BUILD: COURSE 2 | LESSON 4

Scaling, partials and trailing stops

Learning objectives

Calculate the blended risk and R-multiples of scaled entries, partial exits and trailed stops instead of guessing their effect.
Explain what each technique actually trades away (and buys) in expectancy terms — none of them is free money.
Choose one exit-management scheme, define it with exact rules, and keep it constant long enough to measure.

So far the course has treated a trade as one entry, one stop, one target. Real trade management offers three levers between entry and exit: adding to a position (scaling), taking part of it off (partials), and moving the stop (trailing). Each is sold somewhere online as the trick that transforms results. The honest version: every one of them redistributes your P&L between win rate and average win — they cannot create expectancy, only reshape it. Reshaping can still be valuable (smoother equity curves are easier to trade well), but only if you know the arithmetic and apply one scheme consistently.

Scaling in: averaging into strength vs averaging down

Scaling into winners means planning a position as tranches that are added as the trade proves itself. Example: your plan allows 1% total risk on a EUR/USD long. Instead of 0.30 lots at 1.0800 (stop 1.0750, ~$150 on a $15,000 account), you take 0.15 lots at 1.0800, and add 0.15 at 1.0830 with the whole position's stop moved to 1.0790.

Tranche 1: 0.15 lots, entry 1.0800, stop 1.0790 → risk 10 pips × $1.50 = $15
Tranche 2: 0.15 lots, entry 1.0830, stop 1.0790 → risk 40 pips × $1.50 = $60
Total open risk after adding: $75 — half the single-entry version, with the same upside if the move continues.

The cost: when price runs without pulling back to fill your add, you profit on half size; and when it triggers the add then reverses, you've converted an unrealised winner into a scratch or loss. Scaling in lowers risk per unit of conviction but lowers your average win on clean moves — a trade-off, not an upgrade.

Averaging down — adding to a losing position to "improve the average price" — is the same mechanic pointed the wrong way. Each add increases size exactly when the market is telling you the idea is wrong, and it breaks the position-sizing discipline of P2.2: your carefully computed 1% becomes 2%, then 4%, with the stop either widened or abandoned. It is the signature behaviour of blown accounts (and of martingale systems, which are averaging down with a formula). Rule for this academy: adds are only ever pre-planned, only ever with the total position inside your original risk budget, and only in the direction the trade is working.

Partial exits: what "taking half at 1R" really costs and buys

The most popular scheme: close 50% at +1R, move the stop to breakeven, let the rest run to +3R. Feels perfect — you "lock in profit" and get a "free trade". Do the numbers on the three outcomes:

Price hits 1R then reverses to your breakeven stop: +0.5R × 50% banked = +0.5R total (versus −1R for the all-or-nothing trader — this is the scenario partials are for).
Price runs to 3R: 0.5R (first half) + 1.5R (second half) = +2.0R (versus +3R full size — you gave up a third of your best outcome).
Price never reaches 1R and stops out: −1R either way.

So partials raise your win rate and cut your average win. Whether expectancy goes up or down depends entirely on how often the market gives you 1R-then-reversal versus clean 3R runs — which is an empirical property of your setups, not a law. Two honest observations: psychologically, partials make systems much easier to follow, and a system you actually follow beats a theoretically better one you abandon (P2.3's circuit-breaker logic again). Statistically, the breakeven stop is the sneaky expectancy leak — markets routinely retrace to entry before continuing, and "it was a free trade" often means "I systematically ejected from my best trades at +0.5R". If you use partials, journal the exit of both halves so you can see, after 50 trades, what the second half actually earned.

Trailing stops: renting a lottery ticket on trends

A trailing stop moves your exit in the trade's favour as price advances, converting "target" into "follow until it turns". Common mechanics, from tightest to loosest:

Swing trailing: stop below each new higher low (longs) — structural, adapts to the chart.
ATR (chandelier) trailing: stop = highest close since entry − k × ATR, k typically 2–3. Volatility-adaptive, mechanical, backtestable — the natural partner of P2.2.
Moving-average trailing: exit on close beyond the 20/50 EMA. Simple, but lags badly in fast reversals.
Fixed-pip trailing (the platform default): steps the stop a constant distance behind price. Ignores volatility; usually the worst of the four for exactly the reasons fixed-pip stops were in P2.2.

The universal arithmetic: trailing lowers win rate (many trades that would have hit a fixed 2R target get trailed out at +1.2R when a pullback exceeds the trail) in exchange for occasional outsized wins (+5R, +8R) that a fixed target can never capture. Whether that trade is worth it depends on how often your market trends. A tight 1-ATR trail in a ranging market is a machine for converting winners into scratches; a 3-ATR trail in a strong trend can double a system's expectancy versus a 2R target. There is no universally right answer — which is exactly why the trail rule must be fixed in advance and tested (P3.5), not improvised per trade. Improvised trailing is just fear with a user interface.

Pick one scheme and freeze it

The worst exit management is the one that changes trade to trade, because (P2.1) an inconsistent process has no measurable expectancy. So end this lesson with a decision, written into your trade plan:

Entries: single entry, or a pre-defined 2-tranche scale-in with total risk ≤ your per-trade %. No unplanned adds; never against the position.
Exits: choose ONE — (a) fixed target at ≥2R; (b) 50% at 1R + trail the remainder by 2.5 × ATR; (c) full position trailed by swing lows. Write the exact rule, including when (if ever) the stop moves to breakeven.
Freeze it for 30 trades, journal every exit against the rule, and only then compare: what would each alternative have earned on the same trades? Your journal — not this lesson, not a forum — will tell you which reshaping suits your setups. That habit of testing-before-trusting is the entire subject of Course P3.

Key takeaways

Scaling, partials and trailing redistribute P&L between win rate and average win; they cannot manufacture expectancy that isn't in the setup.
Scale-ins reduce open risk on unproven trades but shrink wins on clean moves; adds must be pre-planned and inside the original risk budget. Averaging down is the same lever reversed — maximum size on the worst ideas.
"Half off at 1R, stop to breakeven" turns −1R reversals into +0.5R but caps your best trades (e.g. +3R → +2R); the breakeven stop is a common hidden leak.
ATR/swing trails trade win rate for occasional large R-multiples — powerful in trending conditions, corrosive in ranges; fixed-pip trails are usually worst.
Choose one written management scheme, hold it constant for ~30 trades, and let journal data — not in-trade emotion — decide changes.