Why a stop does not protect you across a gap
A stop order is a trigger level; on reaching it, it becomes a market order filled at the best available price. Across a gap there is no available price at your level, because the market never traded there.
So it fills at the first real price beyond the gap, which can be far away. This is why "risking 1%" is an estimate rather than a guarantee, and the same reason a stop-out cannot promise your balance will not be exceeded.
Cost measured against risk
Execution quality is not a technical detail but a cost line measured in dollars. The same reference account the cost family uses: 0.10 lots of gold at 2,400, where $1 of price is $10 of money, and 1% of a $1,000 account is $10.
On that scale every execution cost is directly comparable with the spread, slippage and swap already published — and the only question that matters is how much of the risk budget it consumes before price moves at all.
A worked example
Buy gold at 2,400 with a stop at 2,380 — a planned risk of $20 per ounce, or $200 on 0.10 lots.
News breaks over the weekend and the market opens at 2,340. Your stop triggers and fills at 2,340 rather than 2,380, making the loss $60 per ounce — $600, three times the plan. The stop did not fail and nobody interfered; there was simply no buyer between 2,380 and 2,340.
Common mistakes with this term
- Computing the worst case on the assumption that a stop fills exactly at its number.
- Carrying a large size across a weekend or a major event without reducing it.