Trend Following Strategy: Rules, Examples and Risk

Trend following is a trading approach that buys markets that are already rising and sells markets that are already falling, then holds the position until the trend shows clear signs of ending. It does not try to predict tops or bottoms. It waits for price to prove a direction, joins late, and exits late, accepting many small losses in exchange for a few large winners.

Quick answer: A trend following strategy enters in the direction of an established move, for example when a fast moving average crosses above a slow one or when price breaks a 20-day high, and exits on a trailing stop. Win rates are often below 50%, so profits depend on letting winners run and keeping each loss small and fixed.

---

Trend following key facts

What is trend following?

Trend following is a rule-based style that treats recent price direction as the main input: long if a market has risen over a chosen lookback, short or flat if it has fallen. The idea is old. AQR's research on trend following quotes the economist David Ricardo's advice to cut losses short and let profits run.

The academic name for the simplest version is time series momentum. It is one of the core trading strategies and a close relative of momentum trading, though trend following tends to hold longer and relies on mechanical exits.

Does trend following work?

Trend following has a long research record, but it does not profit in every period, and individual results depend on costs, discipline and the markets traded.

In "A Century of Evidence on Trend-Following Investing", Hurst, Ooi and Pedersen of AQR built a time series momentum strategy back to 1880 and found it was consistently profitable over the following 110 years. Earlier work by Moskowitz, Ooi and Pedersen (2012) found that time series momentum had been profitable on average since 1985 for nearly all equity index futures, bond futures, commodity futures and currency forwards they studied.

Three caveats matter for a retail or prop trader:

1. Those results come from diversified portfolios of many futures markets. One trader running one or two instruments sees far lumpier results.
2. Drawdowns can be long, especially in range-bound years.
3. Past performance does not guarantee future returns.

How do you identify a trend?

You identify a trend with a simple, objective rule rather than by eye. Three common definitions:

• Price versus a moving average. Price above a rising 50-day or 200-day average is an uptrend. See moving averages for the SMA versus EMA choice.
• Swing structure. Higher highs and higher lows define an uptrend; lower highs and lower lows define a downtrend. The break of structure guide covers how traders mark these swings.
• Channel position. A close above the highest high of the last N candles signals a new uptrend; a close below the lowest low signals a downtrend.

What are the main trend following strategies?

The main trend following strategies are moving average crossovers, Donchian channel breakouts and pullback entries within an established trend. Each has the same skeleton: a trend filter, an entry trigger, an initial stop and a trailing exit.

Moving average crossover

A crossover system buys when a faster average crosses above a slower one and sells or reverses when it crosses back below. A well-known pair is the 50-day and 200-day simple moving average, whose bullish cross is often called a golden cross. Faster pairs, such as a 20 and 50 period exponential average, react sooner but produce more false signals in sideways markets.

The weakness is lag: by the time a 50-day average crosses a 200-day average, much of the move may be gone. Trend followers accept this because the aim is the long middle of a big trend.

Donchian channel breakout (turtle style)

A Donchian channel plots the highest high and lowest low over a lookback period. The breakout version buys when price closes above the upper band and exits when price closes below a shorter lower band. This is the logic popularised by the turtle traders, a group trained by Richard Dennis in the 1980s. Investopedia's account of the experiment describes buying new four-week highs as an entry signal, using ATR to vary position size, and never risking more than 2% of the account on one trade.

In this illustrative chart, the entry comes well after the range low and the exit well after the top. A breakout system gives up both ends of every move to hold the middle.

Pullback entries in a trend

A pullback system waits for an established trend and buys a dip toward a moving average, a prior breakout level or a Fibonacci retracement zone. Entries are closer to the stop, so the reward-to-risk ratio can be better, but some strong trends never pull back far enough to trigger an entry.

How do you set stops and exits in trend following?

You set the initial stop at a volatility-based distance and then trail it, so that the exit is decided by price action rather than by a profit target. Common choices:

• ATR stop. Place the stop two or three times the Average True Range away from entry. ATR measures how much a market typically moves, so the stop adapts to quiet and volatile conditions.
• Channel exit. Exit on a close below the lowest low of the last 10 or 20 candles (the reverse for shorts).
• Opposite signal. Exit when the fast average crosses back below the slow one.

Most trend followers avoid fixed profit targets, because the occasional outsized winner pays for all the small losses. See the stop-loss guide for stop mechanics.

Why do trend followers lose more often than they win?

Trend followers lose more often than they win because most breakouts and crossovers fail; markets spend a lot of time moving sideways. The strategy survives because the average winner is several times the average loss.

Here is the arithmetic with hypothetical numbers. Suppose a system wins 35% of the time, the average winner is 3R and the average loss is 1R (R is the amount risked per trade):

• Expectancy = (0.35 × 3R) minus (0.65 × 1R) = 1.05R minus 0.65R = plus 0.40R per trade

That looks attractive, but the same system will produce long losing runs. With a 65% chance of losing each trade, a common approximation puts the longest expected losing streak in 100 trades at about eight in a row. If you risk 2% per trade, eight losses cost roughly 15% of the account. If you risk 0.5%, they cost about 4%. The risk-reward ratio guide shows how win rate and reward-to-risk combine to break even.

