Prediction markets price the future continuously. Somewhere right now there is a number attached to whether a central bank cuts rates next Thursday, whether a ceasefire holds, whether a film opens above a given figure. This paper collects every such market that settles inside the window you have chosen — seven days, a month, six months or a year — decides what the prices say will happen, and reports it in the past tense as though the week were already over.

The difficult part is not the writing. It is deciding what "will happen" means when the markets contradict each other. Questions that exclude one another — the candidates in one race, the possible sizes of one rate decision — are resolved together rather than separately, so exactly one outcome survives from each contest. Without that step the paper would cheerfully report three different results for the same election.

Where the same question trades on more than one venue, the prices are pooled and weighted by how much money stands behind them, because a deep book is better evidence than a thin one.

Confidence is printed rather than stated. A claim the markets treat as near-certain is set in solid ink; a claim closer to a coin flip is set in the blue a printer uses to proof a page before committing it to the press. Every story carries a bar showing how far ahead the reported outcome ran and what came second, and the line where a reporter's name would sit carries the certainty instead.

Prices move. When the odds beneath a published article change enough to matter, the article is not quietly rewritten — a dated revision is added to it saying what changed. The paper is allowed to have been wrong about a future that had not happened yet.

None of this is a forecast worth acting on, and none of it is a record of anything. It is a reading of a market, written as fiction, in the only tense that makes the reading legible.