TradingView Black Friday: up to 80% off annual plans, 23 November to 1 December. See the dates and the best plan

Kagels Trading
Topics

Trend Following Strategies: Rules, Tools and a 76-Year Test

Contents
  1. Trend following in 30 seconds
  2. What is trend following?
  3. Trend following models: rules instead of judgment
  4. Why do trends persist?
  5. How do you recognize a trend?
  6. Trend following and its alternatives
  7. The main tools of trend followers
  8. A simple example: the 200-day rule
  9. Which timeframe works best?
  10. The seven most common trend following mistakes
  11. Which markets suit trend following?
  12. Pros and cons of trend following
  13. Conclusion: trend following works, but not the way most people hope
  14. Frequently asked questions about trend following
  15. About the author

Trend following means you buy what is already rising and sell what is already falling, and you stay in until the trend breaks. You do not try to call the top or the bottom. You accept that you will miss the start and the end of every move and earn your money in the middle. That simple idea has been used by traders for more than a century and by large futures funds for decades.

In this guide you get the rules, the tools and the honest numbers. We explain how to read a trend, which indicators trend followers use and where the method fails. Then we show what 224 trades of a simple 200-day trend filter on the S&P 500 did from 1950 to 2026. This article follows our editorial policy; it is education, not investment advice.

Trend following in 30 seconds

  • Idea: trade in the direction of an existing trend and exit when it breaks, instead of predicting turns.
  • Typical result: many small losses and a few large winners. In our 200-day test, only 43 percent of trades made money.
  • Main enemy: sideways markets, where the rules buy and sell back and forth at a loss.
  • What it buys you: smaller drawdowns. Our filter cut the largest S&P 500 decline from 56.8 to 24.0 percent, but earned 7.2 instead of 8.2 percent a year.
  • What it needs: fixed rules, a stop and the patience to sit through long losing streaks.

What is trend following?

Trend following is a trading method that identifies an existing price move and trades in its direction instead of forecasting turning points. When a stock or an index keeps rising, you buy or stay long. When it keeps falling, you sell short or stay out. The old market saying “the trend is your friend” sums it up.

The idea is old. Jesse Livermore waited for the market to confirm its direction before he bought and added to positions only when they were already in profit. In 1983 Richard Dennis taught a group of beginners a purely rule-based system that bought breakouts above 20-day and 55-day highs. The full rules are in our guide to Turtle Trading, and Ed Seykota is another well-known name of the same school.

Trend following differs from contrarian trading at the root. A contrarian tries to anticipate a reversal and buys weakness. A trend follower waits until the market has shown its direction and buys strength. Neither is right in every market phase.

Trend following models: rules instead of judgment

A trend following model is a fixed set of rules that derives entries, exits and position size from price data alone. It needs no forecast and leaves no room for a decision in the moment. You decide in advance and then follow the rules.

That is what makes a model testable. Anything written as a fixed rule can be checked against decades of historical prices. A gut decision cannot. Our guide to backtesting shows how to run such a test without fooling yourself, including the dividend trap that makes trend filters look better than they were.

Research finds that past returns of a market tend to continue for a while. Moskowitz, Ooi and Pedersen studied 58 liquid futures on equity indexes, currencies, commodities and bonds in their paper Time Series Momentum (Journal of Financial Economics, 2012). They found that a market’s own past return tended to persist for about a year and partly reversed over longer periods.

This is time series momentum, and it is what trend followers trade. A related effect, cross-sectional momentum, compares assets with each other: past winners tended to beat past losers. Asness, Moskowitz and Pedersen documented it across eight markets and asset classes in “Value and Momentum Everywhere” (Journal of Finance, 2013). Both are historical findings, not guarantees.

Behavior and market structure offer explanations. Investors react slowly to new information, then herd into moves they missed. Large funds spread big orders over days or weeks, and stop-loss orders can push a move further once they trigger. None of this tells you how long the next trend will last.

How do you recognize a trend?

There is no single definition of a trend, so pick one method and use it consistently. The most common definition comes from market structure: a series of highs and lows. What looks like an uptrend on a daily chart can be a downtrend on an hourly chart, so the timeframe is part of the definition.

  • Uptrend: higher highs and higher lows. A rising trendline under the lows holds, and moving averages slope up.
  • Downtrend: lower highs and lower lows. A falling line over the highs holds, and moving averages slope down.
  • Sideways market: price moves between support and resistance, and moving averages run flat and cross each other.

The sideways market is the trend follower’s trap. Every small push looks like a new trend, the rules buy, and price falls back into the range. Our guide to support and resistance explains how to mark such zones, and our range trading guide covers the strategies that work there. How to read trends from highs and lows alone is in our guide to price action trading.

