Does Trend Following Work?
Updated: Nov 3
Does Trend Following Work? offers a useful case study in turning market observations into repeatable trading rules. This updated guide preserves the original Financial Wisdom analysis while separating practical lessons from exceptional results and hindsight.
Trend Following: In This Guide
Key lessons and practical rules
The original Financial Wisdom guide and media
Evidence, limitations and risk controls
Frequently asked questions
Related guides and authoritative sources
Key Takeaways
Translate the central idea into objective entry and exit rules.
Test the method across different market regimes and include costs.
Use price, volume and market context without treating any signal as certain.
Define invalidation and position size before entering.
Original Financial Wisdom Guide
Background and Core Ideas
A 22 year research study provides the answer..
In this video we look at a research study which attempts to answer the question; Does trend following work?
The study was completed by a company called RDB Computing, they offer a high-level back testing service used by hedge funds, charging over six figures in return. Needless to say, the study should be rather extensive, let’s take a look.
The definition of the term Trend Following is in essence a trend following trading style, attempting to capture gains through the analysis of an asset’s momentum in a particular direction, up or down. The study in this video focused on a long only approach, covering over 24,000 stocks over a 22-year period from 1983. The stocks chosen came from the major US markets.
In our previous study, we looked at percentage stop losses to determine which percentage offered the best results when compared to a buy and hold approach. Many viewers commented below that video requesting a similar study, but based on an Average True Range stop loss approach, and that is exactly how this trend following study was performed.
The exit rule in the study used the average true range trailing stop, commonly known as the ATR, which is a technical indicator found on most charting or brokerage platforms. It’s a measure of volatility for an individual security, or for this study, a stock’s price movement. Let’s look at an example of its application from within the study.
The red line on this Nortel Networks chart represents the 10-day ATR, and this trailed price after each trading day. We can see that as price increased, so did the ATR, responding to the average price movements during its journey. The price eventually started to trend down, and the trade was therefore closed at the open, the day after the ATR was breached. The same exit process was followed throughout the study on every stock.
The entry rule was simple, buy a stock as soon as it makes an all-time high.
Another example taken from the study here, provides numerous entry points, including the initial entry at an all time high, followed by other entry points thereafter, each making all-time highs.
Practical Checkpoint
Define the setup, entry and invalidation level before acting.
Compare the potential payoff with risk-reward planning.
Set the trade size using consistent position-sizing rules.
Rules and Practical Application
Some measures were considered regarding the integrity of the data within the study, for example, the database of more than 24,000 stocks considered the survivorship bias, meaning that all stocks that were delisted from the exchanges at any point during the 22 years, were included. In fact, more than half of the database comprised of delisted stocks. Dividends and splits were included, as was a minimum liquidity filter, penny stocks were excluded, a minimum $15 price was used, as was a minimum daily traded volume of $500,000. A trading cost for each completed trade amounted to 0.5% to account for commissions and slippage.
This chart represents the number of stocks which would have passed the filters for any given year during the study.
The results we are about to see are summarised in the study abstract, by stating:
“The empirical results strongly suggest that trend following on stocks does offer a positive mathematical expectancy, an essential building block of an effective investing or trading system”.
Let’s break down the results, and of course if you do like this type of content, please let me know below by hitting the like button.
First, we have the expectancy results.
There were over 18,000 trades placed over the 22-year period, and the expectancy distribution we see here shows the number of trades for each net return. For example, of the 18,000 trades, 175 of them returned a net 100% profit, or 4 of the trades, returned a negative 90% net return.
The clear observation to be made is the proportionate skew to the right of the chart. For example, this area shows all the trades which returned 50% or more, equating to 17% of all the trades, and to the left, this area shows all the trades which had a loss of 50% or more, equating to less than 3% of all trades.
At first glance a near 50% win rate may not seem that impressive, but it is in fact rather good for a trend following system. Remember, although the strategy only had a winning trade half of the time, when it had a winning trade, it provided an average 51% profit, and when a trade lost the other 50% of the time, it only lost 20%. This provided a win to loss ratio of 2.56 to 1, which is a solid foundation to work from and offers a positive mathematical expectancy, which is a bare minimum for any system. In my last video we looked at the importance of positive expectancy, using the example of a casino, which you might find helpful.
