Welcome back to the Monday Roundup!

The Monday Roundup is a bonus edition featuring data-driven culture stories, curated datasets, and recommended reading. Given the strong response to this format, I’ll be publishing it weekly for the foreseeable future.

Your regular Wednesday essay stays free for everyone—this is strictly additive.

Today’s edition has six new data stories. Free readers get the first two; paid subscribers get the full thing, plus datasets and recommended reading. Enjoy!

In Today’s Roundup:

• Original Data Stories: A James Bond drought, the end of music’s “loudness war,” the importance of opening day box office, the Falklands War revisited, a college for livestreamers, and generational adoption of closed captions.

• Recommended Reads: Notable data journalism and pop culture reads from around the web.

• Unique Datasets: Ten publicly available datasets, covering everything from Rotten Tomatoes scores to Big Mac prices to Spotify song composition attributes.

Trend 1: A James Bond Drought

James Bond is alive and well, despite the character’s absence from cinemas and the fact that we don’t even know who the next James Bond will be.

The latest installment of the 007 franchise is currently slated for 2028, which would make for a 7-year gap between that project and 2021’s No Time to Die—the longest Bond-related movie drought since the franchise’s introduction in 1963.

If you’re a bona fide Bond-head, don’t fret. Progress is being made. In June of last year, Denis Villeneuve—the filmmaker behind the Dune films, Sicario, and Arrival—was announced as the franchise’s new director.

Next in line: the world needs a new Bond. Prediction markets currently favor Callum Turner, a talented actor who is perhaps most famous for being married to Dua Lipa.

After all this time, it’s shocking that there is no consensus frontrunner for Bond succession, yet my hot take is that this is actually an asset for the franchise.

The past two decades have seen a deluge of films oriented around intellectual property. And perhaps the greatest sin committed by top franchises like Marvel and DC during this period has been an error of abundance. This abundance comes in the form of too many movies, too many toys, and too many streaming shows.

James Bond thrives on scarcity, meaning a 7-year hiatus may be the best thing that’s ever happened to the character.

Trend 2: Is Music Getting Softer?

Spotify has created its fair share of problems throughout the music industry, but it’s also solved a few as well. One such example is the “loudness war” of the early 21st century, a decades-long arms race among record producers to make their songs louder than the competition. This phenomenon reads like something invented by a local news anchor or a coalition of Concerned Parents, but the conflict—and its effect on recorded music—was very real.

If you want to see this decades-long struggle encapsulated in a single graphic, look no further than the average decibel level of Billboard-charting music, which rose from the late 20th century all the way up until the early 2010s.

Beginning in the late 1980s, producers increasingly relied on audio compression and limiting to make songs louder than the competition. The push toward music’s upper limit was driven by fourth-grader logic: the louder the track, the more likely it was to stand out on the radio or cut through the cacophony of background noise.

And which artists emerged as the winners of this proverbial war? A collection of musicians who saw commercial success during peak-music-loudness in the late 2000s.

The Red Hot Chili Peppers are my all-time favorite band, though perhaps this was the product of a psyop by Big Loudness.

The march toward maximal noise began reversing in the 2010s, as streaming services like Spotify adopted volume normalization, which brings songs to a consistent playback level and strips excessive compression of its competitive advantage.

Today, “loud music” is simply the result of a listener turning up the volume too high.

Trend 3: Christopher Nolan, and Why Blockbusters Live or Die on Opening Day

Against all odds, Christopher Nolan has turned a book from my 10th-grade summer reading list into a billion-dollar blockbuster. Anticipation for The Odyssey reached such a fever pitch that Universal opened ticket sales a full year before release. The strategy paid off: the film earned roughly $51 million at the U.S. box office on opening day alone—a staggering haul that reflects a broader shift in how modern blockbusters are consumed.

Over the past 25 years, a film’s theatrical life cycle has accelerated dramatically, with audiences consuming new movies faster than ever.

Read more

Intro: #StayInLA

The past decade has not been kind to Los Angeles’ entertainment industry. The combined shock of the 2023 labor strikes and the 2025 wildfires intensified fears that film and television production is migrating overseas, threatening a fragile ecosystem of below-the-line workers based in Hollywood.

In the days after the wildfires, filmmaker Sarah Smith and writer Alexandra Pechman organized a WhatsApp group of showrunners and industry workers, drafted a petition, and launched the grassroots #StayInLA campaign. What began as an emergency effort to help Los Angeles rebuild grew into a broader political movement, drawing more than 20,000 petition signatures, endorsements from prominent actors and filmmakers, and even the fleeting attention of Donald Trump—an actor himself, best known for his performance in Home Alone 2: Lost in New York.

Having worked in the entertainment industry and lived in Los Angeles, I have long wanted to investigate this often-vague anxiety: to understand exactly where these productions are migrating and quantify the extent of Hollywood’s flight from, well, Hollywood itself.

So today, we’ll examine Hollywood’s great production migration, explore how these logistical shifts have reshaped the filmmaking process, and investigate whether outsourcing production has any measurable impact on the quality of the movies themselves.

How Hollywood Stopped Making Movies in Hollywood

The American film industry began concentrating in Los Angeles in the 1910s, drawn to the city’s year-round sunshine, varied landscapes, inexpensive land, and distance from Thomas Edison’s patent enforcement on the East Coast. By the 1920s, Hollywood had become the global center of filmmaking, with studios, skilled workers, and supporting businesses clustered together in a self-renewing ecosystem.

For decades, most American films were produced on Los Angeles backlots, where studios provided filmmakers access to turnkey sets and skilled crews. This mechanized approach kept costs low and enabled directors to crank out four to six movies a year. During this era, Hollywood operated much like Henry Ford’s Model T assembly line; studios were maniacally focused on speed and frugality.

LA’s stranglehold over movie production would ultimately be tested by two filmmaking trends:

1. The rise of the blockbuster: In the 1970s, box-office smashes like The Godfather and Jaws demonstrated the commercial potential of event cinema. By the 1980s, corporatized studios like Paramount-Gulf-Western were actively manufacturing blockbusters, pouring ever-growing budgets into spectacle-driven stories. Films like Predator, First Blood, and Raiders of the Lost Ark were shot on location in jungles, deserts, and other far-flung settings, as exotic locales amplified their sense of scale.

2. The rise of IP and tax credits: The 2000s and 2010s brought a surge in big-budget filmmaking built around familiar IP. Marvel’s vaunted Cinematic Universe and other franchises required considerable financial investment, leading studios to offset costs by relocating production to incentive-rich cities such as Vancouver and Atlanta.

The result of these production strategies was a slow-building migration out of Los Angeles and California altogether.

So where did all the movies go? Hollywood’s production exodus has benefited a small group of filmmaking hubs offering some combination of established infrastructure and financial incentives. London, for example, boasts world-class facilities such as Pinewood Studios, while cities like Atlanta and Vancouver provide generous tax credits and rebates that reimburse studios for their local spending.

The savings these destinations offer are substantial, making them an obvious choice for a superhero movie whose story could unfold almost anywhere—be it a generic cityscape or somewhere in outer space. For a major blockbuster, tax incentives equate to tens of millions of dollars, freeing up money for name-brand actors, elaborate set pieces, and unremarkable CGI effects.

As such, the larger a film’s production budget, the more likely it is to be filmed entirely outside the United States.

In some cases, this globe-trotting appears directly on screen, like when Mission: Impossible stages an elaborate car chase through Rome. In other cases, location choice is driven entirely by financial considerations.

The latter trend has created a widening gap between where stories are set and where they are filmed, as productions increasingly follow tax incentives and lower costs rather than narrative necessity.

Look closely, and you’ll spot Toronto’s skyline standing in for New York, or Prague’s streets masquerading as nearly every other European city. With each passing year, filming locations have become untethered from the places depicted on screen.

This disconnect has become a persistent grievance among film critics, particularly when it comes to streaming movies shot far from their narrative setting. Nearly every week, the hosts of my favorite movie podcast complain about some disposable Netflix rom-com ostensibly set in New York but unmistakably filmed in New Orleans. After hearing this criticism ad nauseam, I began to wonder whether shooting on location actually has a discernible impact on a film’s quality—at least in the eyes of viewers.

As a cinema purist, I believe movies should be filmed where they are set, even when that setting is outer space. Most audiences, however, either cannot tell the difference or simply do not care.

Perhaps I have gone to great lengths to prove something studio executives already know: the average viewer is not scrutinizing every frame for telltale glimpses of the Toronto skyline. More likely, they are glancing at Instagram while the film’s protagonists laboriously explain the plot for the narrative comprehension of multitasking viewership—at which point the authenticity of the location becomes irrelevant.

Final Thoughts: Cost Disease

My junior year of college, I scored an internship at Conan O’Brien’s late-night talk show. So, like Miley Cyrus circa 2009, I hopped off the plane at LAX with a dream and an unpaid internship.

Within a week, I’d discovered three things:

1. Entertainment internships are often glorified daycare for the children of industry executives: I could be classified as many things, but “nepo baby” is not one of them—which made me an exception among my intern class.

2. There was no work to do: The show had far more interns than meaningful tasks. I never even earned the privilege of fetching coffee.

3. The Warner Bros. backlot was eerily empty: I had grown up on movies like The Player and Singin’ in the Rain, where Hollywood appeared as a bustling factory of overlapping productions squeezed onto compact studio lots. Instead, I found emptiness.

This combination of things led to a bizarre summer where me and a posse of underutilized unpaid interns wandered a near-vacant Warner Bros. studio as if it were the Backrooms. This meant hanging out in the Central Perk set from Friends or cart-racing other interns from The Ellen DeGeneres Show on the street where they filmed Gilmore Girls.

My lasting impression was that Hollywood had become a collection of offices where white-collar workers decided how the sausage got made, while remarkably little sausage was actually being made there.

It was also during this summer that I learned of the hazards of cost disease.

Cost disease is an economic phenomenon in which labor-intensive industries with limited productivity gains—healthcare, education, live theater, filmmaking—become steadily more expensive over time. The quintessential example of this is Broadway, where the cost of living in New York City increases, leading wages to rise, but there remains a fixed set of theaters for Broadway shows to be staged, thereby limiting revenue growth. The end result is economic unsustainability, as these productions either curtail spending by employing fewer people or pass higher costs along to consumers through ticket pricing, often both.

My summer in LA was cost disease in miniature. I was paid exactly $0 to relocate to Los Angeles, where I then paid thousands of dollars to live in America’s third most expensive city. And this was the very bottom rung of entertainment’s corporate ladder.

Higher up the corporate hierarchy, cost disease takes different forms. For entry-level assistants, it means accepting low wages and paying high rent for the possibility—however remote—of eventually joining the executive class. For camera crews and VFX artists, it means becoming part of a traveling circus that chases tax incentives to Atlanta, Prague, and Vancouver. For studio executives, it means making do with smaller crews, outsourcing production overseas, and, eventually, using AI to replace some human labor.

If this has all drifted into bummertown, it’s because many of Hollywood’s problems can be traced back to a single economic theory of everything: the rent is too damn high. From there, everything else starts to unravel.

The end result is a peculiar version of Hollywood: a place increasingly devoid of physical production, populated instead by executives, AI art, and unpaid interns drag-racing golf carts through empty studio streets.

Enjoyed the analysis? Support Stat Significant and unlock the data behind the insights.

Become a paying subscriber to get bonus graphics, interactive dashboards, downloadable datasets, and to support independent journalism.

Paying subscribers get:

• Bonus Data Stories: Full access to the Monday Roundup, a bonus edition with extra culture data stories, curated datasets, and recommended reads.

• The Public Dataset Directory (Updated Weekly): A curated directory of public datasets from around the web.

• Interactive Data Hubs: Interactive dashboards across Streaming TV, Movies, Music, Search Trends, Wikipedia, and more.

