Commodity Trading Strategies That Work (Until They Don't)
Commodities: Markets, Performance, and Strategies
Editors: H. Kent Baker, Gregory Filbeck, Jeffrey H. Harris
ISBN: 9780190656010
Hilary Till, Joseph Eagleeye, and Richard Heckinger write from the trenches. This chapter is less theory, more “here is what traders actually do, how they blow up, and what you should not copy.”
The two main strategy types
Trend-following CTAs dominate managed futures. Over 70% of funds use systematic trend models. They scan 80+ markets, ride 1-6 month trends, cut losers fast, and lever into winners. Alpha, if it exists, comes from dynamic position sizing, not from predicting fundamentals.
AQR backtested time-series momentum from 1903-2012 across 59 markets. Gross returns averaged 20% annually with a 1.0 Sharpe net of fees. Worked in every decade. Low correlation to stocks and bonds. The catch: behavioral biases (anchoring, herding) and hedging flows may be why trends persist. If those forces fade, so does the edge.
Calendar-spread traders play a different game. Instead of directional bets, they trade the price gap between delivery months. Natural gas summer-winter spreads mirror inventory cycles. Index roll dates create predictable flow that speculators try to front-run.
Three mistakes that kill accounts
1. Targeting returns instead of risk
Gerald Loeb said it best: “Just when you think you found the key to the market, they change the locks.” Strategies have life cycles. If you demand a fixed return threshold, drawdowns get worse.
The natural gas bear calendar spread example is brutal. The strategy worked beautifully from 2004-2006. By mid-2006, profits halved. Traders who doubled position size to maintain returns got destroyed in July-August 2006. Losses were twice their year-to-date gains.
Proprietary firms target risk. Fund investors who chase past track records and pay short-term bonuses without clawbacks invite disaster.
2. Wrong sizing in illiquid markets
Commodity markets have nodal liquidity. Big trades happen when commercials need to hedge, not when you want to exit. If a spread moves several standard deviations with no news, someone is liquidating in distress.
Natural gas keeps appearing in blowup stories: MotherRock, Amaranth, Saracen. The market looks scalable and predictable until it is not. Keep positions small relative to daily volume and open interest. Sounds obvious. Funds routinely ignore it after success brings too much capital.
3. Treating trading as an intellectual puzzle
Ralph Vince’s quote hits hard: “Anyone who claims to be intrigued by the intellectual challenge of the markets is not a trader.” You need tolerance for pain most people cannot handle.
Taleb’s example: a strategy with 15% excess return and 10% volatility only has a 54% daily win rate. If losses hurt 2.5x more than gains feel good, executing the strategy becomes almost impossible.
Amaranth: the case study
Amaranth Advisors lost $6.6 billion in September 2006, mostly on natural gas calendar spreads. The strategy had logic: long winter, short summer, betting on weather shocks. It worked until it did not.
The problem was size, not concept. Senate investigators found Amaranth controlled up to 40% of NYMEX open interest in some winter contracts. Its January 2007 position equaled total U.S. residential natural gas consumption for that month. On one day, it was 70% of trading volume in March-April spreads.
Amaranth entered at extreme spread levels. If spreads reverted to six-year norms, the fund faced 22-36% portfolio losses. That is before the critical liquidation cycle kicked in.
Physical market participants (storage operators, pipeline companies) were the natural counterparties. They had no urgency to unwind at Amaranth’s timeline. Citadel eventually took the book at a $2.15 billion discount and unwound it over weeks.
Robert Greer’s lesson: the market punished size, not rewarded it.
MF Global: a different kind of disaster
MF Global’s 2011 collapse was not bad commodity bets. It was a liquidity crisis from levered European sovereign debt repo-to-maturity trades. Customer funds got commingled with proprietary positions. $1.6 billion went missing.
Unlike Amaranth’s market risk lesson, MF Global teaches due diligence: segregated customer funds matter, and broker counterparty risk is real even in regulated markets. Customers were made whole eventually, but it took years.
What survives contact with reality
Trend-following scales but faces capacity limits. Calendar spreads offer edges but require modest sizing and respect for commercial flow. Every strategy has a shelf life.
The chapter’s honest message: commodity trading rewards discipline, sizing, and emotional endurance more than clever models. The blowup stories are not outliers. They are what happens when smart ideas meet too much leverage in markets without two-sided liquidity.
Navigation
← Previous: Commodities in Emerging Markets (Chapter 19)
Next: Asset Allocation and Commodities (Chapter 21) →