Evidence Check
Is the sample broad enough to include difficult markets?
Do the results include spreads, fees, slippage and failed signals?
Could hindsight, survivorship or selective examples explain the result?
Evidence and Examples
The study also compared other ATR stops ranging from 8 to 12, but there were no significant differences in the results, although a higher ATR had a higher win rate and lower risk reward ratio, whilst a lower ATR had the opposite.
This next chart however, is a far better gauge on performance in my opinion. The distribution normalises each trade by its own risk. If we take a typical trading scenario as an example, price makes an all time high to a price of $10, the 10-day average true range line trails, and is perhaps 20% below the point at which we make a trade, we have therefore established that our risk on this trade will be 20%. If price were to continue to rise by perhaps 40% to $14, we would have returned 200% on risk, calculated by a 40% rise divided by a 20% risk.
If we revert back to the chart, we can see that 803 trades provided a 200% return on risk, and so on. Normalising each trade by its own risk, reduces the possibility that highly volatile stocks, unjustifiably dominate the results with excessive risk.
The skew of data from a zero percent return upwards, represents trend following in its purest form, allowing winners to run to get the maximum potential return, against a controlled trailing amount of risk.
The number of trades taken due to stocks hitting all-time highs can been seen here, for each of the 22 years. Although the numbers could be seen as excessive, remember this is just raw data and serves as a foundation to build on. You could apply additional filters like fundamental analysis, chart patterns and moving averages to get a more manageable number of trades. In fact, the hedge fund that pursued this study devised their own approach built upon the data, providing this hypothetical equity curve, and the S&P 500 as a benchmark.
The portfolio adhered to the following principles;
Losing trades were never added to. Winning trades were only reduced to alleviate risk. New entries were never skipped. Stop losses were always honoured, and open risk was limited to a specific number, although this open risk number was unfortunately not disclosed.
For those interested, a monthly breakdown of the results was presented, with the overall comparison against the S&P 500. The portfolio outperformed the index whilst enduring a far better maximum drawdown.
Overall, the study reinforced the concept from which many of the great traders and systems of the past were built upon, stay within the direction of the trend, control your risk, and hone your craft upon it.
Risk and Limitations
To learn more why not download my strategy or join our forum.
Access Our Group Here - https://www.financialwisdomtv.com/service
My Breakout Scanner - https://bit.ly/3ea6sl8
My Forum - https://www.financialwisdomTV.com/forum
My Strategy Blueprint - https://www.financialwisdomtv.com/plans-pricing
My Brokerage Account (Interactive Brokers) - https://bit.ly/3HVA1nc
Practical Limitations
Historical examples, book claims and indicator results are educational illustrations rather than forecasts. Outcomes depend on the market, timeframe, data, transaction costs, financing, taxes and execution assumptions. Test complete rules across varied conditions and define the maximum acceptable account risk before taking a position.
Put This Into Practice
These ideas become more useful when selection, entry, stop placement, position sizing and trade management form one repeatable process. The Financial Wisdom Trading Blueprint connects those decisions in a practical weekly framework.
Frequently Asked Questions
Does this strategy guarantee profitable trades?
No. It provides a framework with uncertain outcomes and must be tested with complete risk and execution rules.
What should be defined before an entry?
Specify the setup, trigger, invalidation level, position size and conditions for managing or exiting the trade.
Further Reading
References and Further Reading
Educational information only. Trading and investing involve risk, including the possible loss of capital. Examples are illustrative, not guarantees of future performance, and nothing in this article is personalised financial advice.
About the Author
Written by Gareth, MBA
Gareth, founder of Financial Wisdom, has more than 30 years of personal trading experience. He competes in the United States Investing Championship, creates award-winning trading content and has been featured in Technical Analysis of Stocks & Commodities.
Comments