Scalping

Scalping seeks small, frequent moves, often around the spread, the inside of the book, a one-minute level, or a quick reversion to VWAP. The holding time is short on purpose. The target is small, so the costs have to be smaller. Scalping is the most execution-sensitive style in this glossary. A trader with a slow feed, a wide spread, and a hesitant click gives the entire edge, if there was one, to the costs. Scalping is not "day trading but impatient." It is a method in which the spread, the queue, and the ability to flatten instantly are the strategy. The chart pattern is secondary.

Trade only instruments where the spread is a small fraction of the target and the book can absorb your size. Define the scalp in ticks or in a fraction of ATR so you do not let a scalp become a hope. The stop is as close as the noise allows and no closer. If the minimum stop that survives ordinary flicker is larger than the target, you do not have a scalp. You have a poor swing. Flatten when the tape changes, a large print, a halt risk, a spread that widens. A scalp that you "give a little room" is an unplanned larger trade at scalp size, which is backwards: the room should have been accompanied by a smaller size, and you are already in it.

Use scalping only with a platform, a size, and a session you have already measured. Record costs separately from gross points. A week that is green before costs and red after costs is a hobby that pays the venue. Cap the number of scalps so fatigue does not become the signal. If you are scalping because you missed the actual setup and want to "make something happen," you are not scalping. You are fidgeting, and the book will charge you for it.

Scalping a wide-spread or illiquid name because the chart looks active. Also judging the method by gross ticks. And adding to a scalp that went the wrong way, which converts a one-tick risk into a position you did not analyse. The correct losing scalp is a small loss, immediately.

Most retail scalping debates ignore queue position and adverse selection. The fills you get are the ones the other side was willing to give you. Backtests of one-minute bars cannot see this. Scalping also concentrates operational risk: a stuck order, a platform freeze, a mis-click. One operational error can exceed the week's planned profit. If you cannot afford that error, the style is too tight for the account.