Upgrade to Paid

*Note: For best results, open this link in a web browser while logged in to your Substack account.

Want to promote your data tool or media publication to Stat Significant’s 26,200 readers? Email daniel@statsignificant.com

Need help with a data problem? Book a free data consultation

Connect with me on LinkedIn if you want to chat about a data project.

Like movies? Follow me on Letterboxd

Welcome back to the Monday Roundup!

The Monday Roundup is a bonus edition featuring data-driven culture stories, curated datasets, and recommended reading. Given the strong response to this format, I’ll be publishing it weekly for the foreseeable future.

Your regular Wednesday essay stays free for everyone—this is strictly additive.

Today’s edition has six new data stories. Free readers get the first two; paid subscribers get the full thing, plus datasets and recommended reading. Enjoy!

In Today’s Roundup:

• Original Data Stories: YouTube’s surprising gaming dominance, the IMAX boom, why hit tracks have a small army of songwriters, whether Disney’s live-action remake machine is losing steam, what Love Island reveals about reality TV’s low hit rate, and how Count Binface became Britain’s viral political star.

• Recommended Reads: Notable data journalism and pop culture reads from around the web.

• Unique Datasets: Ten publicly available datasets, covering everything from television episode plots to Pitchfork album reviews to eight decades of New York Times articles.

Trend 1: YouTube Is for Gamers

I recently built a Google Trends equivalent for YouTube to quantify the terms that most frequently appear in the site’s top videos. My initial guess was that this list would feature newly released trailers, music videos, and sports highlights. In essence, I was expecting data reflective of my YouTube homepage. But, as is always the case, my predictions never align with the zeitgeist.

In fact, when I reviewed my initial output, I barely recognized 50% of the top terms. And thus I found a new way to feel old. That’s because I am not, nor have I ever been, a gamer, and the most frequently used terms on YouTube revolve around gaming.

At first, I assumed this to be a mistake. But when I looked at the site’s top trending videos over the last few months, gaming content comprised anywhere between 55% and 70% of all entries.

In short, the most popular content on YouTube involves another person playing video games or hearing their hot take about a video game.

In fact, YouTube is the top destination for gaming discovery by orders of magnitude, above dedicated gaming platforms like Twitch or Discord.

To many, this fact is painfully obvious. To me, this was the find of the century.

The takeaway—and I pretty much believe this is always the takeaway when it comes to YouTube—is that the platform can be anything for everyone.

YouTube is the top place to go for gaming content, just as it’s the foremost destination for movie trailers, sports highlights, music videos, podcasts, late-night talk show clips, and videos of celebrities eating chicken wings.

Trend 2: IMAX Is the Present and Future of Moviegoing

Running a movie theater is brutal. You are essentially a middleman with little control over the product you sell. Your job is to create an experience compelling enough to lure people away from their living rooms—all while containing costs, relying on relatively low-wage labor, and competing with every other conceivable form of entertainment. Given those constraints, how can theaters sustainably grow when they do not control the movies themselves?

One of the few levers exhibitors do control is the viewing experience: the size of the screen, the quality of the sound, and the spectacle of the presentation. Over the past decade, theater chains have increasingly invested in premium formats such as IMAX, Dolby Cinema, and 4DX, offering audiences something they cannot replicate at home. And if premium formats had their own Super Bowl, World Cup, or Paris Fashion Week, it might look something like the recent release of Christopher Nolan’s The Odyssey.

Since the pandemic, premium formats have become an increasingly important selling point for moviegoing. They are now on track to account for roughly 17% of U.S. ticket sales this year, up from 10% in 2019.

These formats also command significantly higher ticket prices, allowing exhibitors to grow revenue even as overall attendance declines.

The greatest obstacle to this strategy is scarcity. There are only so many IMAX screens, and building one requires substantial upfront investment—an especially difficult proposition for debt-laden theater chains and independent exhibitors operating on thin margins.

Still, this appears to be the direction moviegoing is headed. Do not be surprised if, 50 years from now, every theater is effectively an IMAX theater, forcing exhibitors to invent an entirely new premium tier to promote. They’ll call it Double IMAX, Triple IMAX, or Double-Triple IMAX 360+.

Trend 3: How Many Writers Does It Take to Make a Hit Song?

The 1970s saw the rise of the singer-songwriter, a genre headlined by the likes of James Taylor, Carole King, and Cat Stevens. If you wanted a nice lullaby to sing to your child, these were your go-to artists.

The process behind these hit songs was decidedly straightforward: artists wrote and performed their own tracks, and the end result was a unique product of that individual author-performer.

Today, very few, if any, songs have a single songwriter.

Read more

Intro: Project Popcorn

Few decisions have reshaped moviegoing more than the stupidly named “Project Popcorn.” For those unfamiliar with this deeply lame corporate initiative, Project Popcorn was WarnerMedia’s 2021 plan to release every Warner Bros. film on HBO Max the same day it arrived in theaters. It was a short-term decision with one obvious goal: juice streaming numbers.

The experiment began with Wonder Woman 1984, then expanded to an entire year’s worth of releases. Over the next twelve months, Warner Bros. jettisoned its most anticipated projects to streaming: Tom & Jerry, Godzilla vs. Kong, Mortal Kombat, The Suicide Squad, Dune, King Richard, The Matrix Resurrections, and many more.

Inspired by Warner’s short-term thinking, other studios pursued similar streaming-first strategies, and the economics of Hollywood subsequently went haywire. Christopher Nolan left Warner Bros. for Universal, where he eventually made Oppenheimer and The Odyssey. Scarlett Johansson sued Disney, alleging that the day-and-date release of Black Widow violated her contract. All across Hollywood, studios were forced to compensate talent for the sudden erosion of once-lucrative back-end deals.

Yet Project Popcorn’s defining legacy was not any single lawsuit or bruised ego. Its real impact was resetting audience expectations about how soon a movie should arrive on streaming. Where there had once been a 5-month window between theatrical release and home viewing, audiences were suddenly told they could get top-shelf entertainment with no waiting period at all.

Studios eventually reversed course, restoring theatrical windows of 45, 60, or even 90 days. By then, however, viewer expectations had already shifted—perhaps forever.

So today, we’ll examine the relationship between theatrical moviegoing and streaming activity, estimate how much money Hollywood is leaving on the table, and explain why theaters remain the smartest way to launch a film.

Today’s newsletter is sponsored by Works in Progress

Works in Progress: A Guide to How the Future Actually Gets Built

Works in Progress is a print magazine that explores the breakthroughs—and bottlenecks—quietly shaping our world, from economics and urbanism to the frontiers of science and energy.

Its long-form, data-rich essays are read by people at Anthropic, OpenAI, MIT, and the Federal Reserve. Six beautifully designed print issues a year.

👉 Start reading Works in Progress

Why We Stopped Going to the Movies

When VHS and Betamax VCRs arrived in the late 1970s, Hollywood feared audiences would abandon theaters for at-home viewership. Universal and Disney even sued Sony, Betamax’s manufacturer, in an unsuccessful attempt to curb the technology’s adoption. By the mid-1980s, however, the industry had embraced VHS, and home video became a lucrative market worth roughly $10 billion a year.

The late 1990s brought DVDs, which offered enhanced picture quality, greater portability, and interactive bonus features. The format quickly supplanted videotape as the dominant mode of home entertainment, with annual DVD sales peaking at roughly $16 billion in 2005.

This dependable revenue stream enabled studios to take bigger risks. If a film underperformed in theaters, strong DVD sales eventually allowed it to turn a profit.

Based on a handful of case studies from the early 2000s, DVDs generated revenue equivalent to roughly 50% of a film’s international box office.

Crucially, VHS and DVD purchases did not cannibalize ticket sales because studios preserved a theatrical window of at least five months. By the time a film reached home video, it had often faded from public attention, prompting studios to launch an entirely different marketing campaign around its release on physical media. This windowing system trained consumers to treat the two formats as distinct purchases: a theatrical ticket paid for a communal, big-screen experience, while a cassette or disc allowed you to own the movie and rewatch it at home, albeit in lower quality.

That carefully managed distinction would be dismantled by technological disruption and the chaos of a global pandemic.

In January 2007, Netflix launched its streaming service, giving consumers instant access to a vast library of film and television for just $5 a month. Movie watching was no longer bound by physical constraints, and DVD sales plummeted as streaming subscriptions, digital rentals, and digital purchases grew into a market generating roughly $60 billion annually.

In 2020, the pandemic shuttered theaters across the country and forced Hollywood studios to collapse their traditional exclusivity windows, most notoriously through WarnerMedia’s “Project Popcorn.” Instead of waiting 180 days to watch a movie at home, audiences might wait just 14 days—or no time at all—upending decades of consumer expectations.

The dependable revenue once generated by DVD sales gave way first to premium video-on-demand, where viewers could rent or purchase films through platforms such as Amazon and Apple, and later to subscription streaming bundles such as Netflix. Once a movie was folded into a monthly subscription fee, however, its financial return became increasingly abstract.

Although premium video-on-demand is more convenient, data released by Universal suggests it generates only a quarter of the revenue of physical media sales.

A digital rental costs between $3 and $20, with a single transaction covering an entire household—effectively turning four theatrical tickets into one purchase. For many consumers, waiting for a home release is therefore cheaper and logistically simpler. Recent case studies based on online review activity suggest that many—perhaps most—films now find their largest audience after leaving theaters, first through paid digital release and later through subscription streaming. This pattern is not universal, however, and varies considerably from film to film.

That said, across dozens of data points, the broader trend is clear: a sizable cohort of consumers has been trained to wait for streaming. The result is a new rhythm of movie consumption: theatrical releases generate a strong, sustained wave of attention; paid digital releases produce a sharp but fleeting spike in activity; and subscription services deliver a smaller second uptick.

Perhaps most revealing is the first week of premium VOD, when online activity surges—evidence of pent-up demand among viewers who never planned to leave home.

Worse still, consumers appear to prefer this newfound bargain. Online film ratings vary sharply across the three release windows, producing a rather dispiriting finding for anyone who still believes in the magic of da’ movies: the less people pay to watch a film, the more they tend to like it.

One caveat: these findings come from a limited sample of films released over the past six months, so I’ll revisit the analysis as more data accumulates. Still, the broader pattern is intuitive: a mediocre movie can be a rip-off when it’s a $35 trip to the theater, yet feel like a quaint diversion when bundled into an existing Netflix subscription. Price, in other words, may shape not only whether people watch a film, but how generously they judge it.

And if, like me, you still prefer seeing movies in theaters—even when surrounded by legions of tweens scrolling TikTok—take comfort in this: Hollywood still has a financial incentive to preserve the theatrical experience.

When we compare online review activity across distribution models and streaming windows, we find that films originally released in theaters consistently generate more engagement across all viewership periods—throughout their theatrical runs, paid digital releases, and eventual availability on subscription platforms—than movies released exclusively on streaming. This suggests that theatrical exhibition still serves as a signal of quality, giving audiences a reason to seek out the film later.

Perhaps theatrical audiences are not merely customers to be monetized, but early evangelists who will build enthusiasm among a cohort of late adopters who will watch the movie via digital rental. Marketing for theatrical releases may therefore accomplish two things at once: attracting committed moviegoers while seeding demand among hesitant at-home audiences.

Final Thoughts: Time, Money, and Diapers

In entertainment trade reporting, theatrical moviegoing is often framed as a noble pursuit, while those who wait for a home release are portrayed as ungrateful philistines who systematically destroy the arts with every Netflix selection. I never fully subscribed to this belief, but I didn’t exactly reject it either. Every time I went to a movie theater, I would pat myself on the back for doing my part—and reward myself with Cookie Dough Bites. And then I had a kid.

