What a trailing stop actually costs

2026-11-23

A trailing stop follows price at a fixed distance, moving only in your favour. It converts open profit into realised profit automatically — and it takes you out of trades that were going to keep going.

Both effects are real. Which dominates is a property of your trades, not of the technique.

How it works

Set a distance — an amount, a percentage, or a multiple of recent range. As price moves in your favour the stop follows at that distance. When price reverses, the stop does not move back. It is hit, and you are out.

The stop is therefore always the same distance behind the best price achieved. Which means you always give back that distance. That is not a flaw; it is the mechanism. But it should be priced.

What it costs

Every trailing stop exit gives back the trail distance from the high-water mark. On a trade that ran 4R and reversed, a 1R trail exits around 3R.

The systematic effects:

Average win falls. You exit before the move ends, by construction.

Win rate rises. Trades that would have round-tripped to your stop now close for a partial gain.

That pairing should look familiar — it is the same trade as partial exits, and it is evaluated the same way. The visible number improves, the one that determines returns may not.

Since expectancy is (win rate × avg win) − (loss rate × avg loss), moving both terms in opposite directions has an ambiguous sign. It has to be computed.

The distance is the whole decision

Too tight and you are stopped by ordinary noise, giving up the trades that would have run. The failure is invisible because you see a small win rather than a loss.

Too wide and you give back most of the move before exiting, which is close to not trailing at all.

The useful reference point is volatility, not comfort. A trail inside the instrument’s normal range gets hit by movement that carries no information — which is the same argument as why tight stops can cost more, applied to exits. A trail set as a multiple of recent range adapts when conditions change; a fixed percentage does not.

When it helps

Trends that run further than you would hold manually. If your discretionary exits are consistently early, a trail that is wider than your patience captures more.

When you cannot watch. A mechanical exit beats no exit.

Distributions with a long tail. If a minority of trades produce most of the return, a trail lets them run while capping the give-back. This is the strongest case.

When it hurts

Choppy conditions. Retracements are normal and a trail converts each one into an exit.

High-R strategies with few winners. If the edge lives in a small number of large trades, anything that truncates them is expensive — and a trail truncates every one.

As a substitute for a target. A trail exits when price reverses, which is not the same as exiting when the reason for the trade is exhausted. Those coincide less often than people assume.

The variant worth knowing

Break-even stop — moving the stop to entry once price has moved a set distance. It is a trailing stop with one step, and it has the same trade-off in sharper form.

It eliminates losses on trades that reached the trigger and reversed, which is the appeal. It also converts a meaningful fraction of eventual winners into scratches, because normal retracements reach entry more often than people expect.

The question is the same: how often does price reach your trigger, return to entry, and then go on to target? If the answer is “rarely”, break-even stops cost you winners for very little. If it is “often”, they are rescuing real losses. Measure it rather than assuming.

Measuring it on your own trades

You do not need to speculate. For your last sample, record for each trade:

The maximum favourable excursion — how far it went in your favour before reversing.
Where your actual exit was.
Where a trail of distance d would have exited.

Then compute expectancy for the actual exits and the simulated trailed exits. It is arithmetic, not opinion, and the answer is frequently surprising in both directions.

The prerequisite is recording the excursion, not just the result — which most journals omit. See logging tool calls for a trading audit trail.

A note on R

A trailing stop makes R ambiguous, because the risk that defined 1R is no longer the risk you are carrying once the stop has moved past entry.

The convention that keeps records comparable: 1R is always the original entry to original stop distance. A trade that trails out at 3R made three times the amount originally at risk, regardless of where the stop ended up. Recomputing R against a moved stop produces numbers that cannot be averaged with anything — the same problem as widening a stop.

FAQ

Is a trailing stop better than a fixed target?

Neither is generally better. A trail captures more of trades that run further than you expected and gives back the trail distance on every exit; a fixed target captures the planned move exactly and nothing beyond it. Which wins depends on whether your winners tend to exceed your targets, which you can measure from your own record.

How far should a trailing stop be?

Far enough to sit outside ordinary price movement for the instrument, which means setting it from volatility rather than from comfort. A trail inside the normal range is hit by movement carrying no information, and the cost shows up as small wins rather than as losses — which makes it easy to miss.

Does a trailing stop increase my win rate?

Usually yes, and that is not sufficient reason to use one. It converts trades that would have round-tripped into partial gains, which raises win rate while lowering average win. Expectancy combines both, and the sign of the change has to be computed rather than assumed.

Should I move my stop to break-even?

It depends on how often price reaches your trigger, returns to entry, and then goes on to target. If that path is common, break-even stops convert real winners into scratches; if it is rare, they eliminate losses cheaply. Both are measurable from your own trades, and the answer varies a lot by strategy.