• 
• 
• 

• Performance
• Services
• Company
• Learn About
• My Attain

What <i>is</i> was Wrong with Trend Following?

October 4, 2004

 

This simple question seemed a good topic to write on right up through the first few hours of Thursday, September 30th. Up until then, the trend following universe was in dire straits. Famous CTAs (commodity trading advisors) with billions of dollars under management such as John W. Henry, Niederhoffer, Campbell & Co., and Dunn Capital were down anywhere between 5% to 40% on the year - with many seeing multiple losing months in a row.

As a refresher, CTAs are professional money managers who specialize in generating investor returns through the trading of commodity futures. While there are not exact stats on this - it is widely believed that the majority of CTAs utilize trend following models in one form or another.

The CSFB HEDG Managed Futures Index, meanwhile, was down -6.99% through the end of August, and had seen six straight months of losses, the longest such streak since its inception in 1994.

Not to be forgotten, long term trend following systems haven't been faring much better in 2004. While some trend followers are positive on the year, each of the trend following systems Attain tracks remains well below its equity highs.

All of this red ink begs the question: "What's wrong with Trend Following?" We set out for an answer, but a funny thing happened on the way to finding out what's wrong - things got better.

Take Meyer Capital Management (MCM) as an example. MCM is one of Attain's recommended CTA programs, with an average annual return of 27.4% and a Max DD of 16.4% prior to this year. Since April of this year, the MCM program saw five straight losing months while posting a new max DD of 21.6%, but as we alluded to above - things got better. The program tacked on nearly 10% in the last few days of September to move significantly off that DD level.

So while we may have to change our question to: "What WAS wrong with Trend Following?" - the new Max DD levels and rather long streak of losing months do demand our attention.

What goes Up, Must Come Down:

The theory put forth by John Joseph of the SEMA4 Group CTA and NextDimension Creative Technologies follows the old adage that what goes up must come down. Mr. Joseph had the following to say:

"The reason trend-followers have struggled is quite straightforward, actually. In the fall of 2003 and spring of 2004, there were very strong and extended trends in nearly all major markets. This unusual confluence of trends produced unusually large profits for most trend-following traders and systems. However, starting in April most of these trends reversed abruptly, causing trend-followers to give back much of these windfall profits. Of course, when your results are marked-to-market giving back open profits is indistinguishable from a loss of capital and often results in a painful drawdown.

Since then (at least until September), there have been relatively few trends as the markets have entered a period of indecision, which is not unusual after such strong and sustained movement. What is unusual, however, is the fact that so many unrelated markets have gone through a similar cycle almost simultaneously. Complicating the issue is the fact that most trend-following portfolios tend to be overweighted in the financial markets (bonds, currencies, metals) and grains (esp. the soy complex), which were among the hardest hit. The best trends were in energies and meats, two sectors that are traditionally underweighted in a trend-following portfolio."

- John Joseph (www.nextdsystems.com)

A Change in Environment:

The developer of the Andromeda system, Peter Waite, feels there is nothing wrong with trend following, and remains devoted to it both as a developer and as an investor. His personal money remains invested in Andromeda and Pegasus. But Mr. Waite has heard investors utilizing his systems put forth several "reasons" for trend following's recent woes, including the following:

"On a different note, there is another argument which is perhaps worth mentioning. I have heard some argue these days that the new presence of terrorism and war has permanently changed trend following. "Things will never be the same again". Those folks argue that we are now in a new era permanently marked with fear and a new lethal danger. Terrorism is a danger that we cannot predict when it will happen, only that eventually it will happen, and it will strike when we least expect it. The terrorists always have the advantage. It is perhaps the most important issue on people's minds these days, especially during these election times. Consequently traders are nervous and quick to take profits. Any young trend is quickly "killed" as traders take early profits rather than taking their chances by letting them run. The reason - terrorism which has changed the dynamics of how we trade for ever.

Does this argument have validity? Only time will tell...........................!!!"

- Peter Waite (www.andromedafutures.com)

Unrealistic Expectations:

The feeling prevalent with more than a few CTA managers was that the only problem with trend following in 2004 has been investors' expectations. They argued that systematic approaches such as trend following are bound to experience long periods of stagnant or no growth in equity as well as the ever present possibility of new max drawdowns. One particular manager mentioned that many investors who piled into alternative investments such as trend following CTAs while chasing after anything with positive returns during the post-bubble down years in the stock market, based their investment decisions on track records often as short as 5 years and rarely if ever longer than 20 years.

This manager theorized that without track records and testing spanning as far back as 50 years (which is impossible in many cases due to the fact many commodity markets didn't exist); there is a build up of unrealistic expectations amongst the public. In short, many investors (and managers for that matter) don't know what to expect over a fifty year time period - as we don't have that much trend following history.

His feeling was that there are several possible return distributions which have never been seen. These could include more extended flat/down periods such as we saw this year, or uncharted territory such as triple digit annual returns or new, very large max DDs.

It's the Markets:

For my two cents, I believe the poor trend following environment in 2004 was a result of the markets in which trends developed. Contrary to popular belief, there were several nice trends in 2004, most notably in Crude Oil, Bonds, and Natural Gas. The issue with these trends is that they emerged out of a high volatility environment, as they were reversals of previous trends nearly across the board.

Trend followers usually catch trends emerging from a directionless, low volatility phase - in which a market or sector is range bound for an extended period of time, before "breaking out" to begin a new trend higher or lower. This range bound environment creates a low risk/high reward trade - as the risk of the market falling back into the range is a lot less than the possible reward of the market setting new all time highs or lows.

In contrast, the trends which developed in Crude Oil, Bonds, and Natural Gas in 2004 did not emerge from congestion phases. In bonds, for example, the up trend emerged after bonds had one of the sharpest down moves in over 20 years. In fact, many trend followers had signaled a short trend just days before the up trend began. This resulted in a high risk/high reward trading environment.

The problem with a high risk/high reward environment is that to mitigate that high risk, a trend following system or manager will reduce exposure by doing a smaller number of contracts or by passing on the trade altogether. Should a trend emerge, it has to be that much larger to offset the risk taken, and larger trends take time to develop. So, when the trends did develop in 2004, the exposure to them was relatively small, the benefit therefore small, and there simply hasn't been enough time for them to become large enough trends to register substantial gains.

Have your own thoughts on what was wrong with trend following? We would love to hear them. Email comments to comments@attaincapital.com

- Jeff Malec