When Mean Reversion Goes Horribly Awry

After reading an article which described a recent hedge fund blowup I began to ponder the market volatility of the last two weeks and its effect upon mean reversion trading strategies. The mean reversion trade is an important tool in most traders’ toolboxes. Some traders focus on mean reversion almost exclusively while others only implement the mean reversion trade when other trading styles aren’t working very well. Before I delve more deeply into the pitfalls of mean reversion in the realm of financial markets, let’s first discuss the concept of mean reversion and how it can be applied to markets with a couple of examples:

“In statistics, regression toward the mean is the phenomenon that if a variable is extreme on its first measurement, it will tend to be closer to the average on a second measurement, and—a fact that may superficially seem paradoxical—if it is extreme on a second measurement, will tend to have been closer to the average on the first measurement.” Wikipedia

“Mean reversion is a mathematical concept sometimes used for stock investing, but it can be applied to other assets. In general terms, the essence of the concept is the assumption that both a stock’s high and low prices are temporary and that a stock’s price will tend to move to the average price over time.” Wikipedia

In technical analysis there are all sorts of methods which help to identify when a particular market or stock has moved abnormally far away from its “mean”, which is usually represented by a moving average. The most common tools which technicians use to identify a “stretched market” are Bollinger bands, Keltner channels, and standard deviation. All of these indicators are variations of standard deviation of price action over the time period in which the analyst is interested. Many technicians also use the technique of monitoring how far (%) a given market/stock is above/below a key moving average such as the 50-day or 200-day simple moving averages.

Mean reversion techniques work most of the time when implemented correctly, however, when risk is not properly managed mean reversion trades can fail spectacularly. Let’s take a look at some examples of successful mean reversion trading before we review a couple of spectacular disasters:

 

Activision- $ATVI has been a mean reversion/channel trader’s delight for the past two years as it has continued to oscillate in a range between 10.50-12.50. Notice how the 200-day SMA has served as an excellent center line to help guide mean reversion trades. Each time ATVI traded .80 (7-8%) or more away from its 200-day SMA proved to be an excellent time to enter a mean reversion trade. Many times the trader would have had to “withstand some pain” but never enough pain to warrant stopping out of the trade at a loss. The win rate from appropriately executed mean reversion trades in ATVI over the last two years would have been 100%. Needless to say, hindsight is 20/20!

 

Mega-cap Microsoft ($MSFT) was a bit trickier for mean reversion traders over the past year. If a trader had the view that MSFT would remain primarily range bound between 25 & 27, and therefore bought each dip to 25 and shorted each rally to 27 he would have found himself in considerable discomfort on at least two occasions which I have circled. Clearly, buying and selling MSFT at pre-defined levels isn’t a very sophisticated or profitable strategy, therefore let’s use Bollinger bands instead:

 

There we go, now we have something which works much better. If the trader faded each move outside of the Bollinger band (20 period with 2 standard deviations) he would have rarely been in the red and shortly thereafter would have found himself nicely in the green and well on the way to benefiting from a larger price reversion to the mean.

Once again, hindsight is 20/20 and any technical analyst can pull up charts and curve fit (backtest) some indicator studies to demonstrate a very high historical success rate for a particular strategy. However, real time trading is never as simple as implementing one or two backtested strategies and finding your money tree which will shower you with Benjamins each and every day that markets are open. More realistically the trader will find several strategies that work over extended periods of time which then spectacularly fail over very short time spans. The difference between remaining a long term winner in markets and finding oneself in the overflowing graveyard of market blowups is one powerful concept: Risk management. Risk management is a much talked about/pontificated upon concept in the trading/investing community. However, there are very few market participants who genuinely practice what they preach. As an experienced market participant with 8+ years of rigorous experience across markets, I still find myself skirting the edge of optimal risk management from time to time.

The fact is that if you are entering into a mean reversion trade with a limited amount of upside then you must have a clearly defined risk which is expressed through the implementation of a predetermined stop-loss for the trade. Those who enter into mean reversion trades without stop losses are inviting themselves to BLOW UP as many have done recently :

 

The above chart ($6S_F)  reminds me of silver’s ($SI_F, $SLV) parabolic ascent which left many battered & maimed mean reversionists/short sellers in its wake:

 

The Swiss franc (going long $EURCHF, $USDCHF) has been a disastrous currency market in which to implement a mean reversion strategy during the past year. Currency markets are well known to exhibit a tendency to remain in powerful trends over weeks & months before reverting to a mean level (purchasing power parity etc.) over longer time frames (years or decades). While the trader is waiting for long term mean reversion to take hold he will usually find himself insolvent in the short term, similar to what has befallen Dighton Capital. I believe that the Dighton Swiss franc story serves as a powerful lesson to traders who believe in continually cost averaging into a position in anticipation of an eventual reversion to the mean. Remember that the “mean” is a moving target which is often powerfully affected by secular trends in macro/micro economics.

Many have been waiting for $BAC to “revert to the mean” since it fell below $13 earlier this year. Now it looks like a bounce back to $9 would be a gift for trapped longs:

 

Even if all of your trade entries offer a high expected value of profit, if you allow losers to bleed into massive hemorrhaging money gushers you will be doomed to failure as a market participant. Always manage risk and never allow a single trade to lead to your demise.

 

The information in this blog post represents my own opinions and does not contain a recommendation for any particular security or investment. I or my affiliates may hold positions or other interests in securities mentioned in the Blog, please see my Disclaimer page for my full disclaimer.

You might be interested in:
blog comments powered by Disqus