Gold, quoted as XAU/USD, is the price of gold in US dollars - a commodity that behaves differently from currency pairs, with larger ranges, safe-haven flows, and its own pip mechanics. Those differences mean a strategy that works on EUR/USD cannot simply be copied onto gold without testing.
What drives gold
Gold is a safe-haven asset. It tends to rise when investors are fearful, when real interest rates fall, and when the US dollar weakens - and it can fall hard when risk appetite returns. Because it responds to macro fear as much as to scheduled data, it can trend powerfully for weeks and then whip violently around events like FOMC and CPI. It is more about sentiment than any one country's rate decision.
Different pip mechanics
Gold does not use the four-decimal pip of a currency pair. It is priced in dollars and cents per ounce, with a much larger contract value per point, so the same-sounding "move" is a completely different dollar amount. This trips up traders who carry their forex pip-value intuition onto gold. A simulator that handles gold's contract size for you removes the guesswork - but you must respect that a "100 point" gold move is not a 100-pip forex move.
Size for the range, not out of habit: gold's large daily range means stops must be wider, so position size must be smaller to keep risk constant. Traders who size gold like a currency pair routinely take far more risk than they intend. Let the stop distance set the size.
How to backtest gold
FxBacktest handles gold's contract size and pip value for you, so the dollar risk shown when you drag a stop is gold's real risk rather than a forex figure applied to the wrong instrument. That single detail is where most spreadsheet-based gold backtests quietly go wrong, and it compounds across every trade in the sample.
Load XAU/USD into a simulator that models its contract size and spread, and run your setup across a large sample that includes both trending and choppy periods. Gold's behaviour changes with the macro backdrop, so test across different conditions - a strategy that shines in a fear-driven uptrend may fail in a quiet range. Watch how its volatility interacts with your stops, exactly as you would on Cable. None of this needs paid software — see how to backtest gold for free, online for the full step-by-step loop.
Respect gold's differences
Gold offers some of the cleanest trends in any market, but only to traders who size for its range and understand its drivers. Backtest it as its own instrument - not as a currency pair with a different name - and you can add a genuinely different, uncorrelated market to your toolkit.
Measured: how gold's daily range changed over eleven years
Gold does not have "a" typical daily range — it has a range that has quadrupled inside a decade. We measured the average daily high-low on XAUUSD across 3,419 trading days:
| Year | Avg daily range | As % of price | EURUSD same year |
|---|---|---|---|
| 2015 | 137 pips ($13.70) | 1.18% | 107 pips |
| 2017 | 109 pips ($10.90) | 0.87% | 67 pips |
| 2018 | 105 pips ($10.50) | 0.83% | 73 pips |
| 2020 | 270 pips ($27.00) | 1.53% | 73 pips |
| 2022 | 223 pips ($22.30) | 1.24% | 88 pips |
| 2024 | 287 pips ($28.70) | 1.20% | 53 pips |
| 2025 | 535 pips ($53.50) | 1.52% | 71 pips |
| 2015-2025 | 216 pips ($21.60) | 1.17% | 71 pips |
Two findings matter for anyone backtesting gold:
- A stop distance from 2018 is not a stop distance for 2025. Gold's average day ran 105 pips in 2018 and 535 pips in 2025 — five times wider. A fixed 150-pip stop was generous in 2018 and is inside the noise today. This is the single biggest reason gold backtests fail to carry forward: the strategy was fitted to a volatility regime that ended.
- In percentage terms gold is steadier than it looks. The pip range exploded, but as a share of price it moved between 0.83% and 1.53% — because gold itself went from roughly $1,150 to well over $3,000. Gold is about 1.8x as volatile as EURUSD in percentage terms (1.17% vs 0.64%), not the 3x the raw pip counts suggest.
Practically: size gold positions off ATR or a percentage of price rather than a fixed pip stop, and when you choose a test window, include at least one quiet year (2017-2018) and one violent one (2020, 2025) so the result is not a single regime in disguise.
Computed 2026-08-27 from the FxBacktest tick-derived OHLC archive: 3,419 XAUUSD and 3,440 EURUSD daily bars, 2015-2025. "Pips" for gold means 0.1 of a dollar, so 535 pips = $53.50. Percentages are the daily high-low divided by that day’s close, averaged over the year.
Backtesting gold FAQ
What moves the price of gold (XAU/USD)?
Gold is a safe-haven asset. It tends to rise on fear, falling real interest rates, and a weaker US dollar, and fall when risk appetite returns. It responds to macro sentiment as much as to scheduled data.
How is backtesting gold different from a currency pair?
Gold has larger ranges, wider spreads, and different pip mechanics - it is priced in dollars per ounce with a bigger contract value per point, so a move worth the same number of points is a very different dollar amount.
Why do traders lose money sizing gold like forex?
Because gold's large range needs wider stops, and carrying a forex lot size onto gold takes far more risk than intended. Sizing must come from the stop distance and gold's contract value, not from forex habit.
Sources
- World Gold Council — Goldhub price data — the gold industry’s reference body for long-run price and demand statistics.
- LBMA — precious metal prices — the London Bullion Market Association, which administers the LBMA Gold Price benchmark.
- BIS Triennial Central Bank Survey — OTC FX turnover, April 2025 — the primary measurement of global FX market size: $9.6 trillion traded per day, up 28% from $7.5 trillion in 2022, with the US dollar on one side of 89.2% of all trades.