Stops, Scaling, and Position Size

The most-quoted rule in trend following is "let your winners run, and cut your losses short." That second half — cutting losses short — is risk management. However good your stock selection, without it a few large losses will sink the account. And even with ordinary selection, disciplined risk keeps you alive for the next opportunity. Risk management isn’t optional; it’s the precondition for survival.

Why losses must be cut short — the asymmetry of recovery

Loss and recovery are not symmetric. Lose 10% and you need +11.1% to get back; lose 50% and you need +100%. The deeper the loss, the exponentially harder the recovery.

That’s why most trend followers cap the loss on any single trade at 7–8%. Cut it mechanically before that line and one mistake never compounds into irreversible damage.

Set the stop before you buy

The most important rule of stops: decide the stop before entering. Set it after buying and, the moment you’re underwater, the "just a little longer and it’ll come back" instinct keeps pushing the line lower. Before you click buy, define the level that says "if this breaks, my thesis is wrong." Common stops sit below the prior low or below a 50-day-line break.

Position size — where real risk control lives

Many equate risk management with the stop alone, but the more fundamental lever is position size. How much of the account you commit to a name determines the hit the whole account takes when the stop triggers. A common approach fixes the loss on any single stop at a set fraction of the account (say 0.5–1%).

Thinking in R — payoff ratio and expectancy

Call the maximum loss you allowed on a trade — the $100 above — 1R. Record every result in R units and your strategy’s true quality becomes visible: stops cluster near −1R, wins spread out at +2R, +5R, and so on. And here is the liberating arithmetic — you can be profitable while losing more often than you win.

A 40%-win-rate strategy that averages +0.4R per trade compounds nicely. This is why trend following is called a game of payoff ratio, not win rate — and why the discipline of always cutting at −1R is the pillar that holds the whole structure up. One neglected −5R loss cancels five good trades.

Scale in and out to ease the decision

Buying and selling all at once bets everything on a single decision. Scaling eases that. Enter part on the breakout and part on confirmed support; exit part into a sharp run-up and the rest when the trend breaks. Especially in a winning position, not selling the whole thing while the trend lives is how you "let winners run."

Trailing stops — moving the exit up to protect gains

Once a trade is profitable, the stop’s job changes. The initial stop (entry −7–8%) is the door out when you’re wrong; as the price advances you raise the line so it becomes the door that locks in what you’ve earned. Common practice: once the gain exceeds 1R, move the stop to breakeven; from there, trail the 20-day line (aggressive) or the 50-day line (roomier) as the exit trigger. Whichever line you choose, the principle is the same — ignore the daily wiggles while the trend lives, and let the exit fire automatically when the trend actually breaks.

What 2022 taught — regime-level risk

Sometimes per-stock stops aren’t enough. In the 2022 bear market the S&P 500 fell about 25% peak-to-trough and the Nasdaq well over 30%, and during that stretch even the best-looking breakouts mostly failed. Accounts with stop discipline escaped with a series of paper cuts; accounts that "held quality through it" took damage measured in years of recovery. Hence the second layer of risk control: when the market environment is hostile, cut back or stop initiating new positions altogether. A rule as crude as "reduce buying while the index is below its 200-day line" would have kept you out of most of the worst stretch. The market breadth article covers how to read that regime.

Selection and risk are one package

The two pillars of trend following are stock selection and risk management. Trend Screener automates the first half — objectively narrowing strong candidates. But stops, position size, and scaling are entirely yours. After you get a good candidate list, always design the "what if I’m wrong" scenario first, then enter. That’s the only way to last in the market.

Frequently Asked Questions

Where does the 7–8% stop rule come from?

It’s William O’Neil’s empirical rule: a correctly chosen leader bought at a proper pivot rarely retraces more than 8% if the breakout is real. A deeper drop usually means the entry was wrong. Note it’s a ceiling for well-placed entries — chart-based stops (below the pivot or prior low) are often tighter, and tighter is better.

My stopped-out stocks keep rebounding. Am I doing it wrong?

Rebounds after stops are an unavoidable cost of the method. What matters is the expectancy of the rule, not any single outcome. The memory of "holding through it worked" is seductive — and that same habit eventually delivers the −40% disaster. If a stopped stock sets up again, buy it again; re-entry costs a commission and some pride, which is cheap compared to a blown account.

Doesn’t scaling just multiply commissions?

At modern commission levels the added cost is trivial. The real value of scaling is psychological: removing the all-or-nothing decision makes both stopping out and holding winners far easier to execute. Beginners especially benefit from pilot buys — start with half the planned size, add only after the market confirms.

How many positions should I diversify across?

In trend following, diversification means "as many as you can genuinely manage," not "as many as possible." The O’Neil–Minervini school typically runs 4–10 concentrated positions. Spread across dozens and you can no longer track each trend — and returns converge to the index anyway. More important than the count is capping the total simultaneous risk: the sum you’d lose if every open stop triggered on the same day, kept to a few percent of the account.