Scaling Out And Taking Partial Profits

A position is up 15% and one of your three thesis legs has played out. Do you hold the whole thing for the other two legs, or sell it all and bank the gain? Most desks do neither — they scale out, selling a slice sized to the leg that resolved and keeping the rest on for the legs that have not.

Tie the slice to what actually resolved

The trap with partial profit-taking is doing it by feel: selling "some" because a position is up and that feels prudent. It is not a risk decision unless it is tied to the same thesis-leg logic used to size the trade in the first place. If a thesis had three legs and one has now confirmed and priced in, selling roughly a third and holding the rest for the remaining two legs keeps the position's risk proportional to what is still uncertain — the same logic as conviction-weighted sizing, run in reverse as the trade plays out.

This also converts an emotional decision into a mechanical one. "Sell a third because leg one confirmed" is defensible at a review six weeks later. "Sold some because I got nervous at the highs" is not, and it is indistinguishable after the fact from having no plan at all.

Worked example

You bought $1.2m of a stock at $40 on a three-leg thesis. It is now $47 (up 17.5%). Leg one — a product launch — has happened and the market has largely priced it; legs two and three, a margin expansion and a re-rating, have not yet shown up.

• Slice to sell: one of three legs resolved → sell one-third of the position.
• Shares sold: roughly 10,000 out of 30,000, at $47, realizing about $70,000 of gain on that slice (10{,}000 \times \7$).
• Remaining position: $940,000 (20,000 shares at current price), still exposed to legs two and three, with a stop that should now be reset off the higher entry basis of the remaining shares, not the original $40.

Two months later leg two confirms as well; another third comes off. What is left — the final third — is a pure bet on the re-rating, sized exactly as large as your remaining conviction in that one leg, no more.

What scaling out is not

It is not a substitute for a stop, and it is not a way to avoid making a decision. If a position is going against you, scaling out does not fix that — that calls for honouring the stop or cutting the whole thing. Scaling out is specifically for winners, and specifically tied to thesis resolution rather than to a price target you picked in advance, because a price target has no relationship to how much of the thesis is actually done.