Lessons 3 and 4 defined where to enter and where the trade is wrong. This lesson turns that distance into a position size. Position sizing decides how much a losing streak costs and whether the account survives long enough for the few large winners to arrive. It matters more to long-run results than the choice between one breakout rule and another.
Risk per trade and R
The starting point is a fixed fraction of the account that a single trade may lose if its stop is hit. Most systematic traders use between 0.5% and 1%; 2% is an upper bound for most. That amount is 1R. The position size follows from the distance between entry and stop:
Shares = (account equity × risk per trade) ÷ (entry − stop)
With $100,000 and 1% risk, the budget is $1,000. A $100 entry with a $96 stop risks $4 per share, so the position is 250 shares. The basic method, and why the stop comes before the size, is covered in stop-loss and position sizing; this lesson adds what is specific to trend following: volatility-based stops and sizes, adding to winners, and the limits that keep a portfolio of breakouts from becoming one large bet.
ATR stops: let volatility set the distance
A stop placed inside a stock's normal daily noise gets hit by noise. The average true range (ATR), J. Welles Wilder's measure of the typical daily range including gaps, scales the stop to each stock: a common initial stop for breakouts is 2 × ATR below the entry. A quiet stock gets a tight stop, a volatile one a wide stop, and the size adjusts so that both risk the same amount.
The calculation of ATR and the chandelier variant are explained in ATR stop. Two practical points for breakouts: a structural stop (under the base or the setup day's low) and an ATR stop can be combined by taking the higher of the two, as several presets do, and any stop should be capped as a percentage of price, because a very volatile stock can otherwise justify a stop 20% away.
Volatility sizing: the Turtles' N
The Turtles took the idea one step further. They called the 20-day ATR N and sized every position in units so that a move of 1 N changed the account by 1%. Their stop was 2 N, so a full loss on one unit cost about 2%. Because every unit had the same volatility in money terms, a quiet market and a wild one contributed equally to the portfolio's swings.
For stocks the translation is direct: shares = (equity × fraction) ÷ ATR. The fraction must be smaller than the Turtles used, for two reasons. Their futures accounts could hold positions larger than the account itself; a cash stock account cannot. And individual stocks gap on earnings and news in ways that diversified futures rarely did. A useful check is the position's share of the account: at 0.5% per ATR, a stock with an ATR of 2% of its price already becomes a position worth 25% of the account.
Pyramiding: adding to winners
Trend followers add to positions that work, never to ones that do not. The Turtles added a unit each time price rose ½ N above the previous entry, up to four units, and raised the stop on every unit to 2 N below the newest one. Pyramiding lets a strong trend become a large position without committing the full size at the riskiest moment, the breakout.
The figure shows the hidden cost. After four adds in a rising market, a full reversal to the stop loses 2.5% rather than 1%, and the position has grown to more than half of the account. Pyramiding increases both risk and concentration precisely when a trend is mature. Safer rules for stocks:
- add only after the first position's stop can be raised to break-even or better;
- make each add smaller than the one before, so the average cost stays close to the first entry;
- add on a new, valid setup (a breakout from a new base or a volume pivot), not on a fixed price step;
- keep the total loss if every stop is hit below a cap you set in advance.
Portfolio limits and losing streaks
A breakout system often triggers in several stocks of the same industry in the same week. Five semiconductor breakouts at 1% each are not five independent 1% risks; they are close to one 5% bet. The Turtles capped the number of units in one market, in closely related markets and in one direction. The stock equivalent is a cap on positions per sector and on total open risk, often called portfolio heat, for example 5% to 8% of equity across all stops.
Risk per trade decides how deep an ordinary losing streak goes. At 1%, ten full losses in a row cost about 9.6%; at 5%, they cost about 40%, which requires a 67% gain to recover. Since lesson 1 showed that streaks of eight or more losses are likely over a few hundred trades, sizes that look modest per trade can be dangerous per streak. The Turtles also cut their trading size by 20% for every 10% drawdown, which slows the damage while the system is out of step with the market.
How to apply it
Entry
Before the order, compute the size: risk budget ÷ (entry − stop), then check the result against a maximum position (for example 20% of equity) and against sector and portfolio limits. If the position is capped, accept the smaller risk rather than tightening the stop.
Stop
Use the higher of a structural stop and 2 × ATR below the entry, capped at a set percentage (8% to 10% is common for breakouts). Set it on entry and never move it down.
Target
Sizing does not set a target. Profits are managed by the exit rules in lesson 6.
Invalidation
If a gap opens below the stop, exit at the open; the loss will exceed 1R, which is why the per-trade risk must leave room for it. Skip entries just before earnings if a gap through the stop would cost more than you are willing to lose.
In Clear Tape
Each preset on the Trading systems page reports its entry, stop, target and the reward-to-risk ratio with every signal. The stops are fixed by the rules: 20-Day Breakout, entry − 2 × ATR(20); 52-Week High Breakout, the higher of the day's low and entry − 2 × ATR(14), capped at 10%; Volume Base Breakout, 8% or one cent under the 10-day low if higher; NR7 / Inside-Day Breakout, one cent under the setup day's low; Volume Pivot, one cent under the day's low, capped at 8%. The backtests size positions the way this lesson describes: $100,000 to start, each trade risking a set share of current equity (1% by default, adjustable from 0.1% to 5%), shares = equity × risk ÷ (entry − stop), at most 20% of equity in one stock, no margin, and a signal skipped when there is not enough cash. The presets do not pyramid, so the backtests measure single entries. Rerunning a backtest at 0.5%, 1% and 2% shows directly how risk per trade changes the drawdown. Backtests are available to Tools and Pro members.
Common mistakes
- Choosing the number of shares first and placing the stop wherever the loss "feels" acceptable.
- Copying the Turtles' 1% per N into a stock account, where it produces positions larger than the account.
- Pyramiding by fixed price steps until the position dominates the account.
- Treating several breakouts in one sector as independent trades.
- Raising risk after a losing streak to "win it back".
Summary
Position size follows from the stop: risk budget divided by the distance to the stop. ATR stops adapt that distance to each stock, and volatility sizing gives every position a similar effect on the account. Pyramiding can turn a strong trend into a large gain but raises risk and concentration, so it needs strict rules. Portfolio limits and a modest risk per trade keep normal losing streaks survivable. The next lesson covers how to exit once the trend is working.
For educational purposes only. Not investment advice.