Scaling In and Out
Building or reducing a position in pieces rather than at once. It changes the distribution of outcomes and it is frequently confused with averaging down.
MadStockAlerts Research · Updated August 28, 2026
What to take away
- Scaling in builds a planned position in tranches decided in advance.
- Averaging down adds to a losing position, which is a different decision.
- Scaling out reduces variance and reduces the average exit price in a trend.
- The full position size must be decided before the first tranche.
- Costs rise with the number of transactions.
MAD Academy Training Video · 0:45
Splitting the Decision Into Pieces
Scaling trades a better average for a smaller position, and both directions have a cost worth naming.
This lesson is part of a Stock Alerts + Tools plan.
The distinction that matters
| Scaling in | Averaging down | |
|---|---|---|
| Decided | Before the first entry | After the position moved against you |
| Total size | Fixed in advance | Growing |
| Trigger | A defined condition for each tranche | The price having fallen |
| Risk | Known from the start | Rising with each addition |
The second row is decisive. Scaling in cannot increase the risk beyond what was planned, because the total was fixed. Averaging down increases it every time, and the increases occur precisely when the original reasoning is under question.
What scaling in costs and buys
- It reduces the consequence of a single badly timed entry, which is the main argument for it.
- It produces a worse average entry in a position that works immediately.
- It multiplies transaction costs by the number of tranches.
- It requires a defined condition for each tranche, or it becomes discretionary adding.
The fourth point is where it usually breaks down in practice. Without a written condition for each addition, the tranches get added on a feeling, which is averaging down under a different name.
Scaling out
Reducing a position in pieces has a different profile from exiting at once: it guarantees participating in some of a further move and guarantees not participating fully.
Scroll the chart sideways to see all of it.
- Exit all at once
- Scale out in thirds
The behavioural argument is stronger than the mathematical one. Taking part of a position removes the pressure that causes a full exit at the first pullback, and holding some of it is what allows a large move to be captured at all.
The rule that keeps it honest
The full intended size is decided before the first tranche and the risk is measured against that total. A plan that sizes the first tranche and leaves the rest open is not a scaling plan; it is a position with no defined size.
The arithmetic of an average entry
Scaling changes the average entry price, and the effect on risk depends on whether the exit moves with it.
| Approach | Average entry | Risk if the exit is unchanged |
|---|---|---|
| Full size at once | The entry price | As planned |
| Thirds, at higher prices | Above the first entry | Larger, since more shares are above the exit |
| Thirds, at lower prices | Below the first entry | Larger in total, though each share risks less |
Both scaling rows increase total risk relative to the first tranche alone, which is the point that is frequently missed. The risk is measured against the full intended position, and a plan that measures it against the first tranche has understated it by the number of tranches.
The corresponding discipline is to compute the risk on the full intended size before the first entry, and to size each tranche so that the total lands where it was meant to.