November 2004–May 2005: The Commodities Reflation Boom and Its Decision Training Legacy

On November 15, 2004, copper opened at 1.44 per pound and natural gas at 7.44 per million British thermal units. Six months later, on May 12, 2005, those same contracts would close at 1.47 and 6.51 respectively. The bare price comparison tells a story of modest moves—2.1% for copper, -12.5% for natural gas—but the number that mattered most was the journey between those endpoints: drawdowns that reached 9.7% in copper and 27.4% in natural gas, with single-bar swings of 8.6% and 17.5% tearing through positions regardless of directional conviction.

This was the heart of the 2000s reflation trade, a period when monetary expectations were shifting and commodity markets absorbed that signal with unusual sensitivity. Nineteen distinct futures markets—covering base metals, precious metals, energy, grains, and soft commodities—all entered the same November-to-May window, each driven by overlapping macro forces while retaining their own idiosyncratic volatility profiles. The collective data, spanning 120 to 124 trading days per contract, forms a rich decision-training laboratory.

Energy markets set the tone early. Natural gas dropped 12.5% net over 122 trading days, yet endured a 27.4% maximum drawdown—a gap that reveals how concentrated losses can punish even trend-following approaches. WTI crude oil moved more modestly, closing up 3.6% but with an 18.2% drawdown and 7.4% single-bar moves. Heating oil, starting November 16, added another 19.6% drawdown layer on top of a 3.8% net gain. The message across the energy complex was consistent: the direction was there for some, but the process of capturing it demanded extraordinary tolerance for interim pain.

Agricultural markets delivered the most extreme statistics of the window. Oats Historical Replay: 2004-11 to 2005-05 captures a contract that fell 12.4% over 122 trading days while experiencing a 28.6% maximum drawdown—the largest of any single contract in the entire dataset, according to FinFable's scenario data for this window. The largest single-bar move in oats reached 11.3%, meaning roughly an entire year's directional move could be erased or amplified in a single session. Corn declined 4.6% with a 14.1% drawdown; wheat fell 5.3% with an 18.1% drawdown; live cattle slipped 1.2% with a 16.4% drawdown. Across the grain corridor, the pattern held: directional moves were subdued, but the internal volatility told a much harsher story.

The soft commodities offered the window's most dramatic performance divergence. Coffee Historical Replay: 2004-11 to 2005-05 stands out as the strongest performer: a 30.8% gain over 122 trading days, yet participants still endured an 18.2% drawdown and 6.7% single-bar swings along the way. Cotton gained 22.6% with a 13.7% drawdown; orange juice rose 26.9% despite a 12.3% maximum pullback. Sugar, by contrast, fell 4.9% with a 13.1% drawdown and 7.6% single-bar moves—demonstrating that losing markets could still punish process discipline as severely as winning ones. Cocoa dropped 17.5% with a 21.7% drawdown and 8.3% largest bar. The lesson embedded in these numbers is not about predicting which soft commodity would win; it is about understanding that the process of being right required surviving drawdowns that often dwarfed the net directional outcome.

Precious metals occupied a calmer but still volatile corridor. Gold declined 3.5% with a 9.5% drawdown and 3.3% largest move; platinum fell just 0.6% with a 7.6% drawdown and 4.2% single-bar swings. Silver, however, returned 8.2% loss over 120 trading days with a 19.9% maximum drawdown and 9.4% single-bar moves—the most severe volatility-to-return ratio among the precious metals group. The relatively contained drawdowns in gold and platinum reflected their traditional safe-haven role, while silver's industrial sensitivity exposed it to sharper reversals.

The base metals segment, anchored by copper, illustrated how even a winning direction could feel like losing. Copper gained 2.1% but fell as much as 9.7% before recovering; the 8.6% largest single-bar move meant that any position sized without accounting for intra-period turbulence risked forced liquidation at the worst moment. Soybeans gained 12.1% with an 11.0% drawdown and 5.3% largest bar; soybean meal surged 23.8% despite an 11.3% drawdown and 5.9% single-bar moves; rough rice rose 6.6% with an 18.0% drawdown and 5.4% largest bar. Across all nineteen scenarios, the common thread was not the direction of travel but the turbulence of the journey—drawdowns ranging from 7.6% to 28.6% demanded that participants maintain process discipline even when their directional thesis appeared to be failing.

The macro backdrop for this episode is widely attributed to the transition from the deflation concerns that dominated the early 2000s toward reflation expectations. Commodity markets, historically lagging indicators of inflation dynamics, began repricing across the board in late 2004. The US dollar's path, Federal Reserve signaling, and global demand signals all contributed to the shifting landscape. Yet the material record makes clear that these macro forces manifested differently across individual markets—some rising, some falling, most experiencing drawdowns far exceeding their net directional moves.

The decision architecture of this window was remarkably consistent across all nineteen scenarios: 21 decision points per replay, with 7 designated as critical junctures. This uniform structure allowed participants to compare their process performance across vastly different market outcomes—from coffee's 30.8% gain to cocoa's 17.5% loss—while confronting the same psychological pressures. The calibration question was identical in every market: when the drawdown reaches X%, do you hold, reduce, or exit?

For decision training today, this window offers several structural lessons. The first concerns the relationship between net outcome and process quality: coffee's 30.8% gain could easily be misread as evidence that holding through an 18.2% drawdown was the correct decision, yet oats' 12.4% loss with a 28.6% drawdown demonstrates that the same holding-through-drawdown process produced a catastrophic result. The differentiating factor was not the process but the underlying market direction—a reminder that disciplined execution and favorable outcome are separate variables that only converge over many repetitions.

For markets today, this episode established the template for how commodity exchanges would later position themselves as inflation-sensitive diversifiers. The divergence between energy, agriculture, and precious metals during a single macro regime underscored that commodity indexing is not a monolithic bet on direction but a basket of idiosyncratic risks requiring active management.

For institutional frameworks, the 2004–2005 window prompted re-evaluation of risk models that had historically treated commodity volatility as secondary to equity risk. The drawdown magnitudes—27.4% in natural gas, 28.6% in oats, 21.7% in cocoa—demonstrated that commodity positions could inflict equity-equivalent losses while generating fractionally small net returns. Risk allocation frameworks gradually incorporated commodity-specific volatility regimes rather than applying uniform risk weights.

For traders training decisions today, the enduring lesson is the asymmetry between process and outcome. The nineteen scenarios in this window collectively demonstrate that disciplined process is necessary but not sufficient—market direction is a separate variable that modulates results independently. The 7 critical decision points per scenario represent the moments where this tension was most acute: holding through a 20%+ drawdown requires either extraordinary conviction or robust position sizing that accommodates extended adverse moves. The window's educational power lies not in revealing which direction was correct, but in showing how the same decision framework produces radically different results depending on the underlying market environment that remains unknown at the time of each decision.