How do you size positions in a trend following strategy?

You size each position so that hitting the stop costs a fixed fraction of the account, which means volatile markets get smaller positions and quiet markets get larger ones. The formula is:

Position size = (account × risk %) / (stop distance per unit)

The turtle traders used a version built on ATR, so every position had a similar impact on the account and no single market dominated the risk. The position size calculator does the arithmetic for you, and the risk management in trading guide covers how to cap total open risk across several positions.

Worked example: trend following gold on a 25,000 USD account

This hypothetical example uses a 25,000 USD CLASSIC 2-Step evaluation at Velotrade, a multi-asset prop trading firm offering simulated evaluations. Prices are illustrative.

Account limits

• Static maximum drawdown: 10%, so the account must stay above 22,500 USD for its whole life. Total loss budget: 2,500 USD.
• Daily loss limit: 5%, set at 00:30 UTC from the higher of balance or equity. On day one that is 1,250 USD.

The trade

1. Gold (XAU) closes above its 20-day high at an illustrative 3,000 USD per ounce.
2. The 20-day ATR is 45 USD, so a 2 ATR stop is 90 USD below entry, at 2,910 USD.
3. Risk per trade is 0.5% of 25,000 USD = 125 USD.
4. Position size = 125 / 90 = 1.39 ounces, rounded down to 1.3 ounces. Actual risk: 1.3 × 90 = 117 USD.
5. Notional value is 1.3 × 3,000 = 3,900 USD, far inside the 6x challenge leverage available on gold. Commission on commodities is 0.01% per side, about 0.39 USD each way here.

Two outcomes

• The trend runs. Gold climbs and the 2 ATR trailing stop is finally hit at 3,270 USD. Profit = 270 × 1.3 = 351 USD, or 3R.
• It fails. Price falls back into the range and hits the stop. Loss = 117 USD, or 1R.

The losing streak test. Ten consecutive losses at 117 USD cost 1,170 USD. The account would sit near 23,830 USD, still 1,330 USD above the static floor. At 2% risk (about 470 USD per trade), roughly five losses would use the whole budget. That difference is why trend followers in an evaluation usually risk 0.25% to 0.5% per trade.

How does trend following fit prop firm rules?

Trend following fits prop firm rules well when position risk is small, because the main threats are losing streaks and giving back open profit, not single blow-ups.

Points that matter specifically for trend followers:

• Floating profit counts. Because the daily limit is set from the higher of balance or equity at 00:30 UTC, a large open gain raises the reference point. If equity is 26,600 USD at the reset on a 25,000 USD CLASSIC 2-Step account, the daily floor is 26,600 × 95% = 25,270 USD. A sharp reversal that gives back 1,330 USD of open profit in one day breaches the limit, even though the account is still in profit overall. Tighten trailing stops or reduce size when open profit is large relative to the daily limit.
• The drawdown floor never moves. Profits build a cushion above a fixed floor. See static maximum drawdown explained.
• Time. There is no maximum time limit, which suits a slow method; a daily-chart system may take months to reach a 10% target.
• Holding. Positions can be held overnight and over weekends, except single stocks, which carry earnings, dividend and holding restrictions.
• PRO 1-Step. With a 3% static drawdown, even a 0.5% risk per trade allows only six full losses. Trend followers on this plan usually cut risk to 0.25% or lower.

Because trend following holds for days or weeks, compare the conditions in best prop firm for swing traders and see the full plan details on the challenges page.

Which markets suit trend following?

Liquid markets capable of sustained moves suit trend following. On Velotrade's simulated accounts that includes crypto perpetuals such as Bitcoin and Ether, forex majors, gold, silver, crude oil and the SPY, QQQ and IWM index ETFs. Velotrade does not offer CME futures, so treat the futures-based research as background, not a forecast for any single instrument.

Common trend following mistakes

• Cutting winners early. Taking profit at 1R or 2R removes the large winners the method depends on.
• Moving the stop away from price. The stop should only ever trail in the direction of the trade.
• Over-optimising the lookback. A 47-day breakout that backtests best is probably curve-fitted; see backtesting.
• Abandoning the system during a drawdown. Losing streaks are built into the method; size so you can sit through them.

Trend following is one of several approaches in the trading strategies library. If you prefer buying strength over shorter horizons, read breakout trading; if you prefer fading stretched moves, compare mean reversion trading. This article is educational and not investment advice. Trading involves risk, and simulated results do not guarantee future performance.

Sources

• AQR, "A Century of Evidence on Trend-Following Investing": Hurst, Ooi and Pedersen, time series momentum back to 1880.
• AQR, "Time Series Momentum": Moskowitz, Ooi and Pedersen, Journal of Financial Economics (2012).
• Investopedia, "Unlocking the Secrets of the Turtle Trading Experiment": turtle breakout rules, ATR sizing and drawdown risk.
• Investopedia, "Understanding Donchian Channels": how the channel is calculated.
• Investopedia, "Average True Range (ATR)": the ATR formula and its use in stops.
• Investopedia, "Understanding Crossovers in Technical Analysis": moving average crossover signals.
• Investopedia, "Trend Trading": definition and common tools.