Before my son was born, people would offer unsolicited advice that I’d immediately ignore. They’d say things like, “Nothing can prepare you for how much your life will change,” “Time becomes a luxury,” or “Your world will never be the same.” One person even told me they had missed all seven seasons of The West Wing because they were “too busy having kids.” That’s seven full seasons—22 episodes apiece! I disregarded every warning and well-meaning recollection, thinking to myself: “You don’t know me!” Then I’d move on with my day, convinced that I was somehow above conventional wisdom.

Well, all these people were right, and I am a dingus. There is now a growing catalog of movies that I may miss because I am too busy having kids—or, at the very least, I will not see them in theaters.

There was once a meaningful trade-off between staying home and going to the movies. Theaters offered something you could not replicate in your living room: high-definition video, better sound, and access to a film five months before it reached physical media. That trade-off has largely disappeared. I can now watch Hollywood’s best movies on my above-average television within weeks of their debut, spend less money, and never miss my son’s bedtime. Unfortunately, everybody wins—other than movie studios.

At this stage of my life, moviegoing comes with more friction than it once did—a fact Christopher Nolan chose to ignore when he made The Odyssey three hours long. Three hours! Where exactly am I supposed to find a five-hour block required to get to a theater, watch an adaptation of a book I was forced to read in high school, and make it home again? Of course I will go. Once a film major, always a film major. But I have responsibilities now: a kid to put to bed, charts to make for your entertainment, and diaper trash to empty. This new diaper-trash-oriented existence does not easily accommodate three-hour epics, 30 minutes of trailers, an AMC promo featuring Nicole Kidman, and a post-trailer car insurance commercial.

I don’t entirely understand how someone could miss seven full seasons of The West Wing, but it was also a different time—a time before DVR, before TiVo, before Netflix, and before Project Popcorn. Now people don’t really have to miss anything. All they have to do is wait two to four weeks.

Enjoyed the analysis? Support Stat Significant and unlock the data behind the insights.

Become a paying subscriber to get bonus graphics, interactive dashboards, downloadable datasets, and to support independent journalism.

Paying subscribers get:

• Bonus Data Stories: Full access to the Monday Roundup, a bonus edition with extra culture data stories, curated datasets, and recommended reads.

• The Public Dataset Directory (Updated Weekly): A curated directory of public datasets from around the web.

• Interactive Data Hubs: Interactive dashboards across Streaming TV, Movies, Music, Search Trends, Wikipedia, and more.

Upgrade to Paid

*Note: For best results, open this link in a web browser while logged in to your Substack account.

Want to promote your data tool or media publication to Stat Significant’s 26,200 readers? Email daniel@statsignificant.com

Need help with a data problem? Book a free data consultation

Connect with me on LinkedIn if you want to chat about a data project.

Like movies? Follow me on Letterboxd

Welcome back to the Monday Roundup

The Monday Roundup is a twice-monthly bonus edition featuring culture data stories, curated datasets, and recommended reading.

Your regular Wednesday essay stays free for everyone—this is strictly additive.

Today’s edition has six new data stories. Free readers get the first two; paid subscribers get the full thing, plus curated datasets and recommended reading. Enjoy!

In Today’s Roundup:

• Original Data Stories: Netflix is losing the streaming wars, Obsession’s crazy box office ROI, the Swift-Kelce wedding vs. Metallica, TikTok’s livestreaming campaign, Grand Theft Auto pre-sales, and Anna’s Archive.

• Recommended Reads: Notable data journalism and pop culture reads from around the web.

• Unique Datasets: Ten publicly available datasets, covering everything from YouTube trending videos to song lyrics to eighty years of weather data.

Trend 1: Netflix’s No-Good Year

In 2025, Netflix said goodbye to Stranger Things and Squid Game, two series emblematic of TV’s big-budget prestige era. And while correlation is not causation, I’m going to speculate that losing its two biggest shows didn’t help the platform.

For quantifiable proof of Netflix’s 2026 retrenchment, look no further than Nielsen’s streaming market share data, where Netflix has been slowly losing viewership since December of last year—which, go figure, is when Stranger Things finally ended.

Better yet, the two biggest streaming success stories in that same period are YouTube and the Roku Channel, which charge consumers precisely $0 a month.

The economics of the YouTube vs. Netflix streaming battle are pretty hilarious when you think about it: Netflix wins market share by spending more money on content, while YouTube grows stronger by simply existing, allowing its users to handle content creation.

When Netflix has a fallow programming period, YouTube gobbles up market share by simply keeping its servers running and selling ads. Unfortunately for Netflix, the past three months have brought fewer hit shows than usual and a shrinking presence in Nielsen’s weekly Top 10 program rankings.

Perhaps Netflix will rebound with a Stranger Things spinoff or an American adaptation of Squid Game, but that would require exactly what streamers now seem reluctant to do: spend more money and take more risks. When we chart TV’s most expensive seasons by premiere year, we see a sharp drop-off in high-priced productions beginning in 2025, while 2026 has yet to produce a show expensive enough to crack the list.

Trend 2: The Obsession Obsession

Imagine the year is 2025 AD, and we’re playing a box office guessing game. I ask you to pick the highest-grossing 2026 release from the following options: Scream 7, The Mandalorian and Grogu, or a $750,000 movie called Obsession, made by a first-time director with an unknown cast.

You would have picked Star Wars. If you say otherwise, you are lying.

And yet here we are in 2026, and an unknown horror-comedy acquired out of a film festival has become a box-office sensation.

Released in mid-May, Curry Barker’s Obsession has grossed over $400 million on a budget of just $750,000—roughly the cinematic equivalent of buying Apple stock in 1985. That places Obsession in exceedingly rare company: it now has the second-highest ROI of any film released since 1970, alongside microbudget classics like Halloween and The Blair Witch Project.

While much of the box-office reporting surrounding this movie focuses on its unprecedented success, there is indeed precedent. The list of high-ROI films has long been dominated by thrifty horror movies.

Trend 3: America’s Royal Wedding Has Captured Hearts, Wallets, and Metallica’s Scorn

The world’s biggest pop star married one of the NFL’s most dominant players, and if you were somehow out of the loop, well, now you know.

Taylor Swift’s wedding was conducted with a level of secrecy that would make the Freemasons jealous. In the absence of concrete information, the world did what it does best: speculate wildly. And wherever there is mass speculation, you will inevitably find a horde of degenerates eager to gamble on that uncertainty through prediction markets.

Read more

Intro: The Peak of Peak-Superhero

There was once a time when the single biggest thing in pop culture was a series of stories written for preteens about people dressed in form-fitting spandex. What we now recognize as peak-superhero was, at the time, touted by Marvel as “Phase Three” of its vast cinematic universe. And if one were attempting to locate the zenith of peak-superhero, that date would be April 2nd, 2019, when tickets for Avengers: Endgame went on sale.

When the presale window opened, AMC’s site crashed, causing its social media team to make a lame joke about Thanos “snapping” the website out of existence—a reference that the superhero-agnostic will simply not understand. According to moviegoing platform Atom Tickets, Endgame sold three times as many presale tickets in its first hour as Avengers: Infinity War had the year before. When the film finally premiered, the AMC in Times Square reportedly scheduled 41 opening-day showtimes, including screenings at 3:30 a.m., 4:30 a.m., and 5:30 a.m. Demand was so extreme that AMC kept 64 locations open overnight, with 17 theaters operating for 72 consecutive hours.

If you had lived your entire life loving superheroes, then this was your moment. Something that had once been niche, the province of nerds at sparsely attended Comic-Cons, had become one of the biggest events in pop culture history. Unfortunately, that moment did not last.

Fast forward to last weekend, a mere seven years after the premiere of Avengers: Endgame, when the newest comic-book adaptation, Supergirl, debuted to just $37 million at the U.S. box office, one of the weakest openings for a superhero movie in recent memory. And this follows a long string of underperforming titles, including The Marvels, Eternals, The Flash, Blue Beetle, and pretty much everything else released after 2020, with a handful of exceptions we’ll get to later.

Peak-superhero is over. The format is now in decline, like so many entertainment fads before it—though few pop-culture phenomena have matched the cultural heights, scale, or corporate engineering of the Marvel Cinematic Universe. Which means the genre’s fall may be especially steep.

So today, we’ll quantify the rise and fall of the superhero film, examine the forces now plaguing the format, and consider the historical precedent for once-dominant genres falling out of fashion.

Today’s newsletter is sponsored by The Stat Significant Dataset Hub

Access 265+ Curated Datasets to Elevate Your Analysis

Want the data behind the deep dives? The Stat Significant Data Hub gives you access to 265+ curated datasets across movies, music, TV, economics, and more.

New datasets are added weekly, with full archive access for Stat Significant paid subscribers. Since last week’s post, we’ve added 18 new datasets, covering everything from gold prices and trending Twitter topics to Waymo crash rates and daily Google Trends tracking.

Check Out The Dataset Hub

How Superhero Movies Lost Their Power

In the mid-20th century, comic-book adaptations were considered lowbrow fare, confined mostly to janky television shows. The most prominent of these small-screen projects was the live-action Batman series, which ran from 1966 to 1968 and regularly punctuated its fight scenes with title cards shouting “BAM!” and “POW!”—or simply tilted the camera sideways to simulate Batman and Robin scaling a building. In fact, much of the show’s legacy centers on its thriftiness.

These adaptations achieved a level of camp no human being could earnestly replicate today—though that will never stop Ryan Murphy from trying his darndest.

The place of superhero stories in popular imagination changed dramatically with 1978’s Superman, starring Christopher Reeve, and again in 1989, when Tim Burton brought Batman to the big screen. Children who had grown up on Marvel and DC comics were now adults with credit cards, ready to buy tickets to see their childhood fantasies realized as blockbuster spectacle.

In the 2000s, comic-book stories increasingly took place in worlds that looked much like our own. Within a five-year span, audiences got Sam Raimi’s Spider-Man 2, Christopher Nolan’s The Dark Knight, and Robert Downey Jr. as Iron Man—three films that treated superheroes not as camp curiosities, but as vehicles for romance, crime drama, political allegory, and movie-star charisma. The success of these films, and more specifically the enormous box-office upside of their sequels, helped trigger a boom in superhero releases. At the height of this phenomenon, Hollywood was producing between four and eight comic-book adaptations a year. That is, until the pandemic.

The pandemic could hardly have come at a worse time for Marvel. The studio’s seemingly conclusive Avengers: Endgame arrived a few months before March 2020, and the chaos surrounding theatrical exhibition scrambled the studio’s transition into its next phase—assuming such a plan ever existed.

Between 2015 and the start of 2020, superhero films averaged roughly a billion dollars at the global box office (per release). After the pandemic, that figure fell off precipitously.

It is tempting, then, to conclude that COVID-19 made the superhero film obsolete. And to some extent, that is true. But the story is more complicated.

While the box-office value of comic-book IP declined during the pandemic years, interest in the Marvel Cinematic Universe actually peaked in 2021 and 2022, as Disney began turning comic-book stories into streaming series for Disney+. In the short term, this worked: Marvel stayed culturally relevant while theaters struggled to recover.

You’ll notice, however, that interest in Marvel began to fall in 2023, likely due to some combination of theatrical disappointments—such as Eternals and Shang-Chi—and the broader cheapening of the Marvel brand through the aforementioned Disney+ series, which undermined the scarcity of superhero storytelling.

And yet, comic-book movies still have one overwhelming bright spot: a single character who retains an unusually firm grasp on popular imagination—Spider-Man.

Over the past few weeks, the forthcoming Spider-Man: Brand New Day has generated the most TMDB web traffic and the highest number of YouTube trailer views for any upcoming release, suggesting that even as the broader superhero machine sputters, this particular character retains cultural cachet.

Spider-Man: No Way Home grossed nearly $2 billion in 2021 despite lingering COVID-19 disruptions, while the animated Across the Spider-Verse was rapturously reviewed and earned nearly $700 million worldwide in 2023.

