A stop is not "tight" or "wide" in the abstract — it is either inside the market's normal noise or outside it. Enter your distance and see what share of real holding windows would have taken it out.
Loading…
Read this before trusting the number: the curve is built from entries taken at random moments, not from your setup. It answers "how volatile is this instrument over this holding time", which is the floor your stop has to clear. A good entry does better than random; a bad one does worse.
You type in a stop distance. It tells you what share of real historical holding windows would have travelled at least that far against you.
That is it. One number, measured from 559,812 sampled four-hour windows across five instruments, with the whole curve drawn underneath so you can see how the answer changes as you widen or tighten the stop.
Almost every new trader is told "keep your stop tight, cut losses fast". Taken literally, that advice destroys accounts — not through big losses, but through a long string of small ones, each one a position that was right about direction and wrong about distance.
Here is the mechanism. Gold's typical four-hour window sees price travel about 45 pips against a random entry before it goes anywhere. If you set a 20-pip stop on gold and hold for four hours, you are not managing risk — you are buying a lottery ticket that pays out only when the market happens to move your way immediately, with no wobble at all.
The table shows how quickly the odds change with distance. Every figure is the measured share of four-hour windows that reached that far against a long position:
| Instrument | Stop hit 50% of the time | Hit 25% | Hit 10% | Hit 5% |
|---|---|---|---|---|
| XAUUSD | 45 pips | 98.7 pips | 201.8 pips | 308.3 pips |
| GBPUSD | 13 pips | 24.9 pips | 42 pips | 55.9 pips |
| EURUSD | 9.8 pips | 19.2 pips | 32.4 pips | 43.8 pips |
| US30 | 62.6 pts | 134.5 pts | 242.7 pts | 333 pts |
| NAS100 | 38.1 pts | 83.5 pts | 155.9 pts | 215.4 pts |
Read the gold row across. Going from a stop that gets hit half the time to one that gets hit only one time in twenty means widening from 45 to 308.3 pips — about 6.9 times further. There is no free version of this. A tighter stop is genuinely cheaper when it is wrong and genuinely more likely to be wrong. The tool exists so you can choose that trade-off with a number in front of you instead of a slogan.
The technical name for what is measured here is maximum adverse excursion, usually shortened to MAE. In plain words: from the moment you enter, what was the worst point you sat through before the clock ran out.
Three deliberate choices are worth flagging, because they are the difference between a real measurement and a marketing number:
This is not your strategy's stop-out rate, and the tool will never claim to be. If your entries have genuine edge — you enter after a rejection, at a level, in the right direction — your real hit rate should come in below the curve. If it comes in above, your entry timing is worse than random, and that is far more valuable to learn than any stop-placement rule.
Sara reads that professionals risk 1% per trade. Her account is $2,000, so she is willing to lose $20 per trade. She wants to trade gold at 0.10 lots, where each pip is worth $1. That gives her a 20-pip stop.
She enters the numbers here: gold, long, four-hour hold, 20 pips. The tool reports that roughly 74% of real four-hour windows travelled at least 20 pips against a random entry. She will be stopped out most of the time regardless of whether her read was right.
She has three honest options, and the tool makes all three visible. She can cut the position size to 0.03 lots and take a 98.7-pip stop for the same $20 risk. She can shorten the holding time to one hour, where the adverse moves are much smaller. Or she can trade a smaller instrument — the same $20 buys a far more comfortable stop on EURUSD than on gold.
What she cannot do is keep the 20-pip stop, keep the four-hour hold, keep gold, and expect it to work. That was never a risk management decision — it was arithmetic she had not done yet.
This curve is the baseline every strategy has to beat. The only way to find out whether yours does is to replay the same instrument bar by bar, take your actual setups, and record where the stop sat when each one resolved. That is exactly what the free backtester is for.
