A Roll Gap Is Not a Contract Return

Keep both contract identities on the page

Create two wholly fictional delivery contracts for the same generic underlying. Contract A is the earlier delivery month and contract B is a later month. A is quoted at 100 on day zero and 101 on day one. B is quoted at 103 on day one and 104 on day two. No exchange, commodity, actual delivery calendar or observed market history is implied by these invented values.

Stipulate one long contract throughout each holding interval. The assumed actions are to open A at 100, close A at 101, open B at 103 on the roll day and close B at 104. Each price is an explicit teaching assumption, not a claim that an order could have executed there. The example contains enough information for its narrow ledger, but not for an actual instrument backtest.

Identify what the splice actually changes

A joined series that selects A on day zero and B from day one onwards reads 100, 103, 104. Its endpoint difference is 104 minus 100, or four points. The first displayed transition includes both A's one-point movement and the two-point difference between the contracts on day one. A chart retaining only the selected contract each day no longer displays that decomposition directly.

At the roll timestamp, B minus A equals 103 minus 101, or two points. This is a cross-contract level difference, not a change in A between two dates or a change in B between two dates. Calling the four-point endpoint difference a held-contract gain silently treats different delivery commitments as one unchanged instrument. The numerical precision of the joined series cannot repair that identity substitution.

Build the same-contract ledger

For the first holding interval, A's change is 101 minus 100, or plus one point. For the second, B's change is 104 minus 103, also plus one point. Summing the changes of the contracts actually stipulated as held gives two points. Algebraically, four splice points equal one A point plus two gap points plus one B point. The gap belongs to the joining operation, not to either holding interval.

Assign both fictional contracts a generic multiplier of $10 per price point and hold one contract at a time. The resulting conditional gross amount is one times $10 plus one times $10, or $20. CME's money-calculation documentation grounds the relevant distinction: settlement variation is calculated against the trade or preceding settlement price of the applicable contract. It does not substitute another delivery month's quote as that contract's earlier price.

Do not book the gap as an entry payment

In the fair-entry valuation convention assumed here, the newly opened B futures commitment starts with zero contract value before transaction costs. Its quoted futures price of 103 is not a premium paid to acquire an asset worth 103. Margin remains a separate funding requirement. Closing A at 101 and opening B at 103 therefore does not, by itself, create an immediate two-point cash loss for the long position.

It does not create an immediate two-point gain either. If B remains at 103 immediately after entry, its price-based variation from entry is zero: 103 minus 103. Future changes in B are then measured from its own relevant reference. A's previously accumulated movement remains in A's ledger. Keeping price quotation, contract value and settlement cash distinct prevents a roll gap from being counted as an invented transfer.

Charge the four stipulated sides

Choose a fictional $2 charge per contract side. The sequence contains four sides: opening A, closing A, opening B and closing B. Four times $2 equals $8. Subtracting this selected charge from the conditional $20 gross amount leaves $12 toy net. Charging only the two roll-day sides would omit the initial entry and final exit, producing an inconsistent whole-example cost ledger.

The $12 is conditional on the exact assumed prices, quantities, multiplier and chosen charges. It excludes any additional spread, slippage, collateral income, financing effects, taxes and other fees. It is not historical investment P/L, a backtest, an annualised return or evidence that rolling is attractive. No capital denominator has been supplied, so converting it into a percentage investment return would require an additional explicit assumption.

Expiry and carry require separate evidence

CME's explanation of expiration and contract rolls describes rolling as offsetting an existing contract and establishing a later one. Actual last-trading dates, delivery obligations and settlement procedures depend on the product. This fictional A-to-B switch specifies none of those real-world details. It cannot support a claim about a particular commodity's delivery exposure, appropriate roll date or achievable liquidity during an expiry transition.

Nor can these four quotations identify a full carry decomposition. The higher later-month quote does not, on its own, measure financing, storage, income, convenience benefits or subsequent convergence. Those concepts require the relevant underlying, contract terms and time horizon. The example demonstrates an accounting distinction only. It would be equally inappropriate to infer a guaranteed future loss from the gap or to dismiss every economic effect associated with maintaining futures exposure.

Preserve the audit trail before transforming prices

A reproducible roll-aware record keeps the contract identifiers, observation times, selected roll rule, both roll-day prices and the position quantities. A joined reference can remain useful for a limited descriptive purpose, but the original identity map is needed to separate same-contract movements from joins. Changing the series through an adjustment method changes its interpretation; it does not create missing orders or evidence of actual execution.

The final reconciliation should remain visible: four points in the unadjusted splice, two points of stipulated held-contract movement, $20 conditional gross and $12 after only the chosen $8 charge. The two-point roll gap is neither an instant long-position loss nor a gain. Readers can challenge every assumption because the ledger exposes them, rather than asking a continuous line to stand in for contract-level investment accounting.

Source record

1. CME Group: Understanding Futures Expiration and Contract Roll
2. CME Clearing: Money Calculations for Futures and Options
3. CME Group: Mark-to-Market