Pyramiding — the position-scaling strategy for riding trends

In March 2023 a fellow trader opened a full-size short on USD/JPY — six lots on a hundred-thousand-dollar account. Price ran thirty pips against him, stopped him out, and only then did the real trend begin. The same session, another trader started with one lot, added half a lot once direction was confirmed, then 0.3 lots on the next breakout. The 1.8-lot equivalent position was held a week and closed for nearly three times the profit of the first trader. The difference came down to pyramiding.

What pyramiding is

Pyramiding splits the target trade size into three or four smaller tranches. The first — usually the largest, about half of the target — enters on the original signal. A smaller second tranche is added once direction has been confirmed, a third in the most powerful phase of the trend. The largest part of the capital works in a proven direction for the longest time, and the smallest part is exposed only in the phase of highest technical risk.

The principle is adding to winners. Each new tranche is allowed only when the existing position is already in profit and the trend structure remains intact. Adding to a losing position is averaging-down.

Classic scaling schemes

The 50-30-20 split on a concrete example

Illustrative example: a long EUR/USD position with a target size of one lot and an initial stop loss fifty pips below entry. The first tranche of half a lot enters on a breakout of local resistance at 1.0850, stop loss at 1.0800 (risk roughly €250 on a €10,000 account). The second tranche of 0.3 lots is added after a pullback to the 20-period moving average around 1.0890; the combined stop loss moves to 1.0840. The third tranche of 0.2 lots enters once the next swing high at 1.0930 is broken, with the stop trailed up to 1.0880. Weighted average entry is 1.08775; closed at 1.1000 the profit is €1,122.5. A full one-lot position opened at 1.0850 in one click would have produced €1,500 nominally, but in roughly seventy percent of comparable historical setups such a position is stopped out within the first two hours on a thirty-pip retracement.

“It is not whether you are right or wrong that’s important, but how much money you make when you’re right and how much you lose when you’re wrong. That’s why I add to winners and never to losers.” — Stanley Druckenmiller, in Jack D. Schwager, The New Market Wizards, HarperBusiness, 1992.

Risk cap and a combined stop loss

The most important rule of pyramiding is keeping a combined risk cap of one to two percent of equity on the entire scaled position, measured from the weighted average entry and the current stop loss. In practice this comes down to three rules.

• Every add requires trailing the stop up — without this, net exposure rises without compensation. The new stop sits at the most recent confirmed swing low or at the break-even of the previous tranche, plus a ten-pip buffer.
• After the third tranche the stop sits above the weighted average entry — the first two tranches are effectively risk-free, only the smallest, final tranche still carries downside. Total exposure may double the original target while the risk cap stays within one percent of equity.
• Two ATR multiples as a buffer — even in the strongest trend, price pulls back one to two ATR. The stop should sit at least two ATR away from the most recent high.

Add-on triggers in practice

Adding to winners only works with a precise definition of when to add. Without a trigger, pyramiding collapses into improvisation, and improvisation inside an open winning trade is the worst possible psychological combination — euphoria mixes with greed. Four mechanical triggers narrow that randomness: a pullback to the 20-period EMA or SMA on the hourly chart, a break of the next swing high by at least half an ATR, a breakout from a four-to-eight-bar intra-trend consolidation, and confirmation from a higher time frame (a close above key resistance on the daily or four-hour chart).

The most common mistakes

• Adding to a losing position. From the outside it looks exactly like pyramiding — the decisive difference is whether the position is in profit or in the red. This is averaging-down, the shortest path to bankruptcy.
• Failing to trail the stop loss after an add. A stop left in place after the second entry pushes combined risk from one to about 1.6 percent of equity; after a third tranche it climbs to two percent.
• Over-aggressive scaling. Equal or growing tranches raise exposure in the phase of higher technical risk — that is not pyramiding, it is an inverted pyramid.
• Ignoring correlation between currency pairs. Three long positions on EUR/USD, GBP/USD and AUD/USD behave like a single large short on the US dollar. Pyramiding on each independently multiplies the underlying currency risk by three.
• Adding during the climax of the move. The last tranche enters after a long run — precisely when the probability of a correction is highest. Without a clean trigger it often runs into the pullback and gives back the profit.
• Discretionary overrides of the scheme. “It is going so well, I’ll add more than planned” is the moment when a mechanical technique mutates into gambling. The split is binding even when euphoria says otherwise.

When pyramiding does not make sense

Position scaling is a trend tool. In range trading any add after a breakout is likely to be stopped out as price returns to the middle of the channel. In short-term scalping on five- or fifteen-minute charts moves are too fast for a three-stage technique. Around high-impact macro releases (US non-farm payrolls, FOMC, the inflation print) volatility doubles or triples and even two-ATR stops are swept by single candles. Pyramiding increases profit potential only when the underlying strategy already has positive expectancy on a single-entry baseline.

What to do tomorrow — your first steps

1. Write down the exact definition of the add-on trigger you will rely on for the next thirty days — pick exactly one of the four mechanical signals above and commit that no add will be executed outside that trigger, even when the chart looks tempting.
2. Decide on the maximum number of tranches (three or four, never more) and a concrete size split — 50-30-20, 60-25-15 or the Curtis Faith Turtle scheme. Pre-compute how many lots each tranche represents for the typical stop-loss sizes you use.
3. For a full month, trade pyramiding only on demo at the same notional size as your future live account, journaling every tranche and tallying results against a hypothetical single-entry baseline.
4. Add a section to your journal that captures the stop-loss math after each tranche — weighted average entry, new stop level, combined percentage risk after the add. That section is the best detector of emotional exposure increases.
5. Before going live, set a hard circuit breaker — if in any single week three consecutive pyramided positions are closed at a loss, revert to single-entry positions for four weeks and re-audit your triggers.

Related reading: the anti-martingale system as the foundation of sizing; trend-following systems as the context for pyramiding; ATR-based trailing stop for calibrating stops in changing volatility. For the wider context, see the risk management section on ForexMechanics.com.