Most automated accounts are not undone by a bad strategy. They are undone by good strategies run with bad risk management — position sizes too large, no cap on total exposure, and no plan for the losing streak that statistics guarantee will eventually arrive. Before a single automated trade fires, the risk framework should already be in place.
Size by risk, not by lots
Think in terms of how much of the account a trade can lose, not how many lots it uses. A common discipline is to risk a small fixed percentage — often well under two percent — of equity per position, with the lot size derived from the stop distance. That way a wider stop automatically means a smaller position, and no single trade can meaningfully dent the account.
Cap total exposure
Per-trade risk is not enough if the system can open twenty correlated positions at once. Set a ceiling on aggregate risk and on how many positions can run simultaneously, especially across instruments that tend to move together. Correlated trades are, in effect, one big trade wearing several costumes.
Plan for drawdown before it happens
Every strategy has losing streaks; the question is whether you have decided in advance how you will respond. Define a maximum drawdown at which the system pauses for review, and understand that recovering from deep drawdowns is mathematically brutal — a fifty percent loss needs a hundred percent gain to get back to even. Protecting capital is the whole game.
Leverage cuts both ways
Leverage is what makes small accounts interesting and what makes them disappear. It amplifies gains and losses in equal measure, and it turns an ordinary adverse move into a margin call. Use less than the platform allows, and size as if the leverage were lower than it is.
These are general educational principles, not financial advice. Trading — automated or manual, leveraged or not — carries real risk of loss; set your own limits deliberately and never risk money you cannot afford to lose.