While the universe of commodity trading advisors had its best performance in recent years in 2007, many of the stalwart long-term trend followers continued to struggle. Campbell & Co., John Henry and Dunn all experienced double-digit losses on some programs, but the larger CTA universe experienced strong returns as the Barclay CTA index returned 7.57%, its best year since 2003.
The year started out poorly for most CTAs as bonds chopped around, but soon many long-term trends emerged. Crude oil bottomed out at about $50 in January and moved basically straight up to $100 by year-end. Grain markets, particularly soybeans, had a sustained rally for much of the year; most currencies appreciated against the dollar and bond markets, after chopping around for the first two quarters, experienced a smooth trend for the second half of the year.
“It was a good year. The obvious reason is that you had more tradable trends.” says Sol Waksman, president of Barclay Trading Group.
“The thing that has been delightful for me this year is that this is the first year where I don’t have to depend on one market or another,” says Bob Pardo, principal of Pardo Capital Limited, whose diversified program returned 63.73% in 2007. “It has been entertaining to watch. When I don’t make money in one area, I am making money some other place. Everything is moving. For example, soybeans in April were around $7.25 and by July they were up to $9.25.”
Most defining for 2007 was what happened in the equity index sector, not so much for what the trends and corrections — which were major — did for trading programs as much as the general return of equity market volatility did for trading models. The Dow Jones Industrial Average, S&P 500 and Russell 2000 each hit their all-time high, and also saw a couple of their largest single day drops in history.
Many nimble managers were able to profit in the volatile equities sector, but those who didn’t benefited from the trends in other sectors set off by equity dislocations. “To a very large extent currencies, interest rates and energies all trade off of the stock market,” Pardo says.
Michael Clarke, whose numerous programs all performed well in 2007 (see HYPERLINK "Futures/Monthly%20Issues/Issues/2008/03/Editorial/Trader%20Profile/TraderProfile.xml" Trader Profile), says the added volatility in equities is generally good for managed futures.
“It could be very good for us — gold could go through the roof and bonds can get cratered.” He adds cautiously that the volatility is a double-edged sword. “It could reverse on a dime. Say people get real frightened and bid the interest rate products up and then all of a sudden the stock market rallies, the Dow rallies 500 points in a day — we could easily sell these things back off severely in one day, even if it might be bogus. Volatility can hurt us too,” he says.
Generally the programs that prospered in 2007 were short- to medium- term traders who were able to jump on trends and get out with a profit. “Look at a weekly chart of the S&P and you had a pretty good rally in 2007 and you had a big break and then you had another rally. Those were all eminently tradable,” Pardo says.
Whether a manager profited in equities or got caught in the corrections depended on the pace of their trading, Pardo says. “If you look at bonds you had a very clear uptrend ever since June of ’07 and there is no sign of it abating. There are some kinds of markets that if your systems can’t make money in it, it is probably not going to make money in any market. Same thing was true with oil. It has to do with the pace at which people trade.”
Roger Pegorsch, principal of Commodity Futures Services, had his best year since 2002, earning 61.7%. The program earned more than 21% in the energy sector alone.
“I did well in energies, I did well in interest rates, I didn’t do really well in ags and metals,” Pegorsch says.
One of the reasons he did so well was his diversification. His IPATS system trades 35 markets and has a significant allocation to physical commodities. He also allocates 16% to non-U.S. interest rates. One of his most successful markets was Japanese Government Bonds.
Ram Management Group’s Aggressive program earned 58% in 2007. Principals Bob Moss and Jeff Earle attribute the success to their diversification and reliance on both trend and counter-trend elements. Ram earned strong returns in gold, corn, wheat, five-year notes and Dow futures. The program was able to catch both the trends and major reversals in the stock indexes.
Going electronic
Most managers we spoke with said the transition to electronic trading in traditional commodity markets had a positive impact on many programs. Pegorsch began trading agricultural markets for the entire day, which improved execution on reversals. “A lot of the time we would get filled at the open. If news in Asia moved the market, we would have to wait until the next day’s open to get in, usually with the price going through our entry or exit level. In the past you had some news from Asia at three in the morning and wait. It has been a positive.”
