Commodities

Commodity Trend Following on MCX: The Oldest Systematic Trade

Saral Team · 1 August 2026

Managed futures is the oldest systematic strategy still running at institutional scale, and commodity trend following is its core. It predates factor investing, survived the arbitrage of most other published anomalies, and rests on a mechanism that is unusually easy to state.

Physical markets adjust slowly

An equity can reprice instantly. A discounted cash flow revision is arithmetic — new information arrives, the model updates, the price moves in a tick.

A commodity cannot do that, because the thing being priced is physical.

Copper is short. The price rises. Does supply respond? Only when someone opens a mine, and a new copper mine takes the better part of a decade from discovery to first production. Wheat is short — supply responds next season, at the earliest. Crude is short — a refinery cannot flex output overnight, and drilling programmes are budgeted quarters ahead.

So a genuine shock to physical supply or demand produces a persistent price adjustment rather than an instant repricing. The market spends months finding the new clearing level, and the path there is a trend.

The reverse holds too: excess supply cannot be switched off, so gluts persist and prices grind lower for extended periods. Trend following captures both directions of this.

Moskowitz, Ooi and Pedersen’s work on time-series momentum tested 58 futures markets over 25 years and found the effect in essentially all of them, with returns that held up across sub-periods and market conditions.

Continuous futures, and the caveat that comes with them

A commodity future expires. To get a multi-year price series you have to stitch contracts together — hold the front month, roll into the next before expiry, repeat. The result is a continuous futures series, and it is what almost all commodity research uses.

It is also not the return of anything you can hold. Two things get baked in:

Roll yield. When the futures curve is in contango (further contracts more expensive), rolling means selling cheap and buying dear — a persistent drag. In backwardation the reverse: a persistent tailwind. For some commodities in some regimes, the roll dominates the spot move entirely. Natural gas has spent long stretches in steep contango where a long position lost money while spot prices were flat.

Stitching convention. How you splice contracts — on volume, on open interest, on a fixed calendar day, with or without a price adjustment at the seam — changes the series. Different providers produce different histories for the same commodity.

The practical reading: treat a continuous series as a signal series, good for deciding whether copper is trending. Do not treat its cumulative return as the return you would have earned, and expect real roll costs on deployment.

Risk parity is not optional here

This is the design decision that matters most, and equal weighting actively breaks the strategy.

Commodity volatilities differ enormously. Gold has historically run in the mid-teens annualised. Natural gas has run several times that, with regular periods above 100%. Crude sits somewhere between and occasionally spikes past both.

Allocate 25% of capital to each of four commodities and the risk allocation is nothing like 25% each — a natural gas position contributes several times the variance of a gold position of the same rupee size. You have written down a diversified commodity basket and are running a natural gas bet with a commodity label.

Risk parity sizes each position so it contributes equally to portfolio variance: large positions in gold, small ones in gas. It looks lopsided on a holdings table and is correct on a risk report.

Stacking absolute and relative conditions

The template applies two trend tests rather than one:

Absolute: price above its 200-day average. A regime filter: is this commodity in an uptrend at all?
Relative: rank the survivors by three-month momentum and hold the strongest four.

Neither alone is sufficient. Relative-only holds the best of a falling set, which in commodities means being long a bear market. Absolute-only holds everything trending, which in a broad commodity bull market means holding everything and diluting the signal.

Together they answer both questions: is this market trending, and is it among the strongest of those that are?

MCX specifics

The exchange is smaller than its global peers. MCX is India’s dominant commodity exchange, but volume concentrates heavily in gold, silver, crude and natural gas. Base metals and agricultural contracts are much thinner, and a strategy that ranks into them will find fills worse than the backtest assumes.

Contract sizes are large and lot-based. MCX contracts have specified lot sizes, and the notional per lot in gold or crude is substantial. This is not a strategy you can run in a small account, and the mini contracts that exist for some commodities have their own liquidity profile.

Commodity derivatives are taxed differently from equity. CTT applies rather than STT, at different rates, and the income treatment differs from equity delivery. Model your own costs; the defaults here are generic.

Agri contracts have carried regulatory risk. SEBI has suspended derivatives trading in several agricultural commodities for extended periods. A backtest over a suspended contract is trading something that was not available.

Try it

The MCX Commodity Momentum template requires price above its 200-day average, ranks survivors by three-month momentum, holds the top four under risk parity, and rebalances monthly.

The instructive change: switch the weighting to equal and re-run. Returns may look better in a period when the most volatile contract happened to trend. Look at the drawdown, and at which single holding explains most of the equity curve’s shape — that is the concentration risk parity was removing.