STOP LOSS? I DON’T NEED NO STINKIN’ STOP LOSS
And, you don’t need a stop loss if you believe in the Efficient Market Hypothesis (EMH) and Random Walk Theory (RWT). First, as a reminder, the EMH states,
“it is impossible to “beat the market” because stock market efficiency causes existing share prices to always incorporate and reflect all relevant information. According to the EMH, stocks always trade at their fair value on stock exchanges, making it impossible for investors to either purchase undervalued stocks or sell stocks for inflated prices. As such, it should be impossible to outperform the overall market through expert stock selection or market timing, and that the only way an investor can possibly obtain higher returns is by purchasing riskier investments.” From Investopedia
And, the RWT states,
“The random walk theory is the theory that stock price changes have the same distribution and are independent of each other, so the past movement or trend of a stock price or market cannot be used to predict its future movement.” From Investopedia
If the above theories are correct, a stop loss is pointless and would actually hurt your returns. So, are the theories correct and are stop losses pointless? If a stop loss is valid, what are the general ground rules regarding them?
Central to this point is whether momentum exists in the stock market. Momentum is when prices continue to rise or fall for extended periods of time. According to the EMH, if an event occurs, the stock price would adjust and again become random. No momentum should exist.
Lo and Mackinlay (1988) reported significant positive serial correlation in weekly returns, suggesting that momentum is present for short time-frames, which counters EMH. At the same time, Fama (developer of EMH) and French (1988) report negative correlation in observation intervals of three to five years, suggesting a mean-reverting processes in the long term.
What do the two 1988 studies imply? That momentum can exist as an anomaly and in the longer term the price reverts back to the mean, making a stop-loss inefficient, in the long the term.
What about the short-term?
Kaminski and Lo asked this exact question in 2007, “When Do Stop-Loss Rules Stop Losses? by “Kathryn M. Kaminski and Andrew W. Lo. They showed that under the RWH, simple stop-loss rules always decrease a strategy’s expected return, but in the presence of momentum, stop-loss rules can add value. This study examined U.S. equities from 1950 to 2004 using monthly data with an event window of 3 – 18 months.
This landmark paper by Lo showed that in the intermediate-term (monthly data points over 3-18 months), stop loss rules do stop losses. But the average trader trades in the here-and-now. Daily, maybe weekly – not over years. In 2008, Robin Erdestam and Olof Stangenberg published, “Efficiency of Stop-Loss Rules: An Empirical Study of the Swedish Stock Market.” This study examined the efficiency of stop-loss in the short term and included the 2008 stock market downturn.
Erdestam showed that the stocks in the OMX Stockholm 30 Index show a strong mean-reverting behavior implying that stop-loss rules are inefficient. However, their study differed from Lo: Erdestam studied 30 months of data with a window of 10 minutes to 2 days; Lo used 54 years of data with a window of 3-18 months. Erdestam probably is seeing mean reversion over short times compared to longer-term momentum for Lo.
Lo repeated his study in 2013 using data from 1993 to 2011 on U.S. stocks and bonds. He again showed a benefit for stop-loss strategies with the best model using a -1.2 to -1.5 standard deviation from the 60-day mean.
A similar study was, “Performance of Stop-Loss Rules vs. Buy-and-hold Strategy,” by Bergsveinn Snorrason and Garib Yusupov in 2009. Once more, when price movements follow random walk or mean-reversion then stop-loss rules are inefficient. But if the price follows a trend, i.e. momentum, then stop loss rules do add value. Snorrason also compared a fixed stop loss to a trailing stop loss. The best performing rule was a trailing stop loss between 15% and 20% loss level limit.
In August, 2015, Han published “Taming Momentum Crashes: A Simple Stop-Loss Strategy.” Using a definition of momentum Han examined if a stop loss improved performance. For this study, momentum was defined by Jegadeesh and Titman (1993) of buying past winners and selling past losers and holding for 3 – 12 months. This strategy produced 1% monthly returns. Han modified the momentum strategy by applying a 10% stop-loss rule. This single rule raised the monthly return from 0.99% to 1.69%. On average, 30% of the stocks do stop out.
Unfortunately, there is not a lot of research into the benefit of using stop-losses. Yet, every trade book and article suggests using a stop-loss, without any instructions on how to use them. Early on, when I was just beginning to study the markets, I ran multiple back-testing experiments on stop-loss size. I found essentially the same thing these researchers did. A tight stop loss always lost money. The looser and farther away the stop loss was, the fewer times I stopped out before hitting the target.
What the articles do show is that a loose trailing stop loss (10%-20%) is beneficial when the stock is showing momentum. If there is no momentum do not use a stop, or a very loose stop, is the best course of action.
- Stop-loss rules are efficient for a stock that has momentum.
- For a stock that follows mean-reversion, stop-loss strategies are not efficient.
- Use a wide stop-loss for determining your position size and your “R.”
- If a stock is “broken” or shows strong negative momentum, exit.
- If the stock is not “broken” and is correlated to the market in general, do not use a stop-loss.
- Use a large stop loss equal to either: 1.5 standard deviation (STD) of 60-day period; or 1.5 STD of volatility 60-days out; or a simple 15% to 20% decline from your purchase price.
- 1.2 STD = 77% of the moves
- 1.5 STD = 86.5% of the moves
- 2 STD = 95% of the moves
- Stop-losses are mental, never enter a stop loss into your trade platform. When you want to exit, enter a “hidden” stop-limit order at the price.