The one calculation that sets the position
Take the cash you accept losing, divide it by the stop distance in dollars, then divide by one hundred. If you accept losing fifty dollars and the structure puts your stop five dollars away, fifty divided by five is ten ounces, which is 0.10 lots. Widen the stop to ten dollars and the same fifty dollars of risk now buys 0.05 lots. The balance did not change. The chart did.
Most traders run this backwards because the order ticket asks for lot size first and leaves the loss as a surprise. Treat the loss as the input and the lot as the output and the decision stops being emotional. The formula also exposes setups you cannot afford, which is useful information rather than a problem to be solved with leverage.
Why the gold lot behaves differently from a currency lot
On gold the contract is one hundred ounces, so every one dollar move in the price is one hundred dollars on a single lot and one dollar on 0.01 lots. Gold also travels in dollars that feel small and are not. A quiet day can cover ten to twenty dollars of range and a data day can cover far more, so a one lot position is exposed to a swing of a few thousand dollars without anything unusual happening.
Traders coming from major currency pairs often carry over a lot size that looked tame there and discover that the same number on gold produces four or five times the swing. The fix is not a rule of thumb. It is the same division as above, run fresh on each trade, with the stop measured on the chart rather than guessed.
Three capital tiers and what they can carry
The percentage you risk should be roughly constant while the cash it represents scales with capital. One percent is a common and defensible figure. On a five hundred dollar account that is five dollars of risk, on a five thousand dollar account fifty dollars, on a fifty thousand dollar account five hundred dollars.
- Small capital, a five dollar risk budget and a five dollar stop gives 0.01 lots, which is the floor on most accounts.
- Mid capital, a fifty dollar budget with the same stop gives 0.10 lots and room to split the entry.
- Larger capital, a five hundred dollar budget gives one lot, at which point fill quality starts to matter more than sizing.
Measure the stop before you think about size
The stop belongs behind the thing that would prove the idea wrong: the swing that must hold, the far side of the zone, the wick that swept liquidity before the reclaim. That distance is handed to you by the chart, and it changes with volatility. The same zone can need a three dollar stop in a quiet Asia range and an eight dollar stop an hour after a data release.
Measure it on the live chart, in dollars, before the order ticket is open. Our Today Levels zones give a concrete edge to measure from, since each zone has a width and a side that invalidates it, and a stop placed just beyond that side is a number you can defend rather than a round figure you liked.
If the required stop makes the lot smaller than your platform allows, the trade is not available to you at this capital. That is the honest answer, and taking it anyway at a size the account cannot absorb is the single most common way small accounts end.
Where small accounts actually break
The minimum size on most retail accounts is 0.01 lots, which is one dollar of loss per dollar of adverse move. On a two hundred dollar account a ten dollar stop at that minimum is a five percent loss, well past any sane limit. The arithmetic says trade smaller, the platform says you cannot, and the gap is where accounts get destroyed.
Raising the risk percentage is not on that list. A string of four or five losses is ordinary, and at five percent each it removes a quarter of the account before the method has had a chance to show anything. More detail on the mechanics of pip value and sizing sits in our guide on pip value and lot size on gold.
What changes when the position gets large
Above roughly one lot the limiting factor stops being the formula and becomes execution. During the New York data window the book thins, the spread widens for a few seconds and a market order can fill several dollars away from the last print. A stop order in that window can do the same, which means the loss you calculated is a best case rather than a fixed cost.
The practical adjustments are dull and they work. Split the entry rather than sending it in one clip, avoid adding size in the first minutes after a scheduled release, and size off the assumption that your stop will fill worse than it reads on the screen.
What the formula cannot protect you from
The calculation controls one loss on one trade under normal conditions. It says nothing about the weekend, when gold can open away from Friday close and straight through a stop. It says nothing about a spike on a headline, where the price trades through your level and the fill lands past it. In both cases the realised loss can be larger than the number you accepted.
It also says nothing about correlation. Two gold positions, or a gold position plus a position in a dollar pair, are usually one bet expressed twice, and the combined exposure is what matters. Sizing each independently at one percent can leave you carrying three percent on a single theme.
And it cannot stop the behavioural failure it was built to prevent. A trader who moves the stop or adds to a loser has left the arithmetic behind. Our wider notes on process live in the gold risk management guide.
A sizing routine that survives a bad week
Fix the risk percentage once, write it down, and leave it alone for a full month of trading regardless of how the last few trades went. Doubling after a loss and halving after a win turns a flat method into a losing one.
Before each entry, do four things in order: mark the invalidation level, measure the stop distance in dollars, convert the cash budget into a lot with the division above, then round down. The order matters because any other sequence lets the lot you wanted dictate where the stop goes.
FAQ
What lot size should a beginner use on gold?
Start at the platform minimum of 0.01 lots and stay there until the record shows a method that holds up across at least thirty trades. At that size a ten dollar adverse move costs ten dollars, which is survivable while you learn. Increase only by recalculating the cash budget from a fixed percentage, never by rounding up because a setup looks convincing.
How many lots can a small account trade on gold?
Divide the cash you accept losing by the stop distance in dollars, then by one hundred. On a small balance that calculation often lands below the platform minimum of 0.01 lots, which means the trade is genuinely unavailable rather than merely inconvenient. Waiting for setups with a close invalidation level is the only honest way around that limit.
Why does the same lot size feel bigger on gold than on a currency pair?
Because a gold lot is one hundred ounces, so a one dollar price move is one hundred dollars per lot, and gold routinely covers ten to twenty dollars in a session. A position size that produced modest swings on a major pair can produce several times that on gold without the market doing anything unusual.
Should lot size change with volatility?
Indirectly, yes. Keep the cash risk constant and let the stop distance do the work. When expected range widens, the stop that respects structure gets wider, the division returns a smaller lot, and the position shrinks automatically. That is the correct response. Holding the lot fixed while the stop widens quietly multiplies the loss you agreed to take.
ⓘ See these ideas on real price: open the free XAUUSD live chart.