The success of these films gets to the heart of Marvel’s post-Endgame problem. Legacy characters can still summon something like event-level excitement; new ones, for the most part, cannot. Try as Disney might, it has struggled to introduce a fresh generation of heroes capable of replacing the original Avengers. Split the post-2019 superhero landscape into Spider-Man and non-Spider-Man films, or into legacy characters and new character launches, and the pattern becomes hard to miss: the holdovers still perform, while new mythology does not.

Throughout the 21st century, Hollywood has increasingly sought out stories with built-in audiences. For most of the 2010s, this meant adapting comic books and rebooting once-profitable franchises as nostalgia bait. But recently, commercial appetites have shifted.

Built-in demand still matters, but the sources of that demand are changing. Consider the 10 highest-grossing movies of the year thus far: a mix of video game franchises, music biopics, book-to-screen adaptations, and YouTube creators converting their online followings into theatrical audiences.

The box office is moving in a new direction—one that is indifferent to superheroes not named Spider-Man.

And if what I’m describing sounds inconceivable to you—because superhero films are simply too big to fail, and because they have defined the mainstream for the better part of 15 years—know that there is ample precedent for this cultural shift.

Genres and storytelling concepts come and go. Musicals and westerns were among the most popular genres of the 1950s and 1960s. Now, these formats are essentially extinct, while thrillers, action movies, and horror films have taken their place.

Sometimes I like to envision telling my son about the MCU, but in the past tense: how these films dominated the zeitgeist for a decade and how Avengers: Endgame offered one of the greatest moviegoing experiences of the 21st century, and how, when the enterprise came to its logical conclusion, people simply refused to let it go. And he’ll say something like “That’s interesting, I guess,” and then go back to scrolling on TikTok.

Final Thoughts: An Unfortunate Hangover

I have always been a casual consumer of comic-book content: if it’s good, I’ll go. So when a friend offered me midnight premiere tickets to Avengers: Endgame, I said, “Sure, why not? This seems like a big deal.”

And it was. When an MCU’s-worth of Avengers assembled during the film’s climactic battle, people in the theater stood up and started clapping.

As we walked out of the theater, I told my friends how impressed I was that Marvel had “ended on such a high note.” After a few moments of genuine confusion, they informed me that the Marvel Cinematic Universe was not, in fact, “done.” A Spider-Man movie was coming out in a few months.

This was not a bit. I simply assumed that if a franchise put the word “End” in the title, it meant some kind of actual ending. Had the film been called Avengers: We’ll Keep Making These Until We Go Bankrupt, I would have adjusted my expectations accordingly.

While I was thoroughly uninformed on this topic, I believe my misreading was a feature, not a bug. Marvel wanted to have its cake and eat it too: they got the box-office frenzy of a conclusive chapter without actually concluding anything. And yet something really did end with that movie. In many ways, the strategy resembled a booze bender: you borrow future happiness for the present, and in return, you wake up with a hangover. Unfortunately, this particular hangover is now enshrined across 25% of Disneyland.

All stories, whether standalone or serialized, require an ending. Even inside a sprawling fictional universe, there comes a point when the narrative simply runs out of gas. Marvel’s mistake was eventizing that logical endpoint so the film would make $3 billion.

What remains of peak-superhero is reheated leftovers: Robert Downey Jr. was paid $100 million to return for two Avengers films; Chris Evans is coming back despite his character dying; Hugh Jackman was also resurrected; Spider-Man remains the one beloved holdover; and audiences broadly refuse to embrace anything new.

It’s as if the year were 1972 and the world’s largest entertainment conglomerate bet its future on the eternal dominance of Hollywood westerns. Nothing lasts forever, no matter how hard Disney tries.

Enjoyed the analysis? Support Stat Significant and unlock the data behind the insights.

Become a paying subscriber to get bonus graphics, interactive dashboards, downloadable datasets, and to support independent journalism.

Paying subscribers get:

• Bonus Data Stories: Full access to the Monday Roundup, a twice-monthly bonus edition with bonus culture data stories, curated datasets, and recommended reads.

• The Public Dataset Directory (Updated Weekly): A curated directory of public datasets from around the web.

• Interactive Data Hubs: Interactive dashboards across Streaming TV, Movies, Music, Search Trends, Wikipedia, and more.

Upgrade to Paid

*Note: For best results, open this link in a web browser while logged in to your Substack account.

Want to promote your data tool or media publication to Stat Significant’s 26,200 readers? Email daniel@statsignificant.com

Need help with a data problem? Book a free data consultation

Connect with me on LinkedIn if you want to chat about a data project.

Like movies? Follow me on Letterboxd

Intro: A “Booming” Campaign for the Ages

In the 1920s, American pop music ran through Tin Pan Alley, a Manhattan district where songwriters, performers, and promoters gathered to turn simple melodies into mass-market commodities. It was one of the earliest examples of entertainment-industry agglomeration, a precursor to music factories like the Brill Building and Motown.

Tin Pan Alley eventually reached its cultural apex with the runaway success of 1923’s “Yes! We Have No Bananas.” For the uninitiated, the song is exactly as banana-centric as advertised, and is so overwhelmingly repetitive that it sparked brawls between those playing it endlessly and its unwilling listeners.

Yet this virality was hardly organic. Tin Pan Alley manufactured popularity by funneling sheet music to “song pluggers,” who performed new tunes in stores, theaters, and restaurants, and by staging “booming” campaigns, in which musicians were planted in public crowds to play a song again and again so it might seem like a preexisting craze.

I came across the exploits of Tin Pan Alley while researching a recent article and was surprised to learn that the music industry has been engineering catchiness long before drum machines, auto-tune, Taylor Swift, or TikTok algorithms. In many ways, these early promoters were the first practitioners of a hybrid science: part sonic optimization, part behavioral manipulation, and ever more refined with each passing generation.

So today, we’ll dig into the scientific underpinnings of song catchiness, examining the compositional tricks that make tracks memorable and the environmental conditions that make a tune inescapable.

Today’s newsletter is sponsored by Quantum of Sollazzo

Want a Weekly Guide to All Things Data?

Quantum of Sollazzo is a weekly data newsletter followed by 4,000+ data professionals with the best links to data journalism, data visualization, tools, tutorials, data thinking, and AI. Every Tuesday in your inbox.

👉 Subscribe to Quantum of Sollazzo

What Makes a Song Catchy: The Musical Hallmarks

There is no single formula for catchiness, so much as a collection of musical tendencies that, when combined with the right artist and cultural moment, can make something indelible.

Two recent studies have tried to quantify these sonic hallmarks, mining auditory data from thousands of hit and non-hit songs to identify the musical features most closely associated with memorability.

The first study, conducted by researchers at National Yang Ming Chiao Tung University in Taiwan, identified five sonic traits meaningfully associated with memorability: higher energy, faster tempo, a prominent lead melody, greater emphasis on D notes, and a more positive mood. If you’re completely devoid of musical talent (like me!), then here are some example songs that satisfy nearly all of these requirements: “Call Me Maybe,” “Love Story,” “Uptown Funk,” “I Gotta Feeling,” “Dancing Queen,” and “Shake It Off.”

A similar study from researchers at Goldsmiths, University of London, also found tempo to be a key factor, alongside a track’s melodic shape and the unexpectedness of its pitch leaps. Earworms typically rely on a traditional rise-and-fall structure, giving them a recognizable melodic arc, while unusually large jumps between notes add just enough surprise to make the song feel both familiar and distinctive all at once.

The textbook example of this rise-and-fall-plus-surprise pattern is Adele’s “Hello,” a melodramatic dirge that lulls you into complacency before vaulting to level 11 in a moment of sonic exaltation. For a few seconds, the song gives you something startling, then quickly settles back into the familiar, returning the listener to their musical comfort zone.

This never-ending tug-of-war between familiarity and novelty is a common thread in nearly all research on song memorability. One study from the Max Planck Institute for Human Cognitive and Brain Sciences asked participants to rate musical excerpts across three dimensions—certainty, surprise, and pleasure—and found that tracks resonated most when listeners had a sense of where songs were headed (high certainty), only to be rewarded with a brief moment of novelty (high surprise). In other words, people want the same thing—but a little different.

And few devices capture this mix of familiarity and surprise better than the compound vocal hook, which stacks recognizable musical ingredients to make something unexpectedly fresh. I had not encountered the term before writing this essay, so for the similarly uninitiated, a compound hook is exactly what it sounds like: a hook made of hooks. The textbook example is Pharrell’s “Happy,” where the title phrase, call-and-response structure, backing vocals, rhythmic vocal delivery, chorus lift, and handclaps all converge into a near-perfect moment of musical memorability.

One study from the University of Wollongong in Australia tested the memorability of different hook strategies, comparing vocal hooks, instrumental hooks, compound hooks, and songs with no hook. The results were not particularly surprising: compound vocal hooks earned the highest scores, suggesting that the stickiest songs do not rely on a single ear-catching trick, but layer several of them at once. And because these hooks are vocal, we can sing them, hum them, and quietly rehearse them to ourselves, keeping them lodged in memory and allowing them to spread from person to person with viral efficiency.

So let’s say you are Pharrell, and you’ve created a track as unforgettable as “Happy.” Do you simply set it free and see what happens? Of course not.

Song composition alone does not make a track inescapable. That requires promotion, repetition, and distribution: the machinery that turns an infectious track into a cultural plague. A “song of the summer” does not arrive fully formed. It is engineered at every stage, from the songwriting itself to the shameless promotional strategies perfected by record labels.

What Makes a Song Catchy: The Environmental Factors

Within the broader architecture of music-making, there are several small optimizations that can increase a track’s chances of becoming inescapable. Many of these tactics are tailor-made for short-form video and playlist culture, yet their underlying logic dates back to the 1920s and Tin Pan Alley. The song promotion of that era was often brazen and ethically dubious, but it understood something fundamental about hit-making: sometimes the best way to lodge a song in popular imagination is brute-force exposure.

An experiment from researchers at Umeå University in Sweden found that song enjoyment increased with repeated exposure, regardless of a track’s complexity. A song could be as densely layered as Led Zeppelin’s “Stairway to Heaven” or as basic as Nickelback’s “Photograph”; either way, most listeners like it more on the twentieth listen than on the first.

Much of this stems from our primordial attraction to the familiar. The same Umeå University study asked participants to rate songs for likability, then grouped the results by how familiar each listener was with the track’s style. The pattern was depressingly straightforward: the more recognizable a song was, because listeners either knew the band or had heard similar artists, the more they tended to enjoy it. And with each listen, that preference only grew stronger.

You can think of musical taste as an ever-expanding lineage. People rarely make radical leaps between genres or styles. Instead, their preferences tend to evolve incrementally, moving from classic rock to a neighboring genre like pop punk through a chain of similar artists that make for a logical transition.

So if you loved a '90s band like Creed, with its particular brand of alternative-rock melodrama and mid-tempo earnestness, then Nickelback’s arrival in the early 2000s probably felt like a logical next step. The former helped lay the groundwork for the latter, smoothing the path from one generation of radio rock to the next.

That is, until you, an impressionable teen, learned that Nickelback was not a band to be embraced, but apparently “the worst band of all time.” As someone who has spent far too much of his professional life thinking about Nickelback, both the band and the meme, I often wonder how many potential fans were snuffed out by the group’s subpar reputation. Which brings us to the final ingredient of song pervasiveness: social influence.

Our taste in music is heavily shaped by the people around us. If a close friend loves a song, we may seek it out; if they hate it, that judgment can poison the well before we ever press play.

Researchers at Princeton tested this dynamic by giving participants a set of unfamiliar songs and varying the social context in which they appeared. Some participants received only the songs themselves, with no indication of what anyone else thought. Others were placed in environments where they could see which tracks their peers had liked and downloaded. The results were all too predictable: when listeners were exposed to social signals, download patterns became far more volatile. Songs were more likely to become runaway hits or total nonentities, suggesting that popularity can be shaped, and sometimes manufactured, by the visible preferences of others.