Trend following and its alternatives

Each method fits a different market phase. Trend following needs directional markets. The table shows where the main alternatives work and where they fail.

Method Works best in Fails in
Trend following Long directional moves Sideways markets
Breakout trading The start of new trends after a range False breakouts
Mean reversion and range trading Stable sideways markets Strong trends
Contrarian trading Major turning points Trends that keep running

Breakout trading is a close relative. It enters when price leaves a range, which is how the Turtles caught the start of many trends. Our breakout trading guide shows how often breakouts fell back into the range in our own count, and why a stop is part of every breakout trade.

The main tools of trend followers

Most trend tools answer one of three questions: what is the direction, how strong is it, and where do I get out. Use one or two of them, not all at once; more tools mean more conflicting signals. Each tool has its own guide on our site.

  • Trendlines: the classic way to see direction and the point where a trend breaks. See trendline trading.
  • Moving averages: a smoothed line of past closes; price above a rising average is a common long-term filter. See moving averages and the golden cross.
  • MACD: shows momentum as the gap between two moving averages. See the MACD indicator.
  • ADX: measures trend strength without direction. See the ADX indicator.
  • Donchian channel: the highest high and lowest low of a period, the breakout tool of the Turtles. See the Donchian channel.
  • SuperTrend: an ATR-based line that flips with the trend and works as a trailing stop. See the SuperTrend indicator.
  • Bollinger Bands: show volatility; a squeeze can come before a new trend. See Bollinger Bands.

A simple example: the 200-day rule

The best-known trend rule fits into one line: hold the S&P 500 while it closes above its 200-day average, and hold cash while it does not. The 200-day moving average is one of the most widely watched lines on index charts. It is simple, needs no judgment and can be tested over many decades.

We tested this rule on the S&P 500 from 1950 to October 15, 2026. The rule: long while the close is above its 200-day simple moving average, trade at the next close, 0.05 percent cost per side, cash earns nothing, price index without dividends. The full test is in our guide to moving averages; here are the main numbers.

S&P 500, 1950 to 2026 200-day filter Buy and hold
Return per year 7.2% 8.2%
Largest decline 24.0% 56.8%
Time in the market 71.6% 100%

Source: own calculation, ^GSPC daily closes from Yahoo Finance, 224 round trips. Without dividends; with dividends, buy and hold gains more, because it never misses one.

Only 43 percent of the trades made money, and that is normal for trend following. Of 224 closed trades, 97 were winners with an average gain of 8.4 percent and 127 were losers with an average loss of 1.5 percent. The best trade ran from November 1953 to May 1956 and gained 80.7 percent. The ten best trades alone made up about two thirds of the combined result of all closed trades, measured in log returns.

Chart of 224 trades of a 200-day trend filter on the S&P 500 from 1950 to 2026. 97 winning trades, 43 percent, and 127 losing trades, 57 percent. The average winner made 8.4 percent, median 1.7 percent, best 80.7 percent from 1953 to 1956. The average loser lost 1.5 percent, median 1.3 percent, worst 6.2 percent in 2015Click to enlarge
Most trades lost a little, a few won a lot. S&P 500 price index, October 1950 to October 15, 2026: long while the close is above its 200-day simple moving average, trades at the next close, 0.05 percent cost per side, no dividends. Own calculation.

Source: Kagels Trading, own calculation (model from our moving averages test).

The filter did not beat buy and hold; it reduced the damage. It earned about one point a year less, mostly because it sat out part of the market and paid for false signals: 58 percent of all trades lasted less than ten trading days. In exchange, it avoided most of the deepest bear markets. Our backtesting guide shows the same pattern on SPY with dividends, where the gap after 2012 grew to almost four points a year.

When does the rule fail? In choppy markets around the line, it sells low and buys back higher again and again. In long bull markets without deep corrections, it simply costs return. If you cannot live with long stretches of small losses while buy and hold looks better, you will abandon the rule right before it pays off.

Which timeframe works best?

There is no single best timeframe, and the number of false signals depends on your rules and the market. On weekly and daily charts, a trend usually spans weeks to months, and each signal is based on more price data. On hourly charts, trends appear every day; many traders find more whipsaws there, and every extra trade adds costs. Test your own rule on the timeframe you plan to trade.

Match the timeframe to your style. Swing trading usually works with daily and four-hour charts and holds positions for days to weeks. Day traders use hourly charts and below. Many traders check the higher timeframe first and trade only in its direction; our guide to multi-timeframe analysis explains how.

