🧮 How Box Office Numbers Are Actually Calculated

You've never read a real Sunday box office number. Every one you've seen was a forecast wearing a final score's clothes; Monday's count is the only one that's actually finished. That gap is why weekend totals shrink overnight. It's why a movie crowned "the winner" on Sunday can lose that crown by Tuesday, no re-release required. The machine producing all of this is smaller than the hype around it suggests, and it starts at the cash register.

Who actually counts your ticket?

Machines, mostly, and one company in particular: Comscore. The instant your card clears, the sale drops into the theater chain's point-of-sale system, straight into Comscore's Box Office Essentials feed, the data source underneath basically the whole industry. By Comscore's own description, that feed covers roughly 99% of US and Canada screens and about 95% of the global box office. Box Office Mojo and The Numbers don't run their own counts. They republish that same pipe under their own mastheads, and when we covered Toy Story 5 crossing $400 million, the number underneath it came from the same place. You've been reading one company's data under a dozen bylines for years.

Those two percentages aren't the same kind of complete, either, and the gap between them is worth sitting with. Domestic coverage runs about four points tighter than global (99% against 95%), which is why domestic totals firm up fast and worldwide ones don't. More on that below.

Why is Sunday's number just a guess?

Because Sunday isn't over when studios publish it. Every weekend estimate goes out Sunday morning, which means the "weekend total" in a Sunday headline is confirmed Friday and Saturday money stapled to a guess about a day still in progress. Outlets run it anyway. Sunday is when this business likes to crown a winner, and nobody's in a hurry to file Tuesday's story on Tuesday.

Even the "confirmed" half is softer than the word implies. Friday is a blend: when a film runs Thursday-evening previews, trackers fold those grosses into Friday's reported total, which is most of why an opening Friday can dwarf the Saturday behind it. The Sunday estimate itself comes from the studio being estimated, not from Comscore. I stopped treating that as neutral bookkeeping a long time ago. Our own weekend results page prints the word "estimate" on Sundays. Studio press releases never do.

How do trackers predict a day that hasn't ended?

They lean on a ratio that barely moves: Sunday typically lands at 70-75% of Saturday's gross, a 25-30% drop that repeats film after film. The reason is unglamorous. School and work resume Monday, so Sunday-night showtimes thin out on schedule, every single week. Trackers call the number before dinner and rarely get burned for it.

Rarely is doing real work in that sentence. A five-point band is still a band, and five points of a big Saturday is real money riding on a rule of thumb. Holiday weekends bend the math further: a Monday off turns Sunday into a second Saturday, not a wind-down, which is exactly when a published Sunday number deserves your least trust of the year.

What actually changes when Monday's actuals land?

Monday swaps the projection for a real count, pulled from the tills of 90% or more of North American theaters. The rest, up to one theater in ten, gets extrapolated — modeled the way a census counts the house that never answers the door, filled in from what the neighbors reported. Even the number wearing the label "actual" carries a modeled sliver. The industry made peace with that. You don't have to.

Here's the part that doesn't get said enough: Comscore's ~99% screen-coverage figure and the 90%-plus theater-reporting figure for Monday actuals aren't measuring the same thing, and the roughly nine-point gap between them is the last bit of modeling still working itself out by Tuesday. The feed already sees almost every screen. It just doesn't have every theater's final tally in hand by Monday morning.

In practice, actuals land within a few percentage points of Sunday's estimate, either direction, and most weeks nobody notices. Some weeks it reshuffles the top of the chart, which is why the biggest opening weekends of all time get settled on Mondays, never Sundays.

No record is real until Monday agrees.

The Sunday estimate that crowned the wrong movie

June 2002 is the closest finish anyone points to when this argument comes up. Minority Report and Lilo & Stitch both finished the weekend convinced they'd won, and it took Monday's actuals to break the tie: $35.7 million for Minority Report, $35.3 million for Lilo & Stitch. Run the arithmetic and the real problem shows itself fast. The margin was $416,913, about 1.2% of the winner's gross, well inside the normal wobble between an estimate and an actual. A perfectly ordinary 3% swing on a $35.7 million weekend is about $1.07 million on its own — more than double the entire gap that decided the race. That Sunday chart wasn't wrong so much as undecided, printed like it was settled anyway.

The finish was a fluke. The margin of error wasn't.

Here's that cycle laid out, and the coverage numbers holding it up.

Why don't domestic and worldwide totals match?

Because only the domestic figure runs through the machinery above. US and Canada numbers are point-of-sale data and can look essentially final within days. Worldwide totals get assembled from dozens of national box office authorities, each running its own schedule, each converting its own currency into dollars before anyone adds it up. A film's domestic run can be fully counted while its worldwide number keeps climbing for weeks, simply because half its markets haven't opened yet.

Currency is the strangest part of the whole apparatus. A worldwide total can move overnight without a single new ticket sold, purely because an exchange rate shifted between two reports. When a headline compares two films' grosses without saying which figure it's using, that omission tells you plenty about how carefully the headline was built. We've laid out the full split in domestic vs. worldwide box office, explained.

How should you actually read this weekend's chart?

Date first, number second, every time. Every tracker we publish, including the 2026 box office chart ranking every release, carries an as-of stamp for a reason: the same page tells a different story on Sunday than it does on Tuesday. One version is a projection wearing a result's clothes. The other is the result.

The caution runs both directions. A soft Sunday estimate is exactly how a movie gets branded a flop a full day before its own opening weekend is even counted, and the real anatomy of a flop is slower and stranger than any single weekend figure, a subject we've picked apart in why movies flop at the box office. Verifying any of this costs nothing, which might be my favorite fact in the whole system: actuals sit in public on every major tracker by Tuesday, for free.

It happens to be true right now: this update posted on a Sunday, July 19, which means every weekend figure in circulation as you read this sentence is, technically, still a forecast.

Next weekend gets the same treatment, and so will the one after that, all the way through Spider-Man: Brand New Day's opening estimates once that Sunday number lands. Trust it exactly as much as this piece just taught you to, no more.

Methodology

This piece leans on Comscore's own account of its Box Office Essentials system, plus Slate's breakdown of how Hollywood box-office estimates work, whose mechanics have held up since it ran two decades ago. The Minority Report vs. Lilo & Stitch finish comes from HowStuffWorks' explainer on predicting weekend box office, and the Thursday-previews-into-Friday convention follows The Numbers' stated practice. We cite primary tracking methodology rather than restate numbers we haven't checked ourselves.

Photo: "Box office of Vue Cinema Longwell Green, Bristol" by Captain Galaxy, licensed CC BY 4.0 via Wikimedia Commons.