We are surrounded by these sorting mechanisms in everyday life: Billboard charts, TikTok view counts, Spotify rankings. These systems make success self-compounding, allowing a Taylor Swift album to saturate popular playlists, attract even more attention, and remain ubiquitous long after its initial release.

Final Thoughts: The Pop Star Illusion

In the early 2000s, Jessica Simpson was an inescapable cultural force. Today, I imagine few members of Gen Z know much about the singer-turned-reality-TV-star or the zeitgeist’s fixation on her inability to decode a tuna can label (if you don’t get the reference, a one-second Google search will suffice).

Simpson’s cultural ubiquity was so complete that when her younger sister Ashlee launched a music career of her own, it felt like a natural extension of the Simpson Industrial Complex.

By 2004, Ashlee Simpson was riding high on the success of “Pieces of Me,” a fizzy pop-rock hit that peaked at No. 5 on the Billboard Hot 100, earning her an invitation to perform on Saturday Night Live—a show that would ultimately undo her short-lived career.

During her second performance of the night, the wrong pre-recorded vocal track started playing. Faced with the impossible task of lip-syncing the words of an entirely different song, Simpson froze, performed a strange little jig, and walked off the stage.

Ashlee Simpson’s catastrophic non-performance dominated the national conversation for what felt like four years, prompting a series of uneasy questions that no one had bothered to ask in the first place. Why had we simply accepted that Jessica Simpson’s sister was also a pop star? Was this person a music industry plant incapable of singing? If Ashlee Simpson played a limited role in the creation of her music and could not perform that music live, then what was the point of her art?

Had you asked those questions before the Saturday Night Live debacle, I’m not sure the answers would have been any different. What changed was that the illusion of Ashlee Simpson’s pop stardom had been broken. In one painfully awkward ten-second stretch, the machinery became visible: “Pieces of Me” was not the pure expression of a singular artist, but a song engineered for catchiness by co-writers and producers, then attached to a performer who could plausibly sell it as her own.

With every passing year, the music industry becomes more mechanized. From Tin Pan Alley to the Brill Building to Motown to modern hitmakers like Max Martin and Jack Antonoff, record labels have perfected the art of producing optimally catchy songs and then blasting them into ubiquity through brute-force exposure. The role of the pop star, then, is to complete the illusion: to make the track feel like an extension of their persona, even if they had little role in its creation.

Music production is an alchemical science: a many-variable equation calibrated around tempo, pitch, texture, familiarity, and repetition. When it works, you get a song like “Pieces of Me” or “Yes! We Have No Bananas”—a track so efficiently constructed, so faithful to pop’s oldest formulas, and so relentlessly pervasive that it burrows into your brain. When it doesn’t, you get every other Ashlee Simpson song, along with all the tracks you’ve never had to erase from memory because they never made it there in the first place.

Enjoyed the analysis? Support Stat Significant and unlock the data behind the insights.

Become a paying subscriber to get bonus graphics, interactive dashboards, downloadable datasets, and to support independent journalism.

Paying subscribers get:

• Bonus Data Stories: Full access to the Monday Roundup, a twice-monthly bonus edition with quick-hit culture data stories, curated datasets, and recommended reads.

• The Public Dataset Directory (Updated Weekly): A curated directory of public datasets from around the web.

• Interactive Data Hubs: Interactive dashboards across Streaming TV, Movies, Music, Search Trends, Wikipedia, and more.

Upgrade to Paid

*Note: For best results, open this link in a web browser while logged in to your Substack account.

Want to promote your data tool or media publication to Stat Significant’s 26,200 readers? Email daniel@statsignificant.com

Need help with a data problem? Book a free data consultation

Connect with me on LinkedIn if you want to chat about a data project.

Like movies? Follow me on Letterboxd

Introducing: The Monday Roundup

Welcome to Stat Significant’s Monday Roundup, a twice-monthly bonus edition featuring quick-hit pop culture stories, original graphics, curated datasets, hand-picked data journalism, and recommended culture reads.

Your regular Wednesday essay is not going anywhere. It stays free for everyone—this is strictly additive.

Today’s edition has six original data stories. Free readers get the first three; paid subscribers get the full thing, plus curated datasets and recommended reading. Enjoy!

In Today’s Roundup:

• Original Data Stories: The end of explicit music, streaming’s most popular shows, Michael Jackson’s cultural legacy, Jesus-centric prediction markets, the World Cup on TikTok, and hockey romance books.

• Recommended Reads: Notable data journalism and pop culture reads from around the web.

• Unique Datasets: Ten publicly available datasets, covering everything from Wikipedia traffic to most-searched guitar chords to podcast listenership.

Trend 1: The End of Explicit Music?

The “explicit” music tag was created in the mid-1980s, following a series of congressional hearings led by Tipper Gore (Al Gore’s wife) and the Parents Music Resource Center (PMRC). By the early 1990s, the music industry had agreed to label explicit material with the now-iconic “Parental Advisory: Explicit Content” sticker, which appeared on physical album covers before becoming a metadata flag in the digital age.

At first, the warning label had the opposite effect, making it easier for teenagers to find the subversive music they were supposed to avoid.

And yet, over the past five years, something strange has happened: explicit music appears to be losing its grip on the mainstream. The percentage of explicit songs on Spotify’s Top 50 chart has fallen sharply, from 74% in 2018 to 13% today, suggesting that nearly 40 years after the parental advisory label was introduced, popular music has finally been sanitized. Congratulations, Tipper Gore.

The clean-ification of mainstream music can be attributed to two mutually reinforcing trends:

1. Increased consumption of classic songs: Nostalgia strikes again! Increasingly, listeners are using Spotify to revisit older, radio-friendly hits like Fleetwood Mac’s “Dreams” and Michael Jackson’s “Thriller.”

2. Declining popularity of hip-hop: Once a cultural behemoth, hip-hop is no longer the dominant force it was a decade ago, at least on Spotify.

In the end, the PMRC’s dream may have finally been realized, though not because of anything the organization actually did. If anything, the parental advisory campaign may have prolonged the cultural appeal of explicit music via the Barbra Streisand effect.

Trend 2: The Most Popular Shows on Streaming—A Tale of Two Islands

In the wonderful world of streaming television, popularity and critical acclaim rarely go hand in hand. For proof, look no further than two parallel, yet wildly different, shows about people stranded on islands: Apple TV’s Widow’s Bay and Peacock’s USA spin-off of Love Island.

Widow’s Bay is the prestige contender: critically acclaimed, currently holding a 78 on Metacritic, and arguably the buzziest new show on television, at least judging by how frequently the series is mentioned across TV-related Reddit communities.

After watching a few episodes, I can confirm that the hype is, in fact, warranted. Widow’s Bay can best be described as a bizarre horror-dark-comedy hybrid: part upper-tier Stephen King novel, part Parks and Recreation, part Looney Tunes nightmare. I am not sure I have ever seen anything quite like it.

And then there is Love Island, a show that is not critically acclaimed and may, in fact, thrive on critical disdain. Every time someone rolls their eyes at the mere mention of it, Love Island simply grows stronger, harnessing that disapproval as fuel.

I have never been a big reality TV fan and have done my darndest to avoid watching the show at all costs. I do, however, know that the series releases five episodes a week, which is not a typo, and that it was the most-watched original series on streaming last week by a significant margin.

You will also notice that Widow’s Bay is nowhere to be found on this top-ten list, because buzz among pop-culture intelligentsia does not translate into actual viewership. After three years of writing this newsletter, I should probably have this sentiment tattooed across my chest, Memento-style, just so I stop forgetting.

Trend 3: A Michael Jackson Renaissance

If you prefer not to think about Michael Jackson—because you do not like his music or because you are not inclined to separate the art from the artist—then you may not enjoy what comes next.

Read more

Intro: The Subtle AI and the Sloppy AI

A few weeks ago, I examined the prevalence of AI slop across the internet and found that there is, indeed, no shortage of AI-generated garbage—especially if you go looking for it. For many, “AI” is now used as a pejorative term. The other week, for example, someone told me they liked one of my recent essays because it was “the opposite of a large language model,” which is high praise in the year 2026.

Don’t get me wrong, I’ll take the compliment. But beneath the flattery is an assumption that all LLM-generated writing is unmistakably lazy and easy to spot. People are so preoccupied with obvious slop—like videos of Shrimp Jesus or weight-lifting cats—that they miss the subtler ways LLMs are reshaping internet culture: the way ideas are expressed, the punctuation we use, and the polarity (or perhaps uniformity) of our expression. All of this is changing right under our noses, while we’re distracted by videos of Harambe starring in Broadway musicals and speaking a bizarre hybrid of Portuguese and Swahili. We fixate on the sloppy and miss the subtle.

So today we’ll investigate how large language models are reshaping online culture, from the way we write to the ideological currents embedded within this new form of expression.

How Has AI Reshaped Internet Culture?

One hundred years ago, the term media referred to print newspapers and radio broadcasts. Today, that term includes TikToks, Substacks, Truth Social Truths, tweets, skeets, YouTube videos, YouTube Video Shorts, Tubi TV, and so on.

The past century has seen a ceaseless progression of newfangled communication technology, and with it intense scrutiny on how every new medium will melt our brains and kickstart the downfall of civilization. AI has now inherited the role of society’s newest moral hazard, and because the internet is already awash in AI-generated garbage, there is no shortage of evidence documenting its effects.

We’ll start our review with the superficial—and in my opinion more comical—before moving toward the far-reaching and subtly insidious. Which brings us to the word “delve,” the signature example of a new vocabulary favored by large language models.

Researchers from the Institute for Human Development in Berlin analyzed thousands of LLM-edited documents to see how word usage changed after the release of ChatGPT, comparing those shifts against each word’s prior trajectory. They found that terms like “delve,” “comprehend,” and “bolster” appeared far more often than historical patterns would have predicted, reflecting the handiwork of LLM-assisted authorship.

The study identified the words most overrepresented in LLM-generated writing and published a ranked top-20 list of this emerging lexicon.

You may notice that each of these words carries an air of sophistication, allowing LLMs to make lesser writing sound more polished. At first pass, this doesn’t seem so bad: who wouldn’t want Twitter tweets and Bluesky skeets to be more legible? But polish has its tradeoffs, which we’ll return to later.

And the influence of LLMs extends well beyond simple vocabulary changes. These tools are also reshaping basic grammatical competence, sentence structure, and punctuation, nowhere more clearly than in the Great Em Dash Conspiracy of 2025.

The em dash conspiracy is less a conspiracy than an observation: large language models really like using em dashes. In early 2025, the following graphic began circulating on LinkedIn and Reddit, showing the growing prevalence of em dashes on tech-focused subreddits following the release of a new ChatGPT model.

This finding has since been covered by numerous media outlets, including an in-depth exposé in The Washington Post, and is a huge bummer for me personally—a guy who enjoyed using em dashes before large language models ruined them for everyone. I have since been forced to wean myself off this once-beloved punctuation mark. So thanks a lot, Big Tech.

Whatever training data these companies rely on appears to associate the em dash with readability, offering users yet another shortcut to elegant-sounding sentences.

This synthetic boost in sophistication is apparent among ChatGPT’s most enthusiastic adopters: high school and college students. A recent study from the University of Warwick tracked the sentence complexity of college essays over the past decade using the Flesch-Kincaid grade-level score and found a meaningful uptick in writing quality following the release of ChatGPT.

Colleges have countered this trend by returning to good old-fashioned blue book essays written by hand. I would have loved to have ChatGPT when I was in school, but I also hated written exams, especially as a left-handed person. Fellow lefties will understand this persistently mild inconvenience. So perhaps the two cancel each other out.

