The three fundamental spread mappings form the methodological pivot of oil fundamental research. Using “supply − demand = inventory” as the base equation, they establish a one-to-one correspondence between three types of spreads (inter-regional spread, crack spread, calendar spread) and the three fundamental factors (marginal supply, demand intensity, inventory-change intensity), replacing lagged and hard-to-forecast quantity indicators with real-time price indicators. The core of the framework is a single perspective reversal: from “measuring by quantity” to “measuring by spread.”
The Framework As It Stands
This section is compiled from research working notes: the original framework’s structure, terminology, and key formulations are preserved, including editorial bridging and externally sourced factual annotations; diagrams are drawn by the compiler following the original text’s structure.
The core question: how does fundamental research actually land in practice?
This framework answers one fundamental question: how does fundamental research actually land in practice? The main thesis is a single perspective reversal — replacing “quantity measurement” of supply/demand/inventory with “spread measurement,” because quantitative forecasting is extremely difficult whereas prices transact every moment and data sources are abundant.
The implicit main line — reversing from the “quantity perspective” to the “price perspective” to escape the meter-reader dilemma
Western microeconomic price theory assumes that at equilibrium “supply quantity = demand quantity,” but in reality supply never precisely matches demand; the true relationship is supply − demand = inventory. The conventional approach is to study supply-demand balance tables, tabulate and forecast quantities, then form a directional view — but quantitative forecasting is extremely difficult. With crude oil as the example: U.S. inventory data forecasts have never been accurate. Even using drones to monitor pipeline pump valves and infrared imaging to survey tank liquid levels one by one, forecasts remain inaccurate.
This is the meter-reader dilemma: a car in a garage, with the odometer read once daily at noon. Drawing yesterday’s speed curve is impossible; all that can be done is subtract yesterday’s reading from today’s and divide by 24 hours to obtain an average speed — the average speed and the integral of the actual speed curve are equal in quantity, but the process is entirely fictitious. Judging the current supply-demand balance from quantities and forecasting the future is exactly this meter-reader dilemma.
The solution: convert quantity indicators into price indicators, because prices transact and occur every moment and data sources are highly abundant.
The implicit closing loop — strict one-to-one correspondence between the three spreads and the three factors
Arbitrage is fundamentally about trading spreads. Mapping business models and arbitrage models yields three types of spreads:
- Calendar spread (between spot and futures, or between futures contracts of different delivery months)
- Inter-regional spread (textbook term: cross-market spread; the price difference for the same commodity across different regions)
- Crack spread (textbook term: cross-product spread; the price difference between a refined product and its feedstock)
Four trader practices: inter-regional arbitrage (buy in a cheap region, sell where prices are higher); contango carry (buy spot and store, sell short on the futures board at a higher price); backwardation trade (sell inventory first, replenish via futures if the futures price is cheaper); crack spread arbitrage / tolling (lock in the diesel crack spread).
The standard model — the correspondence between cross-regional commodity fundamentals and the three types of spreads
The standard model has two defining characteristics: ① a commodity market cannot be resolved with a single balance table — it must be zoned by region; ② all quantity factors (supply, demand, inventory) are converted from quantity measurement to spread measurement. Within this framework, the three spreads correspond strictly to the three fundamental factors:
Inter-regional spread → Marginal supply capacity
When supply is deficient, imports compensate; when supply is excessive, exports absorb it. Cross-regional trade changes regional supply and therefore constitutes marginal supply volume. It reacts quickly and flexibly, because traders are purely profit-driven and will act whenever there is a spread and a margin. Hence the inter-regional spread corresponds to “marginal supply capacity.”
Crack spread → Demand intensity
A large crack spread (large gap between refined-product price and crude price) signals strong demand. Analogy: if the price of bread were ten or a hundred times the price of flour, the streets would be full of bakeries, and demand for flour would necessarily be strong. The logic also runs in reverse: if the gasoline crack spread is strong, it will inevitably weaken in the future — because global refining capacity is in surplus, all refineries will increase gasoline output, and within a limited time gasoline will be in surplus; the logic and mechanism correspond one-to-one.
Calendar spread → Inventory-change intensity
Inventory is the result of the supply-demand balance: when supply exceeds demand, inventory rises; when supply falls short of demand, inventory falls — forming a price-side real-time counterpart to the quantity description in The Four Phases of the Inventory Cycle: Diagnosing the Short Cycle. The reason traders willingly engage in contango carry is that storing is profitable and will continuously add to inventory — not because they expect oil prices to rise and therefore store; if supply exceeds demand but there is no contango profit, that market is not genuinely in oversupply. Contango profit reverse-validates “genuine oversupply”; the logic corresponds one-to-one.
