How to Scale Into a Position
To scale into a position, decide the total size first and enter it in planned instalments. Adding as the trade works keeps the risk bounded; adding as it goes against you increases the exposure at the moment the original reason is being disproved.
Scaling in means building a position in instalments rather than all at once. Done with a planned total and a single stop it is a reasonable execution technique. Done without those it is a way of ending up with a position nobody chose.
Before you start
A total size decided before the first entry, so adding is a schedule rather than a decision. The finished position is known at the start.
A rule about which direction you add in, written down. As it works, or as it goes against you. These are opposite strategies with opposite risk profiles.
A single stop for the whole position, recalculated as it grows. Not one stop per instalment, which is several trades pretending to be one.
The steps
1. Decide the finished size first
Whatever your normal sizing arithmetic produces. That number is the ceiling, and scaling is how you reach it rather than a way past it.
2. Split it into few, large instalments
Two or three parts. On this site’s shared series a round trip measures about 2% of the median bar range of 0.493, and each instalment pays it.
3. Add as it works, not as it fails
The second instalment goes on after price has moved your way and the original reasoning is holding. That keeps the largest part of the position in the trades that are working.
4. Recalculate the stop for the whole position
After each addition, the combined position needs a single invalidation level and the total risk has to still equal your intended figure. Usually that means moving the stop up as you add.
5. Never let the total exceed the planned size
A trade going well is the most persuasive argument for exceeding it and the situation the ceiling exists for. Scaling is an entry technique, not a licence.
6. Stop adding once the risk figure is reached
If the recalculated stop means the combined position risks more than your figure, no further addition is available regardless of how the trade looks.
7. Treat it as one trade in the record
One entry in the journal with an average price. Recording each instalment separately makes a single idea look like three trades and distorts every statistic downstream.
How to tell it worked
The total size was written before the 1st entry, and never exceeded.
The position was built in at most 3 trades, not eight small ones.
A single stop covers the whole position, recalculated at each addition.
And total risk stayed at your intended figure throughout, checked after every addition.
Adding to winners against adding to losers
Adding as it works concentrates size in trades that are already right. The risk stays bounded because the stop moves up with the additions, and the worst case is a working trade turning into a small loss.
Adding as it fails concentrates size in trades that are being disproved. The average price improves, the position grows, and the stop distance widens — three things moving the wrong way at once, in a trade whose original reason is currently not working.
Why the instalment count matters
Each part costs a round trip. Eight small entries pay eight spreads to build one position, which on a short-horizon trade can exceed the benefit of the careful entry.
And more parts means more decisions. Each addition is a moment where the plan can be renegotiated, and the whole point of deciding the total in advance was to have fewer of those.
Two or three is where the technique pays. Enough to avoid committing everything at one price, few enough that the costs stay small and the schedule stays a schedule.
What the schedule should look like
Write it as prices, not as feelings. First instalment at the entry level, second after price has travelled a stated distance in your favour, third after another. Three numbers, decided in advance.
Use the instrument’s own movement to set the spacing. On this site’s shared series the ninetieth percentile bar range is 1.101, so spacing additions closer than that means adding on noise rather than on progress.
Decide what cancels the remaining instalments. Usually the stop being hit, sometimes a time limit — but something, or unfilled additions sit as open intentions indefinitely.
And write the schedule down before the first order. A plan produced after the first instalment works is not a schedule; it is the trade going well being used as a reason to commit more, which is the specific failure this whole structure exists to prevent.
The original data
Of the 24,971 unique videos in research/search-study-corpus.jsonl, 5 mention scaling in the title,
at a median of 5,240 views across 5 channels — and only 20% of those are instruction-shaped. Position
sizing appears in 34 at 2,037 and risk management in 410 at 4,079. The counts come from
site/corpus_count.py and site/rank_howto.py.
5 videos in 24,971 on a technique that appears in almost every serious method. Effectively no coverage, and what exists is mostly descriptive rather than instructional — which leaves the distinction between the two directions of scaling almost entirely untaught.
The answer to the question on that chart is that the better price is the market disagreeing with you. Adding there increases exposure while the original reasoning is being contradicted — which may still be right, and is a much larger claim than the first entry was.
When it fails
The failure is scaling in without a planned total, and it produces positions nobody sized. The first entry is small because the setup is uncertain. It works, so more goes on. It works again, so more goes on. Each addition is justified by the trade doing well, none of them was in a plan, and the finished position is several times what the account should hold in one idea — discovered only when it reverses.
The second failure is adding as it goes against you. Exposure grows while the reason weakens.
A third is a stop per instalment. That is several trades, not one.
A fourth is too many parts. The costs outweigh the care.
A fifth is exceeding the planned ceiling. The ceiling was the point.
And a sixth is recording each part separately. Every downstream statistic is then wrong.
Related
Position sizing is where the total comes from. Anti-martingale is the principle behind adding to winners. And risk per trade is the figure that must not change as the position grows.
The rule that keeps this honest is that the finished position can never be larger than the one I would have opened in a single order. Scaling is a way of entering the same size more carefully — the moment it becomes a way of ending up with a bigger one, it has turned into something else.
— Michael Whitman
This page is educational, not financial advice. Test every idea on your own charts before risking money.