Windowing, vindicated
The Covid experiment ended. The data agreed with the received view.
For roughly four years the industry ran a natural experiment it never asked for. Theaters closed. Windows collapsed. Studios pushed films onto premium VOD on day one, then onto their own streaming services, then back into theaters with a 17-day hall pass that was supposed to be the future. The trades wrote eulogies for the 90-day window with the conviction of people who had never sat through a P&L.
The received view was unmoved throughout. The argument never wavered: windowing exists because the math of windowing exists, not because the marquee names of 1997 demanded it. Each viewer carries a willingness-to-pay that is highest in a theater on opening weekend and lowest on a basic streaming tier two years later. A rational distributor exhausts the top tier before tapping the next. Anything else is leaving money on the floor and calling it innovation.
“The data that’s been compiled shows that you need to exhaust the highest per viewer revenue, then move to the next tier, then the next, then the next; just like it used to be.” [2026-03-13]
That is the whole argument in one sentence. Notice what it does not say. It does not say theaters are sacred. It does not say streaming is bad. It says: revenue tiers are real, they are ordered, and skipping the top one to chase a subscriber-acquisition narrative is a transfer of value from the studio’s owners to the streamer’s shareholders.
What the Covid window actually proved
The 17-day window did not save anyone. Day-and-date did not save anyone. The films that the day-and-date crowd pointed to as proof of concept either had no theatrical comp (so the counterfactual was unfalsifiable) or were openly cannibalized. The Mortal Kombat 2021 release, opening to $23M because HBO Max+ was running it the same morning, became a verb in this corpus. To be Kilar’d is to have your theatrical gross rented out from under you to feed someone else’s subscriber chart.
The 2026 reboot of the same title opened past that 2021 number on the same IP, against a marketplace with more competition, with no streaming concurrency. The delta is the answer.
The April 2026 capitulation
In April 2026 Universal walked back to a 45-day theatrical window. The trades read this as “Universal extends window.” It is more precise to read it as: the studio that ran the most aggressive shortened-window experiment, with the most data, looked at its own internal numbers and quietly reverted.
The trade press called it a return. It reads better as the natural experiment ending.
What this means operationally
If you operate at a specialty or indie level: the windowing question is no longer a strategic toggle. It is a default. The default is theatrical-first, then PVOD at the right price point, then transactional, then ad-supported. Films that skip a tier should have a specific, articulable reason, not a vibe, not a streamer’s term sheet that buys you out of the question.
The corollary is harder to swallow. If a streamer offers to “give you a theatrical release” as part of a deal, ask what they mean by theatrical. The honest answer is rarely the one you want.
“Theatrical is really hard, it’s kind of like the difference between saying ‘let’s have a High Speed train from LA to SF’ and actually doing it.” [2026-05-01]
The window exists because the work to make a top-tier theatrical release exists. Removing the window does not remove the work. It just removes the revenue.