Scaling in: how adding to a trade changes the risk math

[Amanhasnoname]

Adding to a winning position seems prudent, since it reinforces a correct idea. Scaling in to a losing one is averaging down, which is a different matter. Both change the risk profile in ways that are easy to underestimate.

When adding to a winner, the average entry price rises, and the stop for the whole position may now sit at a worse level relative to the new average. The risk on the combined position can exceed the original.

Costs also increase: each add pays a spread and commission.

For scalpers, the time window is short, so additions happen quickly and under pressure. Position size may balloon before the trader realizes it.

A disciplined approach pre-defines the maximum total size, the conditions for adding, and the stop for the aggregate. The risk per idea stays within the plan.

Test the scaling rules on tick data. Compare the equity curve and drawdown with the single-entry version. Often, scaling raises variance more than it raises return.

What rules, if any, govern adding to positions, and how were they tested?