Stage analysis is a classic weekly trend method. Stan Weinstein divides a stock’s life into four stages and buys only in stage two, the uptrend. The rules are in our guide to stage analysis, and Mark Minervini built a stricter trend template on top of it.

The seven most common trend following mistakes

  1. Entering too early: you buy the first green candles after a fall and call it the bottom. Wait until the structure shows a trend; missing part of a move costs less than buying into a falling market.
  2. Stops that are too tight: normal noise takes you out before the trend runs. Place the stop beyond a swing low or use a volatility stop based on the ATR; see our volatility stop loss guide.
  3. Trading sideways markets: flat moving averages and a low ADX show that there is no trend to follow. Stand aside until price leaves the range.
  4. Ignoring risk management: no stop and positions that are too large turn one bad trade into months of recovery. Our guide to position sizing shows how to cap the risk per trade.
  5. Emotions instead of rules: cutting winners early and letting losers run is the exact opposite of trend following. Write the exit rules down before you enter.
  6. Ignoring major news: central bank decisions and earnings can end a trend in minutes. Know the calendar before you hold a position through an event.
  7. Too many indicators: five tools that disagree lead to paralysis. One direction filter, one entry rule and one exit are enough.

Which markets suit trend following?

Trend following can be applied to liquid markets, but the character of the trends differs. Whether it pays depends on the rules, the costs and the period. Futures funds trade many markets at once because diversification across stocks, bonds, currencies and commodities raises the chance that at least some of them trend at any time.

  • Stock indexes and stocks: long upward drift over decades, with sharp bear markets that trend filters try to avoid.
  • Commodities: supply and demand cycles can create long trends in both directions, for example in energy and metals.
  • Currencies: trends often follow interest rate differences between countries, but many pairs also spend long periods in ranges.
  • Gold: reacts to inflation, real rates and crises and has had long multi-year trends.
  • Crypto: extreme trends in both directions, but also extreme volatility and news risk; position size matters even more.

Pros and cons of trend following

Pros

  • Clear rules: entries and exits follow fixed signals.
  • Big winners: long trends can pay for many small losses.
  • Both directions: works long and short, and as a filter it can keep you out of bear markets.
  • Testable: fixed rules can be backtested honestly.

Cons

  • Low hit rate: in our test only 43 percent of trades won.
  • Whipsaws: sideways markets produce many false signals.
  • Late entries and exits: you always miss the start and the end of a move.
  • Long losing streaks: they test discipline more than any chart.

Conclusion: trend following works, but not the way most people hope

Trend following does not predict, it reacts, and its edge comes from letting a few big winners pay for many small losses. Our 76-year test of a simple 200-day rule shows both sides: a much smaller largest decline, but lower returns and a 43 percent hit rate.

If you want to use it, start with one market, one timeframe and one rule. Test it with realistic costs and dividends, size every position so a losing streak cannot hurt you, and follow the exits even when it hurts. The method only works for traders who can stay with it through the dull and the losing phases.

Frequently asked questions about trend following

What is a trend following strategy in simple terms?

You trade in the direction of an existing trend and stay in until the trend breaks. You buy when prices are already rising and sell or stay out when they are falling. You do not try to guess turning points.

How does trend following work for beginners?

Four steps: identify a trend, wait for an entry, set a stop and let the trend run. Use market structure or a moving average to find the trend, enter on a breakout or a pullback, risk only a small part of your account per trade and exit when the trend breaks. Our pullback trading guide covers one of the common entries.

Why do most trend following trades lose?

Because the method buys every new move, and most of them fail. In our 200-day test, 57 percent of trades lost, but the average loss was small. A few long trends, such as 1953 to 1956, made up most of the total gain.

Does the 200-day moving average strategy beat buy and hold?

Not in our test of the S&P 500 from 1950 to 2026. The filter earned 7.2 percent a year against 8.2 percent for buy and hold, without dividends. Its benefit was the smaller largest decline: 24.0 percent instead of 56.8 percent.

Which indicators are best for trend following?

There is no best indicator, only tools for different questions. Moving averages and trendlines show direction, the ADX shows strength, and the SuperTrend or an ATR stop shows where to exit. One tool per question is enough; more tools mean more conflicting signals.

How do I tell a real trend from a false signal?

Look for structure, slope and strength together. A real uptrend shows higher highs and higher lows, a rising moving average and often a rising ADX. Flat averages and a low ADX point to a sideways market, where most false signals happen.

This English edition is based on our German edition on kagels-trading.de and has been adapted for international readers.

← All articles