The Hollywood Originality Report

The original screenplay has been in continuous retreat for three decades

This is the chart the "Hollywood is out of ideas" argument has been missing. Two lines, 31 years, one direction of travel.

In 1995, 18 of the 50 highest grossing films in North America were original screenplays. In 2025 the figure was 5. Films built on pre-existing intellectual property, meaning sequels, remakes, and adaptations of books, comics, games, toys, and television, moved from 27 of 50 to 44 of 50 across the same period.

The decline is not a recent shock or a pandemic artifact. Fit a straight line to the original film share across all 31 years and it falls at 1.10 percentage points per year, with a correlation coefficient of negative 0.815 and an r squared of 0.665. In plain terms: two thirds of the year to year variation in Hollywood originality is explained by nothing more complicated than the passage of time.

"Hollywood is out of ideas" is the wrong sentence. The right one is that Hollywood stopped betting on them.

The single most original year in the dataset is 2000, when 28 of the top 50 films were original screenplays, including Gladiator, Cast Away, and Erin Brockovich. The least original years are 2021 and 2025, tied at 5 apiece. That is a swing of 23 slots inside a single generation.

Counting films understates it. Follow the money and the floor drops out

A slot in the top 50 is not a unit of success. The 50th film of the year earns a fraction of the first. When you weight by box office dollars instead of headcount, the collapse is far steeper.

Original films earned 34.2 percent of all money made by the top 50 in 1995. In 2025 they earned 10.9 percent. Sequels went the other way, from 17.1 percent to 52.9 percent, and peaked at 73.7 percent in 2021. The two lines crossed in 2001 and have never crossed back.

The dollar trend is tighter than the headcount trend, which matters for anyone arguing about causation. Revenue share for originals falls at 1.33 percentage points per year with a correlation of negative 0.891, a stronger relationship than the film count produces. Whatever is happening is happening to the money first.

2021 is the true floor, not 2025

The worst year for original films in the entire dataset by revenue share is 2021, when originals took 7.9 percent of top 50 money and sequels took 73.7 percent. Both are records. Spider-Man: No Way Home alone earned $573.0 million, more than every original film in that year's top 50 combined, and only one original reached the top 10 at all.

The 2001 crossover

The two lines cross exactly once, in 2001, the year Harry Potter and the Sorcerer's Stone, The Fellowship of the Ring, and Shrek all landed inside twelve months. Sequels took 21.6 percent of top 50 revenue that year against 34.0 percent for originals, and the following year sequels moved ahead for good. Twenty four consecutive years later they have not surrendered the lead for a single calendar year.

Originals did not just lose ground. They vanished from the top of the chart

The most legible version of this trend is not a percentage. It is a count of how many original films cracked the annual top 10, and how often that count hit zero.

In 1998, seven of the ten biggest films of the year were original screenplays. Between 2017 and 2020 the count was zero in four consecutive years, then zero again in 2022, 2023, and 2024. Nine years in the dataset produced no original film in the top 10 at all, and eight of those nine happened after 2010.

The 2025 top 10, the least original in the dataset

One original film in the top 10. Four remakes, the highest concentration of remakes in a top 10 in the entire 31 year dataset.

Since 2015, only five original films have reached the annual top 10 at all: Inside Out at number 4 in 2015, The Secret Life of Pets at number 4 and Zootopia at number 7 in 2016, Free Guy at number 10 in 2021, and Sinners at number 7 in 2025. Three of the five are animated. The eleven years from 1995 to 2005 produced 42 original films in the top 10, an average of 3.8 per year. The eleven years from 2015 to 2025 produced 5, an average of 0.45. That is a decline of 88 percent in the rate at which an original screenplay becomes one of the ten biggest films of its year.

Four distinct eras, and the break happens in 2010, not 2020

Grouping the years into blocks removes the single year noise and exposes where the structural shift actually landed.

The late 1990s and the 2000s look broadly similar to each other: originals holding somewhere between 37 and 43 percent of the top 50, pre-existing IP hovering just above half. The regime change arrives in the 2010 to 2014 block, where originals drop nine points in a single five year step and never recover. By 2015 to 2019 the original share has fallen to 18.4 percent and sequels alone account for half the chart.

It is tempting to blame the pandemic. The data does not support that. The 2020 to 2024 average for originals is 19.6 percent, marginally higher than the 18.4 percent recorded in the five pre-pandemic years. The collapse was fully complete before anyone had heard of a lockdown. What 2020 actually produced was a temporary spike, because tentpole sequels were pulled from the calendar and mid budget originals like Tenet, Knives Out, and Parasite backfilled the chart.

The sequel count more than quadrupled

Six sequels made the 1995 top 50. Twenty eight made the 2025 top 50, and 29 made 2021, the record. More than half of every commercially significant film in a modern year is now a continuation of a story the audience has already paid to see.

Sequels are not just more common. They are structurally bigger

Across all 1,550 films, the average sequel in the top 50 earned $162.9 million domestically. The average original earned $99.0 million. Medians confirm it is not a handful of outliers doing the work: $121.4 million against $81.4 million. Remakes averaged $127.6 million, adaptations $104.3 million, and true event films $97.3 million.

That gap is the entire economic explanation for the trend in this report. A studio choosing between a sequel and an original is not choosing between two ideas. It is choosing between two distributions, and one of them has a fatter right tail and a much higher floor. The sequel premium peaked in the early 2020s, when the average sequel out grossed the average original by a factor of 2.76.

The counterintuitive finding: the chart got more crowded, not more consolidated

The intuitive story is that a few enormous franchises ate the box office. The data says something more interesting and slightly worse.