Backtest your stop placement free →Stop distance, position size and holding time are one single decision wearing three hats. Change any one and the other two must move:
| If you… | Then your stop… | And your position size… |
|---|---|---|
| Hold longer | must be wider — more time means more room to wander | must come down to keep the same money at risk |
| Trade a more volatile instrument | must be wider in that instrument's own pips | must come down, often sharply |
| Trade a quiet session | can be tighter | can be larger for the same risk |
| Want a tighter stop | gets hit more often — the curve is not negotiable | can be larger, but you will be wrong more often |
The quiet hours in the last row are the same ones mapped by the Session & Volatility Clock. The two tools are two views of one underlying fact: volatility is not constant, and everything you decide about risk depends on it.
How many pips should my stop loss be on gold?
There is no single right answer, but there is a measurable range. On XAUUSD over a four-hour hold, a 45-pip stop was reached in about half of all random-entry windows, 98.7 pips in a quarter of them, and 201.8 pips in one in ten. Pick the hit rate you can live with, read the distance off the curve, then size the position so that distance costs you an acceptable amount of money.
What is maximum adverse excursion?
It is the worst point a trade travelled through before it finished — how far into loss you sat, regardless of where the trade ended up. It is the single most useful measurement for stop placement, because a stop is not hit by the final result, it is hit by the worst moment along the way.
Does this account for my strategy?
No, deliberately. Entries are sampled every fifteen minutes with no filter at all, which produces a neutral baseline for the instrument. That baseline is the thing your strategy has to beat. To measure your own stop-out rate you need to replay your actual setups, which is what the free backtester does.
Why does the answer change so much when I change the holding time?
Because adverse excursion grows with time, and it grows faster than most people expect. More time means more chances for price to wander against you before it resolves. A stop that is comfortable on a twenty-minute scalp can be inside the noise on a day trade, which is why the holding window selector matters more than any other input on the page.
Is the spread included?
No. Every figure comes from mid-price candle data, so your real-world stop is hit slightly sooner than the curve indicates — and noticeably sooner in thin hours when spreads widen. Treat the numbers as a floor rather than a precise forecast.
My stop keeps getting hit right before price goes my way. Is that a stop hunt?
Sometimes the level was genuinely obvious and got swept. Far more often the stop was simply placed inside the instrument's ordinary noise for that holding time, and would have been hit on a random entry too. Check the distance here first. If your stop sits below the 50% line on this curve, the market did not target you — the arithmetic did.
Random entries give you the floor. Your entries should beat it. Replay gold, GBPUSD or the Dow bar by bar, take your real setups, and record how far each one went against you before it worked — then set the stop from your own numbers.
Open the free backtester →One reproducible testing idea, with its rules, limitations, and review questions made explicit. In your inbox every week.
Beginner exploration
Open each answer for a plain-language way to read Stop Loss Reality Checker, test it carefully and decide what to explore next.
This page focuses on “Stop Loss Reality Checker”.Enter a stop distance for gold, GBPUSD, EURUSD, US30 or NAS100 and see what share of real holding windows would have hit it. Measured from 8,838,373 one-minute candles. Free, no sign-in.For “Stop Loss Reality Checker”, a beginner should identify what the interactive tool measures, assumes or teaches before acting on its conclusion.Treat this page's account of “Stop Loss Reality Checker” as a learning reference rather than a prediction, signal or promise of future performance.
For “Stop Loss Reality Checker”, run a baseline with inputs you can verify before experimenting with optimistic or extreme values.While exploring “Stop Loss Reality Checker”, change one input at a time so you can see which assumption moved the result.Keep your “Stop Loss Reality Checker” record honest: save the inputs beside the output because a number without its settings cannot be reproduced.Before leaving “Stop Loss Reality Checker”, treat the result as a scenario to investigate, not as an instruction to place a trade.
Turn one idea from “Stop Loss Reality Checker” into a rule with explicit inputs, dates, costs and pass-or-fail conditions.Ask AI to expose missing assumptions in that “Stop Loss Reality Checker” test, not to guess the next market move.Use the FXAbsolute AI Backtesting Lab to inspect calculations connected to “Stop Loss Reality Checker” and the assumptions behind them.Reproduce any important “Stop Loss Reality Checker” result and reserve unseen data before deciding that an apparent pattern is useful.
Continue your exploration of Stop Loss Reality Checker with the beginner AI prompt guide, or inspect public calculations in the AI Backtesting Lab.