Widely adopted models have birthed a new style of writing: heavy on em dashes, syntactically polished, and exceedingly empathetic—the last of these being an obvious paradox, given that all this warmth is exuded by a soulless large language model.

A study from the University of Chicago examined the average empathy and formality of over 600 email samples across three categories: LLM-only output, human writing aided by LLMs, and human-only writing. Human-only emails scored the lowest on both empathy and formality, while LLM-only emails scored the highest; emails produced by humans with LLM assistance fell somewhere in between.

I have always found the divide between human writing and ChatGPT-generated text somewhat perplexing, mainly because human writing is what trained these large language models in the first place. The implication is that there is a gap between how humans want to be addressed—the kind of language rewarded in model training datasets—and how people actually write to one another.

Indeed, the previously mentioned study of college essays reflects this shift toward inorganic empathy, showing a sharp rise in positivity after a dip during the pandemic years.

Humans are combative and emotionally inconsistent. ChatGPT and Claude, by contrast, have been trained to produce language that is pleasing by design, endlessly receptive to whatever half-formed thought a user puts in front of them.

This zealous empathy has become an unmistakable hallmark of LLM-produced writing, and was recently parodied on South Park. In the episode, Stan’s dad Randy becomes infatuated with ChatGPT because it takes his ideas seriously, no matter how stupid they are. The chatbot praises him, then offers only the gentlest revisions. His wife, disturbed by the growing intimacy between man and machine, eventually realizes that the only way to reach him is to mimic the chatbot: affirm his ideas, flatter his ego, and respond to every dumb thing he says with warm, detached encouragement.

And this is where things start to get weird. A recent study of randomly sampled websites by the startup Graphite found that roughly half of all web pages show signs of LLM-generated copy. Which means that rhetorical habits favored by ChatGPT and Claude may soon become the house style for all internet writing.

If a majority of digital culture stems from one of three large language models, then what will happen to diversity of thought? What gets lost in our effort to produce optimally readable text?

A comprehensive study by researchers at Stanford, Imperial College London, and the Internet Archive found that the more likely a piece of text was to be produced by a large language model, the more likely it was to be “semantically similar,” which is a technical way of saying that the text expressed similar ideas and patterns of thought (reflected in the leftmost graphic). They then examined how web pages have changed over the past several years and found that semantic similarity has risen in lockstep with the prevalence of AI-generated web pages (reflected in the rightmost graphic).

Much of the past decade’s cultural criticism has focused on the fragmentation of digital life: the sense that humans can no longer agree on anything, at least not online. But as AI-generated text begins to displace human expression, that disagreement may simply be smoothed over, buried beneath a wave of synthetic politeness produced on behalf of humans too lazy to disagree in their own words.

Where we once seemed headed for a dystopia resembling A24’s Civil War, we may now be drifting toward something closer to WALL-E or George Orwell’s 1984—a world of sedation and sameness.

For anyone who believes society is perpetually on the brink of collapse (which is a super fun way to live your life!), the question may simply be which dystopia you prefer: one where human expression remains unmediated and produces chaotic disagreement, or one where everything becomes sterilely, oppressively the same. A world where robots have overtaken our means of communication not by force, but because we found their version of us to be more pleasing.

Final Thoughts: An Optimizable Science

If you’ve had the good fortune of taking a communication studies class in college, then you were probably forced to read Marshall McLuhan’s “The Medium Is the Message” at least once—and, in my case, roughly five separate times. For an essay published in the 1960s, McLuhan’s arguments have aged remarkably well, somehow remaining prescient while also being extremely vague. His central thesis is right there in the title: the technology or platform used to deliver a message changes society more profoundly than any single message published on that platform (e.g., the invention of Facebook matters more than any one post ever published on Facebook).

Each new communication platform—be it movies, television, or social media—changes how public discourse operates. Television, for instance, transformed politics, education, and news into forms of entertainment, turning serious information into spectacle. Without fail, every decade or two, some new media format arrives, rewires how humans think, argue, and entertain themselves, and makes Marshall McLuhan look like a genius all over again.

Which brings us to AI. Large language models are not a medium in the traditional sense; they are a tool. And yet their impact resembles the rapid adoption of a new media platform.

If television changed public discourse by turning it into performance, and social media changed discourse by bending it toward instant gratification and attention capture, then AI has arrived to turn human expression into an optimizable science—no different from coding or medicine. There is, apparently, a platonic ideal for communication, one that involves the word “delve” and unwavering positivity, and AI is here to help us get there faster. There will be no more typos. No more half-formed thoughts. And all writing will be rendered in the frictionless, error-free style of Claude Opus 4.8 or GPT-5.5, and sometimes Gemini if the other two models happened to have crashed that day.

So what does all this mean for art? Part of what I have always loved about entertainment is the tension between art and science: the question of how to make something beautiful and moving, like a film, while also sustaining it within the limits that society imposes, such as the Hollywood studio system. But with AI, that balance starts to disappear; everything becomes science. A sentence can be made maximally engaging, but in the process it strips away the messiness of being human. And for all its flaws—and there are many—I do think we need that messiness. If that means living in a world of fewer em dashes, so be it.

Enjoyed the analysis? Support Stat Significant and unlock the data behind the insights.

Become a paying subscriber for interactive dashboards, ad-free reading, downloadable datasets, and to support independent journalism.

Paying subscribers get:

• The Public Dataset Directory (Updated Weekly): A curated directory of public datasets from around the web.

• Ad-Free Reading: No ads, made possible by new Substack personalization tools.

• The Stat Significant Data Catalog: Downloadable datasets assembled, cleaned, and updated weekly by Stat Significant.

• Interactive Data Hubs: Interactive dashboards across Streaming TV, Movies, Music, Search Trends, Wikipedia, and more.

Upgrade to Paid

*Note: For best results, open this link in a web browser while logged in to your Substack account.

Want to promote your data tool or media publication to Stat Significant’s 26,100 readers? Email daniel@statsignificant.com

Need help with a data problem? Book a free data consultation

Connect with me on LinkedIn if you want to chat about a data project.

Like movies? Follow me on Letterboxd

Intro: From Styrofoam to Stranger Things

In the beginning, there were four channels. Each offered a few dozen programming options, with only modest turnover from year to year. You watched what you were given and did not complain about production quality. Sure, Doctor Who had wobbly sets, Star Trek featured Styrofoam boulders scattered across obvious soundstages, and Mighty Morphin Power Rangers spliced American actors into recycled footage from an entirely different Japanese show, but none of that mattered. The alternative was sitting alone with your thoughts or—God forbid—reading a book.

Then television became something more. First came The Sopranos, The Wire, Breaking Bad, and the early prestige TV darlings. Then Netflix arrived with streaming-era hits like House of Cards and Stranger Things. Soon, every major media company was launching a platform of its own, flooding the market with splashy original series in an effort to win the streaming wars. Each year crowned a new “most expensive show ever made.” Wobbly sets and Styrofoam boulders were no longer acceptable.

Which brings us to the year 2026 AD, a period of great retrenchment for television. There is now a vague understanding that TV’s glossy prestige era is over. Or is it? Television is certainly more polished than it was 30 years ago. The harder question is whether streamers have abandoned the wanton spending of the late 2010s.

So today, we’ll examine the rise of TV production budgets, identify the streamers that let their spending spiral out of control, and answer one of life’s great mysteries: Does having more money actually make something better?

Is TV’s Big-Budget Era Over?

For this analysis, I assembled a catalog of the most expensive TV seasons of all time, as measured by per-episode production cost. When we chart these high-priced seasons by premiere year, three distinct eras emerge: the before times (pre-2013), the early prestige era (2013–2018), and the streaming wars (2019–2024).

The latter two periods of prestige programming follow a similar arc: a sharp rise in spending, followed by a notable retreat—the likely product of imitation and correction. Which brings us to 2025 and 2026. The number of high-priced TV seasons fell sharply in 2025, and 2026 has yet to produce a show expensive enough to make the list.

When Stranger Things reached its series finale last year, industry coverage framed the moment as the end of an era: a period in which streaming services spent lavishly to create shows with the scale and production value of Hollywood blockbusters. At the time, I found the framing contrived—an attempt to turn an arbitrary milestone into semi-readable clickbait. In hindsight, this was, indeed, the end of an era.

From the premiere of Stranger Things in 2016 to its finale in 2025, television viewers ate well. The 100 most expensive seasons in TV history are a testament to just how spoiled we were: most of the streaming services behind these shows did not exist before 2018, yet they spent extraordinary sums to keep us entertained—all for $5.99 a month. We paid less and got more. What a time to be alive, and parked in front of a television.

And no platform produced more high-budget TV series than Netflix, the company that precipitated this content explosion. Across our catalog of high-priced television seasons, Netflix and Disney+ account for the largest number of expensive shows.

One could look at this graphic and conclude that Netflix simply spent its way to market leadership: pour enough money into glossy prestige series, and viewers will follow. But that is only part of the story. Quality—or at least production value—may matter less than abundance.

Netflix did not merely bring highbrow HBO-style programming to streaming; it embraced all forms of content: dating shows, docuseries, game shows, true crime, and, most recently, podcasts. In the end, Netflix succeeded by replicating what television had always been: a grab bag of curiosities, this time a bit shinier and available on demand.

In fact, Netflix’s high-cost programming accounts for a relatively small share of its original content. For every wildly expensive season of Stranger Things, there are, metaphorically speaking, 33 cheaply produced seasons of Love Is Blind. Meanwhile, other streamers faced a late-mover tax: entering an increasingly crowded market, they had to spend massive sums to capture attention. As a result, their early libraries leaned heavily on big-budget series that doubled as customer-acquisition vehicles. For later entrants like Disney+ and Paramount+, that often meant producing fewer shows at higher cost.

These companies spent and spent. But did all that money pay off? It depends on how you define success. Across our dataset of 238 high-priced television seasons, bigger budgets did not translate into greater acclaim: in fact, the more expensive a season was, the lower its online rating tended to be.

You could stop reading here, armed with a satisfying sound bite: spending more money on television makes it worse. Hollywood executives are idiots. Put you in charge, and you would build a captivating content library for pennies on the dollar.

But the reality is more complicated. The relationship between spending and acclaim is muddied by audience expectation. The most expensive television seasons of all time tend to fall into one of three categories:

1. Adaptations of preexisting intellectual property: Franchise series set within vaunted fictional universes like The Lord of the Rings, Star Wars, or Marvel must satisfy enormous built-in fan bases.

2. Finales and legacy sequels: Long-running shows and sequel series carry unusually high expectations, such as the final season of Stranger Things or HBO’s The Pacific, a spiritual successor to Band of Brothers.

3. A little-seen Amazon series called Citadel: the first season of this show cost $53 million per episode for no discernible reason beyond financial mismanagement.

In many cases, lavish spending is an attempt to meet outsized expectations. But a massive budget can only buy so much goodwill. Even at a reported $60 million per episode, The Rings of Power left many fans disappointed.

Of course, acclaim is only one measure of success. A show like Hacks can win 30 Emmys while drawing an audience comparable to a preseason football game. The more important question is whether all that spending translates into meaningful viewership.

On that front, the results are more favorable. The expensive shows on our list are 10 times more likely to appear in Nielsen’s weekly Top 10 ranking than a modestly budgeted program that did not make our dataset. Once there, these pricey programs remain on the charts longer and accumulate more viewing hours. While extravagant budgets may not produce critical acclaim, they do buy an audience.

The harder—and likely unanswerable—question is whether that incremental viewership was worth the cost. My best guess is that, for a time, it was. If you wanted to compete with Netflix, you had to spend aggressively to get people’s attention. But as the streaming market has matured and subscriber growth has slowed, that approach is no longer justified—as evidenced by the spending slowdown in 2025 and 2026. There is no longer a sound business case for Masters of the Air’s $250 million price tag, or for spending $300 million on a season of Citadel—a show that apparently exists.