The implicit reverse-deduction — the logic corresponds one-to-one, and spreads can reverse-deduce future fundamentals
Spreads are not only result indicators; they can also reverse-deduce: a strong gasoline crack spread → it will inevitably weaken in the future (global refining capacity in surplus → production increase → surplus); contango profit validates whether a market is genuinely in oversupply. These spreads are encountered in daily work, but once integrated into a system, the understanding of strategy construction, price volatility, and fundamentals becomes considerably deeper.
Core mapping table
| Spread Type | Fundamental Correspondence | Mechanism |
|---|---|---|
| Inter-regional spread | Marginal supply capacity | Cross-regional trade changes regional supply; profit-driven traders react quickly |
| Crack spread | Demand intensity | Large product-crude gap = strong demand; can reverse-deduce future (refining surplus → reversal) |
| Calendar spread | Inventory-change intensity | Inventory = supply-demand result; contango profit validates “genuine oversupply” |
Compiler’s Perspective
Coordinates: Category = Energy and Commodities / axis_h = Methods / axis_v = Why It Is So
Interface layer
What this framework resolves is not “which spread matters most” but the fundamental diagnosis of “why quantitative forecasting is bound to distort.” Errors in application are almost exclusively of two kinds.
The first kind — treating the meter-reader dilemma as a data-quality problem rather than a methodological one. The common reaction is “if I could get better real-time inventory data, that would be enough,” but the framework’s core argument is: even using drones to monitor pump valves one by one and infrared imaging to tally tank levels, U.S. inventory forecasts have long remained inaccurate. The problem is not data precision; it is that point-to-point quantitative snapshots cannot capture the process — today’s odometer reading minus yesterday’s yields only a 24-hour average speed, not yesterday’s speed curve. Spread data is superior precisely because it is the continuous result of transactions, reflecting at every moment the balance of marginal buying and selling forces.
The second kind — treating the three spreads as “three independent indicators” rather than “three readings of one system.” The correct reading sequence is: first establish that the three factors are independent (supply − demand = inventory; the three must be accounted for separately — inventory cannot be subsumed into supply), then use the three spreads to read each dimension individually. If the crack spread is strengthening (strong demand) while the calendar spread shifts to a contango structure (inventory accumulation), this indicates a divergence between the demand side and the inventory side, and further judgment is needed: is it regionally strong demand against global inventory build, or a temporal mismatch? Conflating the three indicators without distinguishing dimensions produces systematic misreads in supply-demand divergence environments.
The inter-regional spread → marginal supply mapping shares its underlying logic with the principle that “price-type high-frequency indicators never lie”: both spreads and prices come directly from transaction results, bypassing the subjective statistical intermediary, so their timeliness and reliability are higher than lagged quantitative statistics. This also forms an operational closed loop with the cognitive method of inducting from the phenomenal layer (spread results) to the mechanism (three-element structure) and then deducing forward-looking views.
Proprietary increment: the reverse-deduction logic of the crack spread (spread strong → will weaken in the future) is valid only under the premise that global refining capacity is in surplus — if global refining capacity is not in surplus, a strong gasoline crack spread may persist for an extended period without triggering a reversal. The framework’s original judgment is that “global refining capacity in surplus” was a fact at the time (2019), and therefore the reverse deduction “strong gasoline crack → will inevitably weaken” holds. But if this premise changes (e.g., large-scale refinery shutdowns), that reverse-deduction chain breaks down. This is the critical boundary condition that distinguishes this entry from other “spread = fundamental signal” arguments: before applying the reverse deduction, verify whether the structural assumption about the refining-capacity supply side is still valid.
See Also
-
The Four Phases of the Inventory Cycle: Diagnosing the Short Cycle — The calendar spread reads inventory-change intensity and is the real-time price-side counterpart to the four-phase inventory cycle; inventory-build/draw signals can be cross-validated against contango/backwardation structures
-
High-Frequency Data: Becoming One with the Running Economy — Crack spreads, inter-regional spreads, and calendar spreads all belong to real-time price data, sharing the same underlying logic as “price indicators never lie” in the high-frequency indicator system
-
The Theory of Cognitive Algorithms: Integrating Deduction, Induction, and Dialectics — The quantity-to-price reversal is a methodological example of inducting (from the empirical regularity of failed quantitative forecasting) to deduction (spread = real-time fundamental signal)
-
The Three-Step Macro Diagnosis: Empirical Regularity, Logic, Data, Pricing — The spread system constitutes the “data layer” of commodity fundamentals; the three-spread mapping in this entry is the operationalization of the three-step diagnosis in the oil market
Sources
- Compiled research working notes · archived 2026-07
External source: External course (identity stripped per de-identification protocol), Section 3.3, Special Topics in Oil Research; date November 28, 2019; course name and instructor have been stripped following the de-identification method; the framework body is preserved in full.