Traditional markets
While 2007 was a strong year for CTAs, not all trend followers did well. Managers who are more sophisticated than simple trend followers seemed to do well, as did programs diversified across numerous sectors. The large long-term trend followers with heavy concentration in the financials did not top the charts. “It was an unusual year. There were some firms that you would have thought would have made money in this environment and they didn’t,” Waksman says. “Does trend following still work? Yeah, but I think it requires firms that are involved in it to be a little more sophisticated.”
It was a year that once again awarded diversification and those trading the more traditional agricultural markets. John Di Tomasso, who operates a value based commodity trading program that earned 41.78% in 2007, expects that to continue. “Food is becoming an issue. We’ve got low carryover stocks, we’ve got potential hoarding, we’ve got speculative interest, we’ve got inflation, a declining dollar, China and India. Holy mackerel — you add up all of the factors and these commodities are not just going to go to historical norms, they are going to go well beyond that,” he says.
Several managers attributed the positive environment for CTAs in 2007 to the return of equity volatility and because of that, expect 2008 to be another strong year. CTAs tend to do well when equities struggle and equities have gotten off to a terrible start in 2008. If your 401(k) is in need of repair perhaps an allocation into managed futures is the place.
QIM: Growing profitably
When Futures profiled Quantitative Investment Management (QIM) in 2005, it was an up-and-coming CTA with a strong, if short, track record and a pretty complex story to tell. QIM principal Jaffrey Woodriff has always had difficulty describing his program — settling on “quantitative behavioral finance,” back in 2005. Well someone was listening, or at least following, the performance as QIM has grown from below $10 million under management at the beginning of 2005 to nearly $3 billion under management today. That type of growth is tough on a CTA and Woodriff attributed a relatively lackluster 2006 performance, 5.3% in the global program, to significant slippage.
QIM would enter all of its trades in the first half hour of trading, which became unworkable as they grew. The program utilizes software that takes relatively simple inputs to create a vast number of trading models. The best 800 models are used and together provide a score for each of the 26 mainly financial futures markets the strategy trades. The result is a score from -500 to 500. In the past, a score of -100 or 100 would initiate a full position, but now in addition to spreading its entries over a much longer time frame, the program scales the size of the position based on the signal score.
Woodriff says the changes have improved the overall performance of the program as well as increasing its capacity and reducing slippage all while QIM grew by a factor of five in 2007. QIM’s global program returned 28.23% in 2007 and its X3 program earned 112.73%. The X3 program has a Sharpe ratio of 2.51 and a worst drawdown of 11.44%.
“Every day the models give us different signals, sometimes [they] want to go with the trends, sometimes it wants to go against the trends. We don’t know what the models are going to tell us to do day in day out,” says principal Michael Geismar. The models are an exercise in data mining and have no bias — trend following, countertrend or otherwise. Trades last on average seven to eight days.
The program performs extremely well in volatile markets, often catching major moves and the resulting corrections. “We made a lot of money in the first two weeks of March after the big [equity drop] in February and we did very well the first couple of weeks in August [after] that big decline. Definitely when the markets got more volatile we tended to do better,” Geismar says.
The three principals: Woodriff, Geismar and Greyson Williams, all attended the University of Virginia but at different times. Prior to working as a prop trader for Societe Generale in New York, Woodriff was head trader for Blue Ridge Trading and hired Geismar. Woodriff left Soc Gen and moved back to Virginia to develop his proprietary trading strategy along with Geismar. They were eventually joined by Williams and the three launched QIM in 2002.
QIM performs best in the most liquid markets and in times of high volatility. Woodriff attributes that to the underlying logic of his data mining approach. “The more liquid the market, the more data and the more participants are generating the trading in that market. And therefore the data that we are using is more statistically significant the more liquid the market is.” While difficult to understand, the results are clear.
Jalex/Kesef Trading: reacting to markets
Randy Shell and Simcha Bluth may look like the original odd couple, but they have managed to build first a successful brokerage business, then a commodity pool and eventually their own trading programs. They are the principals of CTA/CPO Jalex Trading and Kesef Trading. Both programs primarily day-trade 30-year bond futures in a systematic strategy. The Jalex program will trade gold, wheat and copper from a long-term perspective, but 80% of trades are in the long bond.
The Jalex program returned 45.62% in 2007 and has a compound annual return of 19.40% since its launch in November 2003. A year later Kesef Trading, which returned 33.11% in 2007, was launched and has earned a compound annual return of 29.58% with a Sharpe ratio of 1.18 in that time.
