Trailing Stop-Losses Effective at Curtailing Downside Risk

A recent study found that trailing stop-loss orders result in higher risk-adjusted returns and also perform particularly well at reducing downside risk.

A recent study found that trailing stop-loss orders result in higher risk-adjusted returns for investors with normal levels of risk aversion. These types of stock orders also perform particularly well at reducing downside risk.

The study looked at 25,997 common stocks for the period of July 1, 1926, through December 30, 2016, from the CRSP database.

Stop-loss rules, as defined by the study, “involve selling a security when its price drops to a pre-determined threshold and buying the security back when its price rises a pre-specified amount.” The study compared a trailing stop-loss with a standard buy-and-hold approach. The trailing stop-loss rule is more flexible than the traditional stop-loss rule in that “the sell trigger price is adjusted upwards if the price moves higher following a purchase … these rules are therefore designed to protect profits.”

There were four thresholds used for the trailing stop-losses: 1%, 5%, 10% and 20%, representing the percentage a stock must fall from the adjusted price to be sold. Concerning all stocks, the study showed that the buy-and-hold approach has a higher mean return than the trailing stop-loss approach.

The study looked at the potential loss level and the expected value in the worst-case scenario for losses. Each of the four trailing stop-loss thresholds indicated a lower downside risk for both risk measures compared to the buy-and-hold approach. The risk-adjusted return was also better for the trailing stop-loss approaches.

The study calculated an equivalent return figure measuring the guaranteed return that someone would accept now rather than betting on an uncertain higher return in the future. The trailing stop-loss approaches had a higher equivalent return than the buy-and-hold approach.

Finally, examining returns over time and for various subperiods, the researchers concluded that trailing stop-loss approaches “are becoming increasingly effective over time, perform better when the overall market is declining and add more value on stocks with higher volatility, and liquidity and lower book-to-market ratios.”

Source: “Risk Reduction Using Trailing Stop-Loss Rules,” by Bochuan Dai, Ben R. Marshall and Nuttawat Visaltanachoti of Massey University and Nhut H. Nguyen of Auckland University of Technology, March 7, 2019 version.

Discussion

ron from dc posted over 7 years ago:

Which level of risk was found to be most advantages in the study: 1,5,10 or 20%. Thanks


Rowan DeBold from MN posted over 7 years ago:

How can I get the full article.


DBK from NY posted over 7 years ago:

To get the source article, just click on the title, which is a link. That brings up the article as a pdf, which you can save to your hard drive.


Jeremy Viles from New York posted over 7 years ago:

I am a huge advocate of using trailing stops, especially on already profitable positions. However, I am not aware of [any] equities trading firm that offers them on their platform. The exception would be Tradestation, who I do not hold an account with. Even so, I believe industry regulation prevent these types of order from being held on the exchanges servers, as well as on your brokers servers. Which would require you to keep your own computer running, or you hosting your trading platform on the cloud. The other option would be having an API written if your firm allows customer sot run API's. But again you are going to have to keep your platform open in order for your stops to move?. I do know one firm which is Tradier that seems to specialize in investors/traders who want to bring their own platform. This whole conversation brings me to my next point which is. Why is there not an industry standard platform at this point. SO that You can customize your platform, and regardless of who you have your account with your software doesn't change. I know that this [is] the case with Forex trading- with the standard platform being Metatrader.You can use Metatrader and any forex broker. There is one trading platform I know of that comes close which is Medved, this platform is standalone, and you can route your trades to any broker you have an account with. The only caveat being they don't connect with [all] firms.


Bob from Argentina posted over 7 years ago:

I have TD Ameritrade and have all my stocks on a 15% trailing stop loss.


Dave from FL posted over 7 years ago:

The article is incomplete - data? - and seems to contradict itself. "Concerning all stocks, the study showed that the buy-and-hold approach has a higher mean return than the trailing stop-loss approach." I sometimes use it at Fidelity in my IRA's. It's easy to set up. If I actually want to sell the stock in the near future, and I need cash soon, but want to enjoy the ride and some high dividends, I set the level at 5%. If I want to hold the stock "within reason" but still be protected from disaster, I set the level to 12% - 15% to avoid losses that would occur with a (20%) bear market - also preempting a 2007-style 50% collapse while I was not watching. I have wrestled with the level to set over the years and would be interested in what academics have discovered. Perhaps a customized formula based on the standard deviation of each stock/sector?


Ronaldo from MD posted over 7 years ago:

I look forward to reading the source article but here are some first impression comments. For traders, a stop loss is essential for good money management. But that exit point should be based upon whatever trading methodology being used which means that repeatable fixed levels as described in the article probably would not be applicable. For investors, a trailing stop loss level can you out of situations that are slowly moving against the reason(s) you bought the stock in the first place. For example, you bought Stock X because improving fundamentals are being recognized by increasing prices starting from a depressed price level. You had spent time identifying the company so you place a trailing stop loss to move up as the stock price rise. The stop loss should account for the normal price movement noise or volatility to avoid whipsaw actions. In any case, setting an objective exit point help avoid emotion-driven mistakes.


John Bradin from CO posted over 7 years ago:

Where's the answer to Dave from Florida's question about buy and hold returns were better than trailing stop-loss? This made no sense to me given the remainder of the article. Does anybody ever proof this stuff?


Dave from FL posted over 7 years ago:

I attempted to read the full article by clicking on the link at the bottom of the summary. I waded through the introduction which is a Literature Review interspersed with what the authors are attempting to present, devoid of useful new information. It's a heavy read, poor grammar and not the way I was taught to write scientific papers. There was no clue as to what the optimum trigger percentage should be. When I got to page 6 it says that sections 2, 3 and 4 will follow - but there is just a blank space until the reference section appears! Outside of this article my own thoughts are that, if I want to implement trailing stop loss protection cost effectively, then I should have far fewer individual stocks and more ETFs so that I don't end up with huge selling costs when the bears bite! It would be of great interest to see a well written article on this topic that is up to AAII's (usual) very high standard! It should include how to differentiate such protection on ETFs (low volatility, requiring higher triggers) versus individual stocks (higher volatility requiring larger triggers). I think this would lead to "advantage ETF" since losses would be smaller and net commissions cheaper. If you had a presentation on this at the AAII conference I might actually turn up!


Dave from FL posted over 7 years ago:

Sorry, I meant ETFs would generally require smaller triggers (at least compared to their constituent stocks) for example a 5% rather than 15% trailing stop loss percentage.


Bob from CA posted over 7 years ago:

It might seem more complete to include stocks back to 1926 and some 25K issues, but there have been significant changes in the market characteristics over the years and there's a lot of junk issues that are half-dead and thinly traded. I use only the past 20 years as it includes a major selloff and has more current characteristics. I also use only ~2K issues of co's with >$200M in sales and P > $5. I find that for a sell, like Dave, I will issue a trailing order at 3~5%, to ride any upside, unless I'm bailing out of a tailspin. For highly liquid shares to hold, 7~12% works pretty well. Thin issues will get you whipped out at times, but I try to avoid them anyway.


You need to log in as a registered AAII user before commenting.
Create an account

Log In

Get your free copy of our special report analyzing the tech stocks most likely to outperform the market.

Download the FREE Report Here: