Home · Learn · Scalping strategy
Almost everything written about scalping is about entries. The entry is the least important of the five decisions a strategy has to make, and the other four are where accounts are actually won and lost.
Ask most traders to describe their strategy and you get a description of an entry: a pattern, a moving-average cross, a level. Ask what size they take, where the stop sits relative to the day's range, and what happens if the trade is still open at the close, and the answers get vaguer — often improvised in the moment.
That asymmetry is the problem. A strategy is a complete specification of five things, decided in advance:
| Decision | Specified in advance? | Usually |
|---|---|---|
| When you are allowed to trade | Must be | rarely is |
| Direction and entry | Must be | almost always is |
| Position size | Must be | improvised by feel |
| Stop | Must be | set, then moved |
| Target and exit | Must be | cut early on green |
Change any one of them and you are no longer trading the strategy you tested. That is not a figure of speech — a system taken at half size on the trades that feel risky and double size on the ones that feel obvious has a completely different expectancy from the one in its backtest, and the backtest no longer describes anything.
The most commonly skipped specification and one of the most consequential. Liquidity, volatility and the character of the move all change through the session, and a strategy validated on the US cash open has not been validated at midday or overnight.
Test by session rather than pooling every hour into one number. A strategy that is strongly profitable in one window and mildly unprofitable in another looks mediocre in aggregate and good once you stop trading the wrong window. Session structure on the Nasdaq contract is covered here.
The part everyone works on, and the part with the least leverage over your results. It matters — a strategy with no edge is not rescued by good risk management — but the difference between a good entry and a slightly better one is small compared with the difference between correct size and improvised size.
The property worth insisting on is that the entry is defined: a price, not a zone you interpret under pressure. "Around the level" is not a specification, and two traders following it will take different trades.
Where most accounts are actually lost. Size is not a confidence dial — it is a calculation with three inputs: the most you will lose on a trade, the stop distance, and the contract's value per point.
Two pages carry the arithmetic: which contract your stop distance permits, and how many consecutive stops your account survives — a number that, on a funded account, does not improve at all until the drawdown floor locks.
A stop is a commitment made before the trade, not a setting adjusted during it. The single most destructive habit in scalping is widening a stop that is about to be hit, which converts a planned small loss into an unplanned large one and quietly halves the number of losses your account can absorb.
Worth deciding in advance, in writing: what happens if the stop is hit twice in a row, and whether there is a daily loss limit that ends the session regardless of what the chart is doing.
Cutting winners early is the mirror image of widening stops, and it is more insidious because each individual instance feels like prudence. A system with a defined target that you exit at 60% of the way there has had its reward-to-risk ratio cut without its loss size changing — which is often enough to turn a profitable system into a losing one while every individual decision felt sensible.
Whether you build a strategy or buy one, the same checks apply:
Most traders who fail do not fail because their entries were bad. They fail because they made four of the five decisions in the moment, under pressure, with money moving — and made them differently each time.
That is why "just be disciplined" is useless advice. Discipline under pressure is not a character trait you can decide to have; it is a design problem. The strategies people manage to follow are the ones that leave the fewest decisions available to interfere with.
That design problem is the one thing Falcon addresses: it answers all five before the alert reaches you, so there is nothing left to improvise. Whether you want that or would rather build your own is a genuine preference — but the five decisions have to be answered somewhere, by someone, before the trade starts.
See how the five arrive as one ticket