Forex Robot Risk Management: Position Sizing, Stops, and Drawdown

Position sizing is the single most important risk decision, and it is yours to make. The standard for sustainable automated trading is to risk 0.5-2% of account equity per trade, meaning the distance from entry to stop-loss, multiplied by position size, costs no more than that fraction if the trade loses. Conservative operators run 0.5-1%; few sustainable systems run above 2%.

The reason is survival mathematics. At 2% risk per trade, a string of ten consecutive losses costs roughly 18-20% of the account — painful but recoverable. At 5% per trade, the same losing streak costs about 40% and may force the EA into oversized lots to recover, accelerating the damage. Aggressive sizing does not make a system more profitable; it makes the inevitable losing streak unrecoverable. Most EAs let you set risk per trade or lot size directly — treat that input as the most consequential setting you control.

Two stops protect an automated account. The per-trade stop-loss bounds the loss on any single position; a robot that trades 'without a stop-loss' (relying on averaging to recover) carries unbounded risk and should be treated with extreme caution regardless of its track record.

The second, often missing, control is an account-level daily-loss limit — a rule that halts the EA after the account loses a defined amount in a day (for example 5%) rather than letting it keep trading into a bad session. This bounds tail risk at the calendar boundary: the worst a single day can do is the limit you set. Some EAs include a daily-loss cut-out; if yours does not, a separate risk-manager utility or a manual rule serves the same purpose. The discipline to honour the limit — to let the EA stop and review rather than overriding it to 'win it back' — is what converts a temporary drawdown into a recoverable one instead of a permanent loss.

Drawdown — the peak-to-trough decline in equity — is the metric that describes the worst experience a robot will put you through, and bounding it is central to risk management. Every profitable strategy has drawdowns; a system with a 1.5 profit factor will inevitably hit 20-30% drawdown stretches. The goal is not to avoid drawdown but to know its likely size in advance and fund and size so you can survive it psychologically and financially.

Set a maximum-drawdown threshold before you go live — the equity decline from peak at which you will stop the EA and review (commonly 20-30%, lower for conservative profiles). Pre-committing to this number removes the in-the-moment emotional decision. Critically, judge a robot's expected drawdown from its live track record across a volatile period, not its calm-period backtest, because the drawdown that matters is the one the strategy has not shown you yet.

Underfunding is a risk failure disguised as a budgeting choice. Every strategy has a capital floor below which its position sizing cannot respect the 0.5-2% rule without the broker's minimum lot size forcing oversized risk. Run a system below its floor and you are effectively trading at a much higher risk percentage than intended, which raises the risk of ruin — the probability of losing enough capital that recovery is impractical.

Most strategies need a realistic floor of $1,000-$2,000, and higher-volatility instruments like gold or indices need more. The relationship between risk per trade, win rate, and account size determines risk of ruin: small accounts at high per-trade risk have a meaningful probability of ruin even with a positive expectancy, simply through an unlucky early sequence. Fund the account to the strategy's floor or above, and keep per-trade risk modest, and risk of ruin falls toward negligible.

It is worth being explicit about the division of labour. The EA controls trade-level mechanics — entries, its own stop and take-profit, its trade timing. You control the account-level risk envelope that the EA operates inside: how much capital funds it, the risk-per-trade setting, the daily-loss and maximum-drawdown limits, and the decision to pause the system when those limits are hit or when the market regime turns against the strategy.

This is why 'set and forget' is a risk myth. Automation removes the trade-by-trade decisions, not the risk-management ones. The operators who survive treat the EA as one component inside their own risk framework — they size conservatively, pre-define their stop-the-EA levels, fund to the floor, and review on a schedule. The strategy gets the headlines; the risk controls keep the account alive long enough for the strategy to matter.