much a mistake is allowed to cost — and it is settled before the trade, not during it.
Every risk decision reduces to a single figure: the amount of money you are willing to lose on one idea. Not the amount you hope to make, not the margin the platform requires — the amount that leaves the account if the trade goes completely against you.
For a discretionary account, one percent of equity per trade is the standard starting point, and two percent is a sensible ceiling. The reason is arithmetic rather than temperament. A trader risking one percent can absorb ten consecutive losses and still hold ninety percent of the account. A trader risking ten percent is down to thirty-five percent after the same streak and needs a gain of roughly 187 percent just to return to break-even. Losing streaks are a normal feature of any strategy with a real edge; position size determines whether the streak is an inconvenience or the end of the account.
Fix the number in money, not in percent, at the start of each session. Percentages drift as the balance moves, and a trader who recalculates mid-trade is really just negotiating with themselves.
The most common sizing error is choosing a lot size first and then looking for a stop that fits the margin. That inverts the logic. The stop belongs where the trade idea is proven wrong; the position size is whatever makes that stop cost the risk budget.
Units = risk budget ÷ stop distance per unit
Account 5,000, risk 1 percent = 50. Stop placed 8.00 below entry on XAU/USD, so the stop distance is 8.00 per ounce.
50 ÷ 8.00 = 6.25 ounces, which is 0.0625 of a standard 100-ounce lot. At 2,400 an ounce the notional is 15,000, and at 1:100 leverage the margin held is about 150.
Two things follow from that example. First, the margin requirement is not a measure of risk: 150 of margin was committed to carry 50 of planned risk, and the gap is exactly where over-leveraged accounts get hurt. Second, a wider stop does not mean a riskier trade — it means a smaller position for the same risk. Traders who keep the lot size fixed while widening the stop are quietly increasing their risk with every adjustment.
A position size calculator removes the arithmetic, but it cannot choose the stop for you.
A stop is a statement about the market, not about your comfort. Placing it at the exact dollar amount you are willing to lose guarantees it sits inside normal noise, where it will be collected by routine fluctuation rather than by a genuine change in the idea.
Once the stop is set, the only legitimate reasons to move it are a change in the market structure that justified the trade, or moving it to reduce risk on a position already in profit. Widening a stop to avoid being closed out converts a defined loss into an undefined one, which is the single most expensive habit in leveraged trading.
On a product where the stake is the maximum loss, there is no margin call to manage — but the control problem does not disappear, it moves. The risks become stake inflation, simultaneous exposure, and the urge to recover a loss with a larger bet.
Committing the whole account to one outcome is not a display of conviction, it is a decision to make the next hundred trades irrelevant. Even a genuinely good idea with a 70 percent win rate loses three times in ten, and an all-in position cannot survive the first loss, let alone a cluster of them.
The same logic applies across positions. Gold is one market with one dominant driver at a time. A long XAU/USD position, a long silver position and a short dollar position are frequently the same trade wearing three tickets; when the dollar rallies, all three lose together. Exposure should be measured in aggregate, not per ticket, and the aggregate should still respect the one percent budget.
Cash is also a position. Sitting out a session with no clear setup costs nothing and preserves the ability to act when one appears. Traders who feel obliged to hold something at all times are usually paying the spread for the privilege.
Individual stops protect against a bad trade. A loss limit protects against a bad day, which is the more common failure mode: three losses in a row, followed by a larger position to make it back, followed by a fourth loss that erases the month.
A workable rule is a daily loss limit of three to five percent and a weekly limit of roughly double that. When the limit is hit, the platform closes and so does the laptop, regardless of how obvious the next setup looks. The purpose is not to avoid the loss already taken; it is to prevent a normal drawdown from being converted into a decision made under pressure.
Take the same approach to winning streaks. A trader up sharply in a day tends to enlarge positions, and the enlarged position usually arrives just as the market returns to normal. Size should be a function of the plan, not of the morning's result.
Good risk management is undramatic. Losses are small and similar in size. No single day decides the month. The equity curve has drawdowns but no cliffs, and the trader can state the maximum they can lose this week without checking the platform. If any of those statements is false, the problem is size — and size is the one variable that is entirely under your control.
Two related primers: what the bid-ask spread really costs, and how to convert ounces into lots and money.
One percent of account equity per trade is the common professional baseline, and two percent is the practical ceiling for discretionary trading. The number matters less than applying it consistently, because consistency is what keeps a normal losing streak from becoming an unrecoverable one. On a 5,000 dollar account, one percent is 50 dollars of risk, and that figure should be set before the entry is chosen.
The stake itself is the stop, since a losing contract cannot cost more than the amount committed. What is still missing is a limit on repetition. The equivalent controls are a fixed stake size of one to two percent of the balance, a cap on the number of contracts open at once, a daily loss limit, and a rule against increasing the stake after a loss to win it back.
Because drawdown compounds against you. Ten consecutive losses at one percent leave about ninety percent of the account and need roughly eleven percent to recover. Ten consecutive losses at ten percent leave about thirty-five percent and need about one hundred and eighty-seven percent to get back to even. The losing streak is not unusual; the size is what turns it into a terminal event.
Trading on the spot gold price carries risk of loss. This page is educational and is not investment advice. Past performance does not indicate future results.