Is gold a good market for automated trading?

The case in favour is range. We measured a median daily range of $76 on XAUUSD over a year, so a strategy targeting a few dollars has room to work — on a major FX pair, the equivalent target is a much larger share of the day's movement. Gold also respects structure during London and New York hours in a way that gives level-based strategies something to key on.

The case against starts with cost. The XAUUSD spread is wide relative to FX majors, it widens sharply at rollover and around news, and on short targets it consumes a meaningful share of the gross move before slippage. Any gold strategy has to clear a higher cost bar than the same idea on EURUSD.

Volatility is the other side of the same coin. The same movement that creates opportunity destroys accounts sized as if gold were a currency pair — a $2 stop on gold is not tight, it is a coin flip against normal noise. Sizing to ATR rather than to a fixed dollar amount is close to mandatory.

Why we trade only gold

Our engine trades XAUUSD and nothing else, which is a deliberate narrowing rather than a limitation we have not got around to. We tested Bitcoin alongside it and disabled it — the reasoning is written up in why we disabled Bitcoin.

The argument for a single instrument is that every strategy parameter, every filter and every risk setting can be tuned to one market's behaviour rather than compromised across several. The argument against is concentration, and it is a real one: a system trading only gold has no diversification whatsoever.