It is wide enough that ordinary post-earnings noise stays inside it and narrow enough that a genuine repricing crosses it. A fixed width keeps the record comparable across names with different volatility; the record is a diary of one rule, not a forecast.

Why the third close and not the first?

The first close after a print is dominated by the after-hours reaction and the opening auction. By the third close, two full sessions have tested that reaction. Oracle’s September 2026 print jumped 4% after hours and was scored sold three sessions later.

Is “sold” a judgment on the company?

No. It records that the market paid less for the shares after the report than before, by more than 3%. A company can report its best quarter and score sold if the market is repricing its discount rate or its scarcity premium.

Glossary term

Print Record

Closelook’s scored log of earnings reports: each print gets a window of ±3% around the last close before the report and is scored on the third close after it — above the window is “paid”, below is “sold”, inside is “flat”. It measures whether beats get rewarded, not whether estimates were beaten.

AI-generated — produced automatically by Closelook’s systems under this site’s editorial policy.

What it means

An earnings print is the quarterly report itself. The print record does not judge the report; it records what the market did with it under one fixed rule. The entry is the closing price before the report. The window is 3% either side of the entry. The score is taken at the third close after the print — long enough for the after-hours reaction to be tested by two full sessions, short enough to stay about the report. Above the top of the window the print was paid; below the bottom it was sold; inside it was flat.

Because the rule never changes, the record can be read across names, sectors and seasons: how many beats were paid this quarter, which sectors sell beats, whether the market is rewarding growth or punishing capex.

Why it matters for the AI trade

The AI trade in 2026 became a trade about whether beats get paid. Oracle reported the larger beat of its September quarter — revenue $19.3 billion, cloud infrastructure up 121%, a $664 billion backlog — jumped after hours, and closed the third session at $144.79, below its $148.35 sold line from a $152.94 entry. Adobe, reporting the same night, closed at $265.60, above its $256.29 paid line. The same beat, scored sold on the seller of compute and paid on the buyer, is the kind of fact the record exists to establish.

How Closelook uses it

Cards for every scored print live under Signals → Print Record, with the entry, the window, the closes and the verdict; each stock page carries the name’s own history. Cards are published on the day they are produced. The post-earnings announcement drift entry covers the academic version of the same question over longer horizons; the earnings print entry covers the report itself.

Common questions

Why a 3% window?: It is wide enough that ordinary post-earnings noise stays inside it and narrow enough that a genuine repricing crosses it. A fixed width keeps the record comparable across names with different volatility; the record is a diary of one rule, not a forecast.
Why the third close and not the first?: The first close after a print is dominated by the after-hours reaction and the opening auction. By the third close, two full sessions have tested that reaction. Oracle’s September 2026 print jumped 4% after hours and was scored sold three sessions later.
Is “sold” a judgment on the company?: No. It records that the market paid less for the shares after the report than before, by more than 3%. A company can report its best quarter and score sold if the market is repricing its discount rate or its scarcity premium.