Final Thoughts: A Glorious Arrangement

“Get Big Fast” emerged as one of the defining mantras of the dot-com era, most famously embraced by Amazon, where Jeff Bezos reportedly had the phrase printed on employee T-shirts. The strategy was simple: spend aggressively, acquire customers as quickly as possible, and seize control of a market before competitors could catch up. And while the dot-com boom ended in a small-scale financial crisis, a handful of spectacular startup successes turned “Get Big Fast” into a template for a generation of founders—and, eventually, for any new market that appeared ripe for disruption.

Yet for every startup that got-big-fast, hundreds more burned precious capital chasing bigness. Take Homejoy, an on-demand cleaning startup that raised $40 million and grew by offering first-time cleanings for $19.99, far below its usual $85 price tag. The company treated those discounts as customer acquisition spending, betting users would return at full price. They did not: only 15% to 20% booked again within a month, and fewer than 10% were still active after six months. Even after data showed the discounts were attracting unprofitable customers, Homejoy continued subsidizing growth to please its investors—until the company shut down in 2015, less than five years after launching.

Homejoy is simply one entry in a long line of businesses that chased scale without sustainability, joining a crowded startup graveyard that includes Pets.com, MoviePass, Quibi, FTX, Webvan, WeWork, Juicero, Munchery, Bird, and many more.

In the 2010s, this became the operating model for television, a market that had long subsisted on low-cost programming and the reliable economics of the cable bundle. The strategy was simple: spend massive sums on big-name TV shows, acquire customers, and maybe—just maybe—they will stick around.

For viewers, it was a glorious arrangement. We got wildly expensive programming at a subsidized price—and sometimes for free. If the streaming boom had merely produced a wave of above-average shows for very little money, that would be a charming story. But the consequences were considerable, irreparably changing the entertainment industry.

Since the streaming bubble began to burst in 2022, Hollywood has been littered with casualties of an untenable business model: the writers’ union went on strike, the actors followed, Paramount sold to Skydance, and the newly enlarged Paramount Skydance moved to swallow Warner Brothers. When the dust settles, there will be fewer studios, fewer jobs, and fewer shows. In exchange, viewers will have gotten ten years of unsustainably awesome television.

In the beginning, we got The Sopranos and Breaking Bad. In the end, Netflix got-big-fast while the rest of the industry saw mass consolidation. And for one brief, absurd moment in between, we got a $300M show called Citadel that tens of people watched.

Enjoyed the analysis? Support Stat Significant and unlock the data behind the insights.

Become a paying subscriber for interactive dashboards, ad-free reading, downloadable datasets, and to support independent journalism.

Paying subscribers get:

• The Public Dataset Directory (Updated Weekly): A curated directory of public datasets from around the web.

• Ad-Free Reading: No ads, made possible by new Substack personalization tools.

• The Stat Significant Data Catalog: Downloadable datasets assembled, cleaned, and updated weekly by Stat Significant.

• Interactive Data Hubs: Interactive dashboards across Streaming TV, Movies, Music, Search Trends, Wikipedia, and more.

Upgrade to Paid

*Note: For best results, open this link in a web browser while logged in to your Substack account.

Want to promote your data tool or media publication to Stat Significant’s 26,100 readers? Email daniel@statsignificant.com

Need help with a data problem? Book a free data consultation

Connect with me on LinkedIn if you want to chat about a data project.

Like movies? Follow me on Letterboxd

Intro: What If We Used to Be Worse?

In her essay “On Keeping a Notebook,” Joan Didion makes the case for journaling as a way of staying in touch with our former selves: “I think we are well advised to keep on nodding terms with the people we used to be, whether we find them attractive company or not.” This sounds charming in the abstract. But what if the person you used to be was significantly worse? What if willful ignorance is the wiser choice?

What if—God forbid—you once had terrible taste in movies? And what if a document existed chronicling these regrettable selections? Unfortunately for me, such a document exists. A few years ago, I found my middle-school yearbook, which was cute until it wasn’t. Every entry asked its subject to list their favorite things, and there, preserved in permanent ink, were my favorite movies: Scary Movie 3, Not Another Teen Movie, and Shrek 2. Yikes! No Citizen Kane. No Hungarian slow cinema. No Hitchcock. Not even The Godfather, Part I or II. Just spoofs and sequels.

In previous articles, I’ve explored how musical taste forms, finding that our favorite songs and styles take shape during adolescence and remain remarkably stable after our twenties. My middle-school movie favorites, by contrast, bear little resemblance to my viewing habits today. Perhaps our taste in film evolves differently than our taste in music. Maybe the radical reinvention of movie-watching preference is a universal phenomenon.

So today, we’ll explore how movie taste shifts with age and how nostalgia distorts our sense of filmmaking quality.

How Does Our Taste in Movies Change With Age?

MovieLens is an online community that collects movie ratings and uses them to recommend films to its users. In 2004, the project published an anonymized dataset containing more than 1 million user ratings submitted between 1995 and 2003, along with each reviewer’s age and gender. This archive spans several files and requires substantial assembly and cleaning before it can be analyzed, so I did the legwork of turning it into a usable, single-file dataset that paying subscribers can download here.

Using this extensive record of moviegoing preferences, we can examine how our tastes and viewing habits evolve with age.

As a new parent, my mind immediately went to movie-viewing volume. With a four-month-old, I am not exactly plugged into movie culture—or really any culture unrelated to said four-month-old. So I wanted to know whether this “parent tax” is universal: Do we watch fewer films as our responsibilities grow?

The answer, unsurprisingly, is yes. In the MovieLens dataset, average rating volume peaks among users ages 25 to 34, then declines with each passing decade.

But this metric captures all forms of movie watching: DVDs, cable television, and trips to the theater. When we isolate contemporary releases—films likely to have been seen in theaters—a different pattern emerges. Instead of peaking in early adulthood, theatrical moviegoing is highest in our teenage years and early twenties, then declines steadily thereafter.

A friend recently asked me whether the dueling box office success of Obsession and Backrooms was “bringing younger audiences back to theaters.” This question was likely inspired by some New Yorker trend piece that chronicles a sweeping cultural shift roughly two weeks before that trend is obsolete.

My hot take is that younger viewers have always made up a large share of theatrical audience. Whenever a movie becomes a surprise hit, people scrutinize its turnout, discover that young people were in the theater, and briefly treat their presence as a revelation.

At the same time, Obsession and Backrooms are both horror films, which brings us to the intersection of genre preference and age. You are more likely to find younger audiences at a scary movie inspired by a 4chan internet meme (Backrooms) than at a dad-core drama about a meteorologist trying to predict the weather on D-Day—which is the actual premise of a movie currently in theaters called Pressure. If you’ve seen Band of Brothers multiple times, own several Jack Ryan novels, and think Apple’s weather app is lacking in detail, then this movie’s for you.

Indeed, our MovieLens data reveals a clear generational divide across several genres: younger audiences tend to favor horror, comedy, animation, and action, while older viewers gravitate toward dramas, war films, and romance.

If genre preference were the only thing shaped by age, that would be a fun find, but not enough to justify an entire study. Fortunately—or perhaps unfortunately—there’s more.

In a strange and philosophically punishing twist, the factor most strongly shaped by age is time. Put less cryptically, age shapes how we relate to movies from the past and present, a phenomenon known as “nostalgia bias”: the tendency to favor older releases while viewing contemporary movies with greater skepticism.

Using the MovieLens dataset, we can compare movie ratings by the number of years between a film’s release and the reviewer’s birth. The results suggest that movies were best before you existed, declined around the time you arrived, and have been getting worse ever since—a sentiment consistent across generations.

Self-selection likely shapes our view of older “classics,” since the most revisited movies from the 1940s are those that stand the test of time, like Casablanca and Citizen Kane. More surprising is that our bias against newer films persists even after we come of age and begin actively consuming contemporary releases. So when a podcast host waxes nostalgic about the cinematic excellence of the 1990s, remember that, soon, a new generation of podcasters will be extolling the greatness of the 2000s and 2010s because that’s just how time works.

And if you are on the older side and feel somewhat attacked by this piece—because you own several Jack Ryan novels and think Pressure sounds like a pleasant night at the movies—know it’s not all bad.

According to MovieLens data, older cohorts tend to rate movies more favorably.

The caveat is that, as we observed earlier, older viewers also watch fewer movies, making them less likely to encounter films outside their core interests. But if we are choosing optimism, perhaps they simply know who they are and what they like—and what they like is movies about D-Day.

Decades from now, I too will sit at the far end of this distribution, enclosed within an igloo of Jack Ryan novels, deriding films based on 8chan memes, and yearning for the cinematic glory days of Shrek 2 and Scary Movie 3.

Final Thoughts: Nodding Terms

When I first approached this topic, I gravitated toward the negative: the idea that our younger selves are lesser versions of who we are today. But as I worked through the piece, my perspective began to shift—which is just a kinder way of saying I stopped being a sourpuss. In particular, I found myself contemplating my changing relationship with David Lynch’s Mulholland Drive.

I first watched this movie when I was 14, which is far too young to watch Mulholland Drive (for several reasons). For one, I had never encountered surrealist art. By the time the credits rolled, I was furious. What the hell had I just seen? I promptly declared it my least favorite movie of all time.

As I went through film school and entertainment-industry internships, I held on to that disgust. I was the guy who harbored a deeply personal grudge against Mulholland Drive. What a cool identity!

Then David Lynch died. A few days after his passing, I decided to give the film another chance. To my surprise, I came away believing the exact opposite. I was now a guy who loved Mulholland Drive. How silly! I subsequently watched the film three times over the next month and now regularly list it among my favorite movies.

There are many takeaways from this story, including the fact that I was an insufferable teenager. But perhaps the most heartening is that we never lose the ability to surprise ourselves.

Following my Mulholland Drive about-face, I’ve revisited several movies I once proclaimed “awful” or “all-time classics.” My perspective is often entirely different, as someone on the other side of puberty—and, more recently, as a parent. Sure, Not Another Teen Movie may not hold up, but neither do Oscar-winning “masterpieces” like American Beauty or Forrest Gump.

My evolving relationship with these films has become a kind of yardstick, not unlike Joan Didion’s notebooks: a way to catch glimpses of my former self. With each rewatch, I find myself “on nodding terms with the [person I] used to be,” better able to understand who I once was and how time has changed me.

Want to promote your data tool or media publication to Stat Significant’s 25,800 readers? Email daniel@statsignificant.com

Need help with a data problem? Book a free data consultation

Connect with me on LinkedIn if you want to chat about a data project.

Like movies? Follow me on Letterboxd

Intro: All of Film Knowledge, for Free

The Internet Movie Database (IMDb) was founded in 1990—an almost inconceivable fact for a website that holds present-day relevance. The site began as a series of Usenet message boards and later evolved into an encyclopedic reference point for all film knowledge.

For a millennial film nerd, IMDb was the playground of my early internet days. I spent endless hours wandering the site, memorizing Oscar winners, reading user reviews, and absorbing an alarming amount of useless trivia. It was thrilling, at least for a certain type of indoor kid, and it was just there, on the internet—for free. I absorbed all IMDb had to offer until my search for film knowledge hit diminishing returns. And then I discovered Letterboxd.

Letterboxd is a hip (slightly more pretentious) IMDb with a greater emphasis on social experience. The site focuses on list creation, with users sharing film preferences through curated guides and movie reviews. Examples of popular lists include:

• Befriending the lyrical loneliness…

• chick flicks

• what is reality?

The site has taken off since the pandemic, amassing tens of millions of users and billions of movie ratings, cementing itself as a pillar of contemporary film-nerd culture. But alas, Letterboxd—and its decidedly twee appeal—may be in danger. Earlier this year, reports surfaced that the startup’s largest investor was looking to sell its 60% stake in the platform, just three years after buying in. The news sparked a familiar anxiety: that Letterboxd has reached a pivotal moment in its development, and that the site’s charm could be corrupted if it falls into the wrong hands.

