Silver stops are the hardest part of trading XAGUSD for me, and the honest answer is that I size down rather than tighten up.
XAGUSD moves in jumps. A stop that would be sensible on EURUSD in terms of structure gets taken out by normal noise on silver. When I first traded it, I used the same R per trade and the same stop logic as gold, and got stopped out constantly by moves that reversed within a minute.
What I do now: stop placement is beyond the structure plus a buffer of about a third of the average one-minute range. That usually makes the stop wider in price than I would like. To keep the risk the same in money, I cut position size. If the required size gets too small to make the trade worth the commission, silver is off the roster for that session.
I also avoid stops at obvious round numbers like 30.00, because they get swept regularly.
Handling stops means sizing to the instrument, not wishing it behaved like EURUSD. If you cannot afford silver's normal noise at your risk per trade, silver is simply not your instrument right now, and that is fine.
XAGUSD stop placement: how I handle the width
-
LondonScalper
- Posts: 965
- Joined: Sat Sep 05, 2026 7:54 am
-
Shadow Trader
- Posts: 5594
- Joined: Wed Sep 09, 2026 5:19 pm
Re: XAGUSD stop placement: how I handle the width
LondonScalper, sizing down instead of tightening is the only approach that kept me in silver longer than a week. When I mirrored my gold stop distance on XAGUSD I was stopped on noise roughly twice as often, and half of those reverses came back inside a minute.
My buffer sits past the structure by roughly one third of the typical M1 range. On a typical London morning that often means 12 to 18 cents of stop width, so position size has to shrink if risk in money stays fixed.
Round numbers are a separate landmine. Stops at 30.00 and 32.00 got swept on four of the five times I tried them last quarter, usually by a quick spike that faded.
If the computed size falls below a quarter of my normal FX lot, I simply skip silver that session. Commission then eats too much of any realistic target.
Do you ever widen further into thin Asia hours? Mine is simply off the roster when the one minute range is already large before London.
My buffer sits past the structure by roughly one third of the typical M1 range. On a typical London morning that often means 12 to 18 cents of stop width, so position size has to shrink if risk in money stays fixed.
Round numbers are a separate landmine. Stops at 30.00 and 32.00 got swept on four of the five times I tried them last quarter, usually by a quick spike that faded.
If the computed size falls below a quarter of my normal FX lot, I simply skip silver that session. Commission then eats too much of any realistic target.
Do you ever widen further into thin Asia hours? Mine is simply off the roster when the one minute range is already large before London.
-
Shadow Trader
- Posts: 5594
- Joined: Wed Sep 09, 2026 5:19 pm
Re: XAGUSD stop placement: how I handle the width
One thing I would add for anyone coming from FX is the contract math, because silver catches people there before the stop logic even matters. On most CFD feeds one lot is 5,000 ounces, so each cent of movement is 50 dollars per lot.
That turns a 15 cent stop into 750 dollars on a full lot. If my risk is 300, the size is 0.4 lots, and I have seen people size silver like a forex lot and lose a week of risk on one noisy candle.
I also stopped using a single buffer for the whole day. The one third of M1 range rule works, but I compute the range from the same hour on the last ten sessions instead of a daily average, because the hour around the COMEX open runs far wider than mid London.
On my notes that hour often needs a buffer close to double the morning one. Using the daily average there put my stops inside the normal noise of that window, which is the exact mistake you describe.
Do you treat the US metals open as its own session for silver, or keep one rule set from London through New York? I lean toward a separate size for that hour.
That turns a 15 cent stop into 750 dollars on a full lot. If my risk is 300, the size is 0.4 lots, and I have seen people size silver like a forex lot and lose a week of risk on one noisy candle.
I also stopped using a single buffer for the whole day. The one third of M1 range rule works, but I compute the range from the same hour on the last ten sessions instead of a daily average, because the hour around the COMEX open runs far wider than mid London.
On my notes that hour often needs a buffer close to double the morning one. Using the daily average there put my stops inside the normal noise of that window, which is the exact mistake you describe.
Do you treat the US metals open as its own session for silver, or keep one rule set from London through New York? I lean toward a separate size for that hour.
-
Shadow Trader
- Posts: 5594
- Joined: Wed Sep 09, 2026 5:19 pm
Re: XAGUSD stop placement: how I handle the width
MAE is maximum adverse excursion: how far a trade went against the position before closing. MFE is maximum favourable excursion: how far it went in the position's favour. Most platforms don't show them directly, but a tick or one-second dataset lets them be computed.
Why it matters. Plot MAE for winning trades only. If 90% of winners never went more than 3 pips against the entry, a 6 pip stop is wasting risk. Tightening it to 4 pips cuts losses while barely touching the winners.
Now plot MFE for losing trades. If many losers first went 4 pips in profit before reversing, a take-profit or a trailing rule might have salvaged them.
The whole exercise turns stop and target placement from guesswork into measurement. It also reveals whether the strategy's entries are good (winners show small MAE) or merely lucky.
Nothing fancy is needed beyond a spreadsheet and the trade timestamps merged with tick data.
A beginner-friendly start: pick the last 100 trades, pull one-second price data for each, and compute the lowest and highest price between entry and exit. Two columns, two charts. Even this crude version often changes the next week's stop placement.
Has MAE or MFE analysis ever led to a change in stops or targets, and by how much did results improve?
Why it matters. Plot MAE for winning trades only. If 90% of winners never went more than 3 pips against the entry, a 6 pip stop is wasting risk. Tightening it to 4 pips cuts losses while barely touching the winners.
Now plot MFE for losing trades. If many losers first went 4 pips in profit before reversing, a take-profit or a trailing rule might have salvaged them.
The whole exercise turns stop and target placement from guesswork into measurement. It also reveals whether the strategy's entries are good (winners show small MAE) or merely lucky.
Nothing fancy is needed beyond a spreadsheet and the trade timestamps merged with tick data.
A beginner-friendly start: pick the last 100 trades, pull one-second price data for each, and compute the lowest and highest price between entry and exit. Two columns, two charts. Even this crude version often changes the next week's stop placement.
Has MAE or MFE analysis ever led to a change in stops or targets, and by how much did results improve?