The ounce is not the ounce you think
Gold is priced in troy ounces. A troy ounce is 31.1035 grams. A kitchen (avoirdupois) ounce is 28.35 grams. That is a 9.7% difference — enough to wreck any calculation where you mixed them up.
Other units you will meet: a kilogram is 32.151 troy ounces (the standard London bar is about 400 oz, the "good delivery" bar); a tola is 11.66 grams and still used across South Asia; a tael is 37.5 grams in Hong Kong and 50 grams in mainland China.
What a lot actually is
On most CFD platforms, one standard lot of XAU/USD is 100 troy ounces. So a one-dollar move in the gold price is a hundred-dollar move on one lot. Mini and micro lots (10 oz, 1 oz) exist on many platforms — check yours before you size anything.
This is why gold feels expensive. At $4,300 an ounce, one lot is $430,000 of notional. With 20:1 leverage that is $21,500 of margin. A $10 adverse move — 0.23% — is $1,000 on a single lot.
Point versus pip: the ten-times trap
Here is the single most expensive mistake in gold CFDs. Brokers quote XAU/USD to two decimals ($4,300.12), but they disagree about what one "unit" of movement means:
- Convention A: one point = $0.01. So one point on one lot = $1.00.
- Convention B: one pip = $0.10. So one pip on one lot = $10.00.
If you set a "500 point stop" believing convention A and your platform uses convention B, you have just placed a stop ten times further away than you intended — and sized a position ten times too large for your risk. This is not a theoretical risk. It is one of the most common ways retail gold accounts blow up.
The fix is boring and essential: always compute your stop in dollars, then convert. Decide the price level where you are wrong, subtract it from entry, and multiply by ounces. Never size from a pip count you have not verified.
How to check this yourself
Before your next gold trade, open your broker's contract specification for XAUUSD and write down three numbers: the lot size in ounces, the point value, and whether the platform quotes stops in points or pips. Then price the same 500-unit stop under both conventions. If the two answers differ by a factor of ten — and they often do — you have just found the most expensive assumption in your trading.