Knowledge base · Market structure

Futures roll mechanics

Educational reference from the platform knowledge base — written agent-readable first, rendered here for humans. Mechanics, not advice: nothing here is a recommendation to buy or sell any security.

Futures roll mechanics

Definition

Rolling a futures position means closing the expiring contract month and opening the same exposure in a later month, because futures have finite lives and most holders do not intend delivery or final cash settlement. The roll is a simultaneous two-leg trade whose cost or credit is the calendar spread between the months at the time of the roll.

How it works / structure

  • Timing: liquidity migrates from the front month to the next contract in a recognizable window before expiration (index futures concentrate the roll in the days around quarterly expiration; physical commodities roll ahead of first-notice/delivery dates).
  • Cost: roll P&L = (price of new month − price of old month), sign depending on curve shape — rolling long positions in contango pays the spread; in backwardation it collects it. Recurring roll cost/credit is the mechanical component of carry (strategy-futures-carry).
  • Continuous series construction: backtests stitch contract months into a continuous series; the adjustment method (back-adjusted, ratio-adjusted, unadjusted) changes historical P&L and signal values — a stated methodology is mandatory for any futures backtest on this platform.
  • Simulation parameters: roll window rule (days before expiry/first notice), which month to roll into, spread cost at roll, and the price-series adjustment convention.

When it applies

Every futures position held across an expiration cycle, every futures-based index or fund (their published roll schedules are front-run-able and studied), and every backtest on futures data (the adjustment convention silently shapes results).

Risk profile & failure modes

  • Roll congestion: when most of the market rolls in the same window, the calendar spread can move against the roller; large funds with published schedules are most exposed.
  • First-notice accidents: holding a physical-delivery long past first notice can trigger delivery processes the account cannot support (instrument-futures-contract).
  • Series artifacts: signals computed on unadjusted stitched series see phantom jumps at roll dates; on back-adjusted series, long-history percentage returns are distorted. Neither is wrong — unstated is wrong.
  • Liquidity mismatch: rolling into a month with thin liquidity costs more than the visible spread.

Evidence & limits

Roll mechanics and liquidity migration are exchange-documented market structure. The magnitude of roll-window price pressure from scheduled fund rolls has been studied (commodity-index roll literature) with mixed and time-varying results — specific “roll yield capture” strategies are treated as unproven unless cited in their own entries.

Falsifiable-thesis examples

Illustrations only, not signals:

  • “The front-to-second calendar spread on X will widen by at least T ticks during the five sessions before quarterly expiration” — falsified by the spread series in that window.
  • “Rolling position P five days before first notice will incur a total roll cost under C over the next four cycles” — falsified by the summed roll P&L in replay.

Cross-references

  • Instrument basics: instrument-futures-contract, ms-contract-specs, ms-futures-margin
  • Strategies built on the curve: strategy-futures-calendar-spread, strategy-futures-carry, strategy-inter-market-spread
  • Extended: ext-commodities

The agent cites this page.

Inside the platform, this entry is live context: the AI reasons from it, quotes it, and grades against it. Make your case.

Inquire about founding membership