Shell got his start working with Peter Steidlmayer at the Market Logic School in Chicago and he has incorporated some of the market profile logic into his trading programs.
Bluth started out developing complex options strategies in New York. The two began working together, though in separate locations, in 1996 as independent introducing brokers. Shell moved back to his home in Oklahoma, where the two CPOs/CTAs are based, while Bluth operates out of his home in Baltimore. They brought in other brokers as branch offices of their IB.
When they started, they worked with several CTAs but could not offer those managers to many of their clients because of the high minimums. That is when they decided to create CPOs to offer third party programs at lower minimums. Their first CPO was launched in 1998. They still operate that CPO though now their programs are the underlying CTA.
In 2002 the two partners began researching their own methodology. “It took us about three years to evolve what we were doing,” Shell says. “What we do is a pattern recognition type of strategy. It is an action/reaction [strategy] based on years and years of observation. The foundation of it comes out of an intraday, day-trading time frame.”
While the underlying logic may be related to what Shell learned working with Steidlmayer, he says, “What we do now is quantum leaps away from traditional market profile analysis. What we do is a combination of both of our backgrounds.”
Bluth came in and began questioning all of the assumptions in the market profile and from that analysis a unique system began to emerge, one based on recognizing patterns and reactions to those patterns. “The market profile is a subjective tool, it is only as good as the person using it, it is like a paint brush in a painter’s hand,” Shell says.
“There are certain things we noticed that happen in the market,” Bluth says. “We are waiting on those actions that [predict] a certain reaction.” This action/reaction scenario is tested so that only the most reliable patterns are used. “We are looking for patterns that generally are 70% or more reliable, Bluth says. “We knew that we could trade, we just needed to quantify it; get it down to an objective workable format that we could employ on a day in day out basis. It took us three years to do that.”
Tri Global FX: Busting rocks
The only thing more impressive than Tri Global FX’s 2007 performance, 46.78%, is that performance on a risk adjusted basis. Tri Global has a monthly standard deviation of 3.01, a Sharpe Ratio of 3.06 and has had only one losing month in its three-year history.
Despite these gaudy numbers, president and head trader Gregory Cotter takes a very down-to-earth approach to the markets. “I consider this a blue collar market, it’s a job. Put in your time every day trading. Try not to get caught with stuff, get out of stuff that is wrong and just keep moving forward.”
Cotter spent more than 20 years trading the off-exchange forex markets for some of the largest institutional banks, including Societe Generale, Banque Indosuez, Credit Suisse and JP Morgan Chase. He developed a simple short-term approach to trading. “We don’t get into anything exotic, we are just traders. I have been doing this 25 years and the only guys I ever saw on the bank side that made money were traders,” Cotter says.
The Roslyn Heights, New York CTA keeps its head down and trades. Cotter does not attempt to find long-term trends and does not jump on bandwagon trades. “Every Tom, Dick and Harry was trying to start up a managed account or hedge fund based on the carry trade,” Cotter says. That is fine with Cotter, whose biggest profits in 2007 came during the unwinding of EUR/JPY and GBP/JPY carry trades. “We knew eventually the carry trade had to fall apart, and when it did, it was going to get ugly, and it did,” Cotter says. “We were long the carry trade, then we got short the carry trade. Last year for me was really sitting around waiting for the carry trade to die. It just got too easy and you got too many people talking about it. Every time I would go to any Web site that had anything to do with currencies, all these guys knew how to do is buy euro/yen,” he says.
Cotter uses a mix of technicals and fundamentals. The program made money in the downtrend in EUR/JPY and GBP/JPY but Cotter is not shy in switching things up or reversing a trade. “We’ve covered our shorts in sterling/yen and euro/yen and now we are investing in some longs. We have a system here but if I see the stock market [turn], I’m not waiting for the chart to tell me to sell. I hit a bid and take some profits on it.”
While he will hold some positions long term, he is a day-trader at heart. “We have stuff on that we will carry for a month or two and we have stuff that we are out within minutes. Given my druthers I would rather walk in every day with a clean sheet, day-trade and go home at night,” Cotter says.
Cotter sees his 2007 performance as somewhat of an anomaly but not in its consistency. “We look for 1% to 2% a month. That is what we trade for. If we can do better than that, great. If we can do that, we figure we’re good. We keep ourselves in the game and when things occur that we can make more than that we go for it, but we just try and make money every month,” he says.
