What is trend following?

Trend following does not try to forecast. It waits for price to show a trend, joins it, and lets the market decide when to get out. The old trading saying “cut your losses short and let your profits run” is the whole idea in one line: a stop caps the losses, and no fixed target caps the gains.

It is usually applied on daily or weekly charts, with trades lasting weeks to months, and across many markets at once. Many managed-futures funds (often called CTAs) run trend-following systems, and the Turtle trading experiment of the 1980s is its best-known story. It is the opposite bet to mean reversion, which expects stretched prices to snap back.

Simple trend-following rules: filter, entry, size, exit

A trend-following system is a short list of rules decided before the trade, as in the flowchart below. The numbers are common examples, not settings to copy:

  1. Filter: trade long only when price is above its 200-day moving average, or when the 50-day is above the 200-day (see the moving average crossover). Higher highs and higher lows in market structure say the same thing without an indicator.
  2. Entry: buy a close at a new 50-day high, a channel breakout in the style of the Turtles.
  3. Size: place the stop a multiple of the average true range (ATR) away, then size the position so that the stop costs a small, fixed share of the account.
  4. Exit: trail the stop, for example 3 × ATR under the highest close or under the 20-day low, and sell when it is hit. Never lower it.

Turn your phone sideways to read the drawing. Tap to close.

A hand-drawn flowchart read left to right under four headers, filter, entry, size and exit: price above its 200-day average? (no: stay out), close at a new 50-day high? (no: wait), then buy: size from the ATR stop, then trail the stop, exit when it is hit. A grey note reads the numbers are examples, not settings to copy.

Why most trend-following trades lose (and why that is the design)

Most breakouts do not become trends, so most trend-following trades end as small losses at the stop, as in the drawing below. The few that catch a long trend are held for many times the initial risk, and they are meant to pay for the rest. Hypothetical: one trend that runs to +10R covers ten separate −1R losses.

That only works if both halves hold. The losses must stay near 1R, so the stop is never moved down, and the big winners must be allowed to run, so profits are not taken early. A trader who skips signals after a few losses, or grabs a quick profit, can end up with all the small losses and none of the large gains. Measuring results in R makes this visible; the risk/reward ratio lesson explains R.

Turn your phone sideways to read the drawing. Tap to close.

A hand-drawn bar chart of twenty hypothetical trades around a zero line, with no amounts: most bars are short red losses (many small losses, cut short), a few are tiny green gains, and two tall green bars stand out (a few big winners, left to run). A grey note reads one bar = one trade.

Losing streaks and long drawdowns

Losses cluster. In a sideways market, breakout after breakout fails, and a trend follower can go months without a new account high: the long, shaded stretch in the drawing below. That is normal for the method, not proof that it has stopped working, and it is when most traders quit, often just before the next trend.

Plan for it before it happens: decide how deep a drawdown you will accept, trade a size that keeps it survivable, and judge the system on a long record, not on the last ten trades. The Turtle trading lesson shows how the Turtles cut their size during drawdowns, and trading psychology covers the urge to abandon a plan.

Turn your phone sideways to read the drawing. Tap to close.

A hand-drawn account-value curve with no numbers: it rises to a new high (circled), then zigzags lower and sideways for a long time under a dashed line drawn at that peak, with the gap shaded red (drawdown: months with no new high), before climbing through the line to a new high again.

Position sizing: risk a little on every trend

Because losing streaks are built in, the risk per trade is usually small. Size from the stop: shares = dollar risk ÷ distance to the stop.

Hypothetical: a $50,000 account risks 0.5% per trade, or $250. The stock’s ATR is $2.00 and the stop sits 3 × ATR = $6.00 below the entry, so the position is $250 ÷ $6.00 = 41 shares. A volatile stock gets fewer shares and a calm one more, but each stop-out costs about the same. At that risk, ten losses in a row cost about 4.9% of the account. The position size calculator does the math, and risk management covers the limits.

Spreading trades across markets that trend at different times, such as stock, bond and gold ETFs, helps more than ten positions in one sector, which behave like a single bet.

When trend following fails

Trend following struggles most in markets that go nowhere, and the failures stay part of the plan only if the size is small. The usual problems:

  • Chop: repeated false breakouts, each a small loss, can last for months.
  • Sharp reversals: a trend that ends in a sudden V gives back much of the open profit before the trailing stop is hit.
  • Gaps: overnight news can open price far beyond the stop, so a loss can exceed 1R.
  • Tinkering: changing the rules after every losing streak, or fitting them perfectly to past charts, leaves you with an untested system.
  • Cutting winners: taking profits early removes the large gains the method depends on.