TRIED & TRADED
← Field guide · exits

Why the exit changes the answer more than the entry

  • Two traders can run identical entries for a year and disagree about whether the strategy works.
  • The exit is usually the reason, and it is the part most strategy write-ups leave as one clause at the end.

The mechanism

An entry decides which trades you take. An exit decides what each one is worth — and it makes that decision once per trade, on every trade.

  • Change the entry and you get a different set of trades, most of them still broadly similar in character.
  • Change the exit and you keep every trade you had, but each one now pays a different amount.
  • The second change touches the whole sample. The first one only touches the trades at the boundary.
  • That is why the exit moves the result harder. It is arithmetic, not a finding.

The same position, two answers

Take the bull put spread this site uses as its worked example — short the 95 put, long the 90, 45 days out, $93.20 credit. Now put the stock exactly at $95, the strike it sold.

When you lookThe position is worthWhat that is
Day one, 45 days left−$94.53worse than the entire credit you took in
21 days left−$71.37still a loss
7 days left−$26.62nearly back
Expiry day+$93.20the maximum win
  • Same stock price in every row. The only thing that changed is when you looked.
  • An exit rule is a decision about which row you get paid on.
  • A stop that fires on day one books a loss on a trade that finishes at maximum profit.
  • A hold to expiry collects the lot and carries all the risk in between.

The expiry chart is the picture everybody knows. It is also the row you are least likely to be standing on.

Arithmetic from the specification above at 25% volatility and a 4% rate. Not a measured result.

Where it bites

Two studies of the same strategy are not comparable

  • One closes at half the credit, one holds to expiry, and both call themselves a test of the same trade.
  • Their headline numbers will differ, and neither is wrong.
  • Comparing them is the most common mistake in published options research, including by people being careful.
  • If a study doesn’t name its exit, assume it held to expiry — that is the cheapest one to model.

And it makes the entry look more important than it is

  • Entries are visible, arguable and fun to optimise.
  • Exits are boring and get one line.
  • So the effort goes into the half of the strategy that moves the number least.
What this site does about it

Every test article names its exit in the specification, before the run, alongside the delta, the tenor, the width and the minimum credit. And the queue includes a run whose whole purpose is this question: one set of entries, closed seven different ways, on one scale.

Common questions

Does this mean the entry doesn't matter?

No. A bad entry can’t be rescued by any exit. The point is narrower: for a given set of entries, the exit is the bigger lever on the result.

How much difference does the exit actually make?

We haven’t published a number and won’t until the run is done. What this page states is the mechanism, which is arithmetic and doesn’t need a test to be true.

Is there a best exit?

Not in the abstract. It depends on the structure, the tenor and how quickly you can redeploy the capital. Anyone who names one without naming those has skipped the question.

Why do so few articles cover exits?

They are harder to write about — there is no chart of an exit, and the interesting part only shows up across a whole sample of trades.

How this shows up in our tests

This is the finding the site is built to produce, and it is not published yet.

The exit ladder run takes one set of entries and closes them seven ways — profit target at three levels, stop, time exit, expiry, technical, assignment and roll — then reports expectancy per trade for each. Until those numbers exist, this page states the mechanism and nothing more.

The hub

The seven ways
to exit

The umbrella

What is an
exit rule?

The exit

The profit
target