The Marvel Cinematic Universe is in a category of its own: 38 appearances in the top 50 across 18 years and $13.06 billion in combined domestic gross, more than double the next closest franchise. But the number of distinct franchises appearing in the top 50 each year rose from 21 in 1995 to 36 in 2025.

That rules out the simple consolidation story. Hollywood did not narrow to five brands. It multiplied the number of brands competing for a fixed 50 slots, and the thing that got displaced was the film with no brand at all. In 2017 the top 50 contained 37 distinct franchises, an all time high, and just 9 original films.

What this data does not prove

A benchmark is only useful if it is honest about its edges. Here is every reasonable objection to the findings above, and what the numbers say back.

It does not prove fewer original films are being made

This dataset only observes the top 50 grossing films per year. It cannot see production volume. An era could produce more original films than ever and still show a falling original share of the top 50 if those films are smaller. Independent evidence suggests exactly that: the mid budget theatrical original has largely migrated to streaming. What this report measures is commercial prominence, not output.

It does not prove originals are less profitable

Gross revenue is not profit. Originals carry lower production and marketing budgets on average, so a $99 million original and a $163 million sequel are not automatically ranked by return on capital. The dataset contains no budget field. Anyone claiming originals are unprofitable needs a different dataset than this one.

It does not prove quality declined

Originality is a provenance measure, not a quality measure. The Dark Knight, Mad Max: Fury Road, and Top Gun: Maverick are all classified here as sequels. Toy Story 3 is a sequel. Nothing in this report says a sequel is a worse film, and the data cannot support that claim in either direction.

The classifications involve judgment

Roughly 40 films in the dataset are genuinely arguable. Barbie is classified as an adaptation because the property predates the film. Alice in Wonderland (2010) is classified as a remake rather than an adaptation because it re-versions an existing Disney film as much as the Carroll novels. 1917 is classified as based on true events despite a fictional plot. Every borderline call is flagged in the notes field of the underlying dataset and the priority rules are documented in the methodology below.

The originals that beat the trend anyway

Twelve films that arrived with no franchise, no source novel, and no prior audience, and finished the year among the biggest releases in North America.

Two patterns stand out. The first is animation. Six of the twelve highest grossing originals since 1995 are animated features, and five of those six come from Pixar or Illumination, studios whose brand does the work a franchise normally does. The audience is not arriving for a known story, it is arriving for a known studio.

The second is that the live action original blockbuster has not died, it has become rare and event shaped. Inception in 2010, Gravity in 2013, and Sinners in 2025 all cleared $250 million domestically with no pre-existing IP. Sinners is the most recent proof that the ceiling still exists. It finished 2025 at number 7 with $279.7 million, out grossing four films in that year's top 10 that arrived with decades of brand recognition attached, including Avatar: Fire and Ash.

Original films did not lose the audience. They lost the release calendar.

Straight answers to the questions people actually search

The complete year by year dataset

Counts are out of 50 films per year. Total IP is sequels plus remakes plus adaptations. Original dollar share is the percentage of that year's top 50 box office revenue earned by original screenplays. Highlighted rows are the most original year on record, the lowest dollar share year, and the most recent year.

Methodology, definitions, and limitations

Scope and sampling

The dataset comprises the 50 highest grossing films at the United States and Canada theatrical box office in each calendar year from 1995 to 2025 inclusive. That is 31 years and 1,550 films, with no gaps, no duplicate ranks, and exactly 50 entries per year. Ranking and revenue figures are taken from Box Office Mojo calendar year charts, which measure money earned inside the calendar year rather than lifetime gross. A December release therefore appears in two years if it earned materially in both. Re-releases are excluded.

The five categories

Original. An original screenplay with no pre-existing narrative source and no prior film in the same continuity. Toy Story, Inception, Get Out, Sinners. A film that launches a franchise is still original.
Sequel. Continues the story, characters, or continuity of an earlier film. Includes prequels, midquels, spin offs that share continuity, and films within an established shared universe such as the Marvel Cinematic Universe.
Remake. A new screen version of a story that has already been filmed, including live action versions of animated features and continuity resetting reboots. Lilo & Stitch (2025), Superman (2025), The Departed.
Adaptation. First screen version of a story or property from another medium: novels, comics, video games, stage musicals, toys, television. Harry Potter and the Sorcerer's Stone, Spider-Man (2002), A Minecraft Movie, Barbie.
Based on true events. A dramatization of real events or real people. Kept separate from both original and IP. Apollo 13, Titanic, Oppenheimer.

Priority rules for overlapping cases

Many films qualify for more than one category. A single primary category is assigned using a fixed priority order: sequel beats everything, remake beats adaptation, and adaptation beats true events where the source is a specific published work rather than the underlying history. Franchise membership is recorded separately and never overrides the originality classification, which is why Toy Story appears as a franchise film in 1995 despite being an original screenplay.

Known limitations

Four limitations should travel with any citation of this report. First, domestic gross only: worldwide performance, and specifically the Chinese market, would change the composition of a global top 50. Second, revenue not profit: the dataset carries no budget or marketing spend. Third, top 50 only: the report measures commercial prominence and is blind to production volume outside the top 50, including films released direct to streaming. Fourth, classification judgment: roughly 40 of 1,550 films sit on a genuine boundary, all flagged in the source notes.

A note on why shares, not dollars

Total top 50 revenue grew from $3.32 billion in 1995 to $8.54 billion in 2018 before falling to $6.80 billion in 2025. Much of that growth is ticket price inflation and premium format surcharges rather than attendance. Box office figures are nominal dollars and are not adjusted for inflation. Comparing the share of each year's top 50 that originals captured removes the problem entirely, because both numerator and denominator sit inside the same year's pricing environment.

Cite this report

This report exists because the "Hollywood is out of ideas" argument had no source. Use it, quote it, argue with it. Attribution and a link are the only conditions.