1. Those Stop Losses That Got Swept Over the Years—What Was Really Wrong
I've traded XAUUSD for over ten years, watching the Asian, London, and US sessions. Early on, the biggest headache was getting my stop loss swept. The most memorable time was in August 2020, when gold was around 2075. I placed a short order with a stop loss at 2068. A single 15-minute candle shot straight to 2065, swept my 0.5-lot stop, and then price fell back to 2060. I stared at the screen with one thought: why does it always precisely hit my stop and then go in my direction?
After many reviews, I realized it wasn't the market hunting my stop—it was that I placed my stops too casually. Back then, I used to put stops a few points below round numbers like 2068, 1950, 1800, thinking round numbers had support and resistance. But gold's liquidity is often thinnest around round numbers, making it cheapest for institutions to sweep. I analyzed three months of my trading records: over 60% of the time, after my stop was swept, price reversed back in my direction. It's not luck—it's choosing the wrong stop location.
Another pitfall was the mismatch between stop distance and position size. Early on, I traded 0.5 lots with only a $3 stop, thinking I'd lose less. But gold's normal volatility in the first hour after London open can be $8-10, so a $3 stop can't withstand the noise. During that period, I often got swept two or three times a day, losing about $150 each time, and my account shrank 10% in a week. Later, I widened my stop to $6-8 and reduced my position to 0.2 lots. With the same risk amount, the frequency of sweeps dropped noticeably.
2. What I Changed: From "Setting a Stop" to "Setting the Right Stop"
The first change was to set stops based on price action, not round numbers. Now when I trade XAUUSD, I always place my stop beyond the most recent obvious swing low or high, with a $1-2 buffer. For example, in March 2023, gold was consolidating around 1810. I waited for a false breakout below 1805 and a recovery, then went long at 1812 with a stop at 1803—a $9 distance—and a 0.3-lot position. That trade eventually reached 1830, but there were two pullbacks to 1806 that didn't hit my stop. If I had followed my old habit of placing the stop just below the round number 1805, it likely would have been swept again.
The second change was to avoid the worst liquidity periods. In the Asian early session (6-9 AM Beijing time), gold spreads are wide and moves are random. I used to get swept most often during that time. Now I only look for opportunities after London open (after 15:00 Beijing time) and in the first two hours of the US session (20:00-22:00). Price action is more continuous in these sessions, with fewer false breakouts. Not absolute, but my stop-sweep frequency has halved.
The third change was to tie the stop loss to the entry logic. Previously, I entered because I "felt it would go up," and placed the stop randomly. Now I first find key levels, wait for a clear reversal signal (like a pin bar or engulfing pattern) at that level, then enter with the stop below the low of the signal candle. If the signal isn't clear, I don't trade. That way, if my stop is swept, at least it means my judgment was wrong, not that I was taken out by noise.
3. What I Do After Getting Swept: Record and Trade-offs
Now when my stop is swept, I don't immediately reverse or add to my position. I first take a screenshot, note the session, level, stop distance, and position size, then wait at least half an hour before looking again. Most of the time, after the sweep, price continues in my original direction for a while, but I don't chase. Because getting swept means my entry timing or stop placement was off, and chasing would only repeat the mistake.
I review my stop-loss records once a week. If the same instrument and session get swept more than three times in a row, I pause trading that session and just observe. Last year, during a period, I got swept four times in a row in the first hour after the US open, each time with my stop within $3 below a key level. Later, I widened my stop to $5-6 and halved my position, and the problem was solved.
In the end, getting your stop swept isn't the market targeting you—it's a mismatch between your stop location, position size, and entry logic. Gold is volatile with lots of noise, so stops need enough room and positions should be scaled down accordingly. Now when I trade XAUUSD, I keep single-trade risk within 1% of my account, adjust stop distance based on volatility (usually $6-10), and use 0.2-0.3 lots. The number of sweeps is lower, but not zero. That's trading: accept that stop losses are part of the cost, and the key is not to repeat the same mistake too many times.