Stop Loss Distance Tested: 1,007 EURUSD Trades, One Entry, Seven Stops

Almost every test in this series measures an indicator. This one deliberately does not. I froze the entry — a plain EMA 12/26 crossover on EURUSD 5-minute bars — and changed only one thing: how far away the stop loss sat. Same 16,926 bars from June 3 to August 26, 2026, same data behind my moving average crossover test, 1,007 trades in total across the grid.

I expected the stop to be a detail. It turned out to be the whole result.

How was the stop distance tested?

• Entry: EMA 12 crosses EMA 26, long or short, entered at the close of the signal bar. Never varied.
• Stop: a fixed pip distance, tested at 5, 8, 10, 15, 20, 30 and 40 pips.
• Target: two times the stop distance, so every column is a genuine 2R trade rather than a different risk-reward in disguise.
• Exit: stop, target, or 24 hours, whichever comes first. If a bar touches both stop and target, the stop is assumed to have hit first — the pessimistic reading.
• Overlap rule: a new trade only opens after the previous one has closed, so nothing is double counted.

The pip figures below are gross. I also show the total after subtracting one pip per round turn for spread, because on 5-minute forex that cost is not a rounding error.

What did each stop distance actually pay?

Read the win rate column and the money column side by side, because they disagree in the most instructive way possible. The 40 pip stop has the best win rate in the table, 52.5%, and it is the worst line in the study. The 15 pip stop wins less often and makes the most. Win rate on its own tells you nothing about a stop — it only tells you how much room you gave the trade before you were forced out.

The trade count column tells the other half. A 5 pip stop produced 349 trades because most of them ended within twenty bars; a 40 pip stop produced 59 because each one tied up the account for the better part of a day. Same signal, six times the activity, all of it decided by a number typed into an order box.

Why did the tight stop fail so badly?

Because of scale. The median ATR(14) on these 5-minute bars is 2.0 pips, with the busiest ten percent of bars at 3.2. A five pip stop is therefore two and a half bars of completely ordinary movement. It gets hit by the market doing nothing in particular.

That table is why 66.8% of the 5 pip trades ended as losers. It is also why the same trap catches people who tighten stops after a losing streak: the tighter the stop, the more of your outcome is decided by noise instead of by the signal you spent so long choosing. The hour of the day you trade moves that noise level around as well, which makes a single fixed stop even harder to defend at the extremes.

Did the target matter as much as the stop?

No, and this surprised me. Holding the stop at 15 pips and moving only the target:

The spread between the best and worst target is 4.4 pips per trade. The spread between the best and worst stop was 7.4. The stop is the bigger lever, even though the target is what most people argue about.

The 5R row deserves a note of its own: its numbers are identical to running the same trades with no target at all. In 80 trades a 75 pip target was never reached inside 24 hours — not once. The median EURUSD day in this sample only covered 48.4 pips top to bottom, so a 75 pip intraday target is not ambitious, it is arithmetically unavailable.

Would a better stop rescue a worse entry?

I ran the identical grid on a second entry, a 20-bar Donchian breakout, to check whether I had found something about stops or something about that one EMA cross.

Flat. Every setting sits within half a pip of zero. Stop placement amplified an entry that had something in it and did nothing at all for one that did not — the same lesson that showed up when filtering signals with ADX above 25 improved nothing that was not already there. Exits are a multiplier, not a source.

What I would actually do with this

• Set the stop from volatility, not from comfort. On EURUSD M5 the working zone was 10 to 20 pips, roughly five to ten times the median ATR.
• Stop tightening stops to raise your win rate. The best win rate in the table belonged to the worst strategy.
• Do not use a five pip stop on this timeframe. Two and a half bars of noise is not a risk decision.
• Check your target against the daily range. Median EURUSD day: 48.4 pips. Size targets inside that, not outside it.
• Fix the entry before tuning the exit. A flat entry stayed flat at every stop distance I tried.

Frequently asked questions

What is the best stop loss distance for EURUSD on a 5-minute chart?

In this test, 15 pips: 101 trades, 44.6% win rate, +3.38 pips per trade and profit factor 1.43, with 20 pips close behind. That is specific to EURUSD M5 in a period whose median ATR(14) was 2.0 pips. On a quieter instrument the equivalent distance is smaller, on gold it is far larger.

Is a tight stop loss better because the loss is smaller?

Not on this evidence. The 5 pip stop kept individual losses tiny and still finished at -0.03 pips per trade before costs and -358.6 pips after spread, because it converted trades that would have worked into small losses. A stop is only “tight” relative to how far the market routinely moves.

Should the take profit be 2R or 3R?

2R was the better of the two here — +3.38 against +2.59 pips per trade — but the gap between them is far smaller than the gap between a good and a bad stop distance. Get the stop right first.

Does this apply to a fixed stop or an ATR based stop?

Everything above uses a fixed pip stop. The ATR version behaves very differently on a 5-minute chart, and not in the direction most guides suggest — I ran that comparison separately in fixed stop against ATR stop.

How do I turn a stop distance into a position size?

Risk in dollars divided by stop in pips gives your value per pip, and that gives the lot size. The full table for a $10,000 account, plus where to type it in the platform, is in my position sizing guide for MT4 and MT5.

What data was this run on?

16,926 EURUSD 5-minute bars, June 3 to August 26, 2026 — the same set behind my Ichimoku cloud breakout test, my RSI(7) against RSI(14) test and my Stochastic overbought test.