Seasonal futures spreads: how they work

What turns a calendar spread into a seasonal study?

The spread is simply a relationship between delivery months. It becomes a seasonal study when the same relationship is tested over the same part of the calendar across multiple historical years. Repeatability requires the same month pair, calculation order and equivalent dates.

Three layers of evidence

A high win rate with tiny gains and occasional large losses can be weaker than a lower win rate with a favorable distribution.

Why curve regime matters

The same seasonal window can begin in steep contango one year and backwardation another. If a pattern only worked under one regime, averaging all years together can hide that dependency.

Use the forward curve to see whether today's contract pair resembles the historical years that contributed most to the pattern.

How to avoid false confidence

Inspect individual years, test nearby date ranges and compare several history lengths. A seasonal average is most useful when the result remains understandable after those robustness checks.

Example: the same spread can tell different stories in different windows

A December–May spread might strengthen during one part of the year and weaken during another. That does not mean the market is inconsistent; it means the underlying physical forces change through time. Harvest pressure, storage demand and expectations for the next crop can affect the two delivery months differently as the calendar advances.

This is why the seasonal window must be part of the strategy definition. “December–May” alone is not enough. A complete hypothesis also states when the relationship is expected to move and why that timing makes economic sense.

After defining the window, compare how stable the result is if you shift the dates slightly. Robust timing should not depend on one exact day chosen in hindsight.