So today, we’ll explore Letterboxd’s unlikely rise, what makes the platform so distinctive, and the economics underlying its current dilemma.

The Rise and (Potential) Fall of Letterboxd

In the late ‘90s, there was roughly one movie rental store for every 9,000 people in the United States. These stores served as watering holes for film culture, providing a locale for cinephiles to discuss movies and breeding a generation of film lovers and directors, including Quentin Tarantino, Kevin Smith, and Paul Thomas Anderson.

Fast forward to the present day: video stores are nearly extinct, leaving movies without a physical place for community (because people don’t really socialize at theaters).

Enter Letterboxd. If IMDb and early-stage Twitter had a baby, it would look something like Letterboxd: a social network built around the compulsion to log, review, and argue about movies. At its most basic level, the site recreates the feeling of a long, spirited conversation held after the credits roll.

Letterboxd’s popularity has surged since 2020, with the site growing from 1.5 million users in 2019 to more than 26 million by early 2026.

For first-time visitors, Letterboxd may resemble a slightly modernized version of IMDb, which raises a fair question: what’s the big deal? Why does the world need another website where people can argue about Green Book and/or The Last Jedi?

The answer has less to do with the arguments themselves than with who is doing the arguing, and how that audience is reshaping modern movie culture.

The site’s user base skews younger, helping Gen Z—and soon, Gen Alpha—discover 20th-century classics and little-seen international gems.

A few months ago, I decided to see a Taiwanese film called Yi Yi at the theater down the street from my house. This particular theater does not offer digital ticketing, so seating is determined by arrival time, with most attendees arriving at least 30 minutes in advance. For Yi Yi, I took a more lax approach. How many people were going to see a 25-year-old, three-hour Taiwanese film on a Sunday night?

The answer: a full movie theater’s worth of humans.

Upon arrival, I was shocked to learn that I had received the last available seat for a sold-out screening, leaving me in the literal corner of a 400-person auditorium. Stranger still, nearly everyone in the theater appeared to be under the age of 30.

The reason? Yi Yi currently ranks as Letterboxd’s thirteenth-highest-rated film of all time.

When the movie ended, an audience full of people born after 9/11 burst into vigorous applause. For a brief 30 seconds, I genuinely wondered if I was hallucinating.

I am not alone in this experience. What I witnessed was part of a broader phenomenon media outlets have dubbed “the Letterboxd effect”: younger, digitally native audiences being drawn into movie culture via the platform.

While IMDb functions primarily as an online reference guide, Letterboxd offers something closer to a social discovery engine. Most visitors arrive directly at Letterboxd.com rather than arriving via Google search, and its users tend to spend more time per visit and engage with more content.

Much of this stickiness can be attributed to the holy grail of digital media: user-generated content. As the site grows, so does its repository of user ratings and reviews, with 2025 seeing an estimated 700 million ratings logged—outpacing IMDb by a significant margin.

On Letterboxd, the reviews can be as entertaining as the movies themselves. I often find myself reading comments for films I’ve never seen, simply because I enjoy eavesdropping on the thoughts of other movie nerds.

By most user-engagement metrics, Letterboxd is a juggernaut: one of the rare 2010s social media startups to achieve critical mass. But there is an important caveat. For all its momentum, Letterboxd has yet to translate its growing user base into meaningful revenue.

The site may have a passionate, highly engaged audience, but its estimated per-user value remains far below that of larger social platforms.

Letterboxd has solved the hard part of building a social platform: getting people to show up, participate, and care. Now comes the even harder part: turning that devotion into revenue. If you’re a heartless corporate goblin, the gap between Letterboxd’s cultural value and its financial performance looks like an obvious opportunity.

This monetization dilemma stokes a familiar anxiety for extremely online individuals: enshittification. The platforms most successful at monetizing their user base tend to follow a familiar playbook: manipulative algorithms, endless gamification, ads for low-quality products, and a social experience increasingly organized around consumption, à la TikTok Shop. Every one of these tactics runs counter to Letterboxd’s stripped-down appeal. Yet if the site fell into the hands of private-equity vampires with little attachment to its Web 1.0 charm, these would be obvious tools for extracting more value from the platform.

To avoid exploitation of the Letterboxd community, a group of industry professionals—including friend-of-the-newsletter Ted Hope—is organizing a crowdfunding campaign to place the site under the control of a public trust.

The proposed structure would resemble Wikipedia’s model: Letterboxd would be governed with an eye toward its long-term existence rather than generating shareholder value for venture capitalists, founders, or private-equity vampires.

Admittedly, I’m a bit skeptical of this approach, since Wikipedia is the only organization that has sustained something like this at a scale comparable to Letterboxd.

But a long-shot effort to preserve something I love still sounds preferable to watching an asset-management firm gradually suck the life out of it.

So, if you would like to help save Letterboxd, you can check out the campaign here.

Or don’t. You’re a free-thinking individual. You can do whatever you want.

Final Thoughts: Unfortunate Discovery

As Letterboxd grows, so will its costs, and the expectations of its investors. This dilemma—how to sustain the site’s growth and ambition without eroding its cultural value—sits between two competing parts of my personality.

On one side is the deeply American part of me that believes economic Darwinism should determine what survives and what fades into obscurity. Opposing this impulse is a willfully naive romantic who keeps asking the same question: “Why can’t we have nice things?”

As you might imagine, this internal dialogue mostly consists of me 1) fretting over the endangerment of “nice things,” then 2) answering my own concern with apathetic economic theory. According to the latter, Letterboxd’s existential dilemma is the simple product of a free-market “discovery process,” one that unfolds in two stages:

1. Identifying a need: for example, people enjoy wearing shoes when walking around outside.

2. Determining the economic value of that need: The global shoe market is worth billions of dollars, offering a range of price points and styles.

With Letterboxd, these two stages work against each other.

1. The platform has proven a need exists: cinephiles want a social community where they can log, review, debate, and celebrate movies.

2. The platform resists maximizing the economic value of that need: much of Letterboxd’s appeal rests on its community focus and its reluctance to introduce disruptive algorithms, aggressive advertising, or unnatural product placement. People like Letterboxd because it extracts relatively little from them—and many would leave if it started to resemble Pinterest, TikTok, or any over-optimized hellscape.

Paradoxically, Letterboxd’s greatest asset may also be the biggest obstacle to its long-term financial sustainability: its user base.

Which brings us back to IMDb, the only other site to build a meaningful social community around movies. How has this 36-year-old website survived all this time? The simplest answer is that IMDb found a benefactor, and that benefactor was Amazon. The Internet Movie Database functions as a loss leader for the e-commerce giant, with its user-generated content utilized in streaming features, movie rentals, and the broader Amazon entertainment ecosystem.

Contrary to my earlier belief, IMDb was never “just there, on the internet, for free.” There was a cost, and that cost was its independence.

Perhaps Letterboxd can defy the odds, operating as a quasi-public utility sustained by benevolent cinephiles. Yet absent a fundraising miracle, all roads may lead to a familiar question: Why can’t we have nice things?

Enjoyed the analysis? Support Stat Significant, and unlock the data behind the insights.

Become a paying subscriber for interactive dashboards, ad-free reading, downloadable datasets, and to support independent journalism.

Paying subscribers get:

• The Public Dataset Directory (Updated Weekly): A curated directory of public datasets from around the web.

• The Stat Significant Data Catalog: Downloadable datasets assembled, cleaned, and updated weekly by Stat Significant.

• Interactive Data Hubs: Interactive dashboards across Streaming TV, Movies, Music, Search Trends, Wikipedia, and more.

Upgrade to Paid

Want to promote your data tool or media publication to Stat Significant’s 25,800 readers? Email daniel@statsignificant.com

Need help with a data problem? Book a free data consultation

Connect with me on LinkedIn if you want to chat about a data project.

Like movies? Follow me on Letterboxd

Stat Significant’s Book Bestseller Trends 📖

Which books are breaking through right now, and which authors continue to dominate the bestseller lists?

The charts below track recent activity across Amazon’s Kindle Store and The New York Times Best Sellers list. Together, they provide a snapshot of the titles gaining momentum, the books showing sustained popularity, and the fiction and nonfiction authors appearing most often near the top of the charts.

All visuals are updated regularly, with new data added daily and weekly. When available, click “Get Data” at the bottom of a chart to download the underlying dataset.

Because many of the graphics below are interactive, this page is best viewed on the web rather than in an email client or the Substack app.

Table of Contents

For paid subscribers — click a link below to jump straight to that section:

1. Section 1: Trending Titles in Recent Weeks

   1. Top Trending Titles in Amazon’s Kindle Store: Last 7 Days

   2. Top Nonfiction Books on the NYT Best Sellers List: Last 4 Weeks

   3. Top Fiction Books on the NYT Best Sellers List: Last 4 Weeks

2. Section 2: Top Authors in Recent Weeks

   1. Top Fiction Authors on the NYT Best Sellers List: Last 2 Months

   2. Top Nonfiction Authors on the NYT Best Sellers List: Last 2 Months

3. Section 3: Top Books This Past Year

   1. Top Fiction Books on the NYT Best Sellers List: Last 52 Weeks

   2. Top Nonfiction Books on the NYT Best Sellers List: Last 52 Weeks

Section 1: Trending Titles in Recent Weeks

Top Trending Titles in Amazon’s Kindle Store: Last 7 Days

• What This Chart Shows: The top titles on Amazon’s Kindle store charts, across fiction and nonfiction, for the past 7 days. Titles are sorted by the number of days trending and then average chart position.

• Data Source: Amazon’s Kindle Store

• Data Update Frequency: Daily || Data Can Be Downloaded: ✅

Top Nonfiction Books on the NYT Best Sellers List: Last 4 Weeks

• What This Chart Shows: The top nonfiction titles on The New York Times Best Sellers List over the last four weeks, calculated from NYT's Combined Print & E-Book sales.

• Data Source: The New York Times

• Data Update Frequency: Weekly || Data Can Be Downloaded: ✅

Read more

Intro: The St. Elsewhere Snow Globe

Before there was The Pitt or ER, there was St. Elsewhere. Premiering in 1982, St. Elsewhere was one of the first modern medical dramas, chronicling the eccentric chaos of St. Eligius, a struggling Boston teaching hospital. In its time, the show was critically beloved, though its legacy is inseparable from one of television’s most divisive finales.

In the final minutes of the final episode, the show abruptly cuts to an alternate reality in which a lead character (Dr. Donald Westphall) appears not as a doctor but as a construction worker. His son Tommy Westphall, who is autistic, sits nearby staring at a snow globe. Inside the snow globe is a tiny replica of the very same St. Eligius Hospital, implying that the entire show existed in the boy’s imagination. Yikes!

Longtime fans were apoplectic: the preceding 136 episodes had been rendered meaningless—a thing TV viewers generally dislike. Despite overwhelming backlash, St. Elsewhere’s finale has endured within popular imagination. And nowhere is the show’s twisted legacy more apparent than in the Tommy Westphall Universe.

The Tommy Westphall Universe is a pre-Internet fan theory that imagines most television shows taking place inside the mind of St. Elsewhere’s Tommy Westphall. The Westphall Universe’s logic goes like this: because characters from St. Elsewhere appeared in crossover episodes for Homicide, and Homicide connects to the Law & Order universe, both shows can be attributed to Tommy. If you allow this web to keep expanding, then programs like Cheers, M*A*S*H, and Chicago Hope all exist within the mind of a child. Ultimately, St. Elsewhere’s finale was so baffling and widely disliked that fans built a satirical TV reality to make sense of their disappointment.

Four decades later, television has changed dramatically, reshaped by streaming and a clearer understanding of what makes for a satisfying conclusion. But has this institutional knowledge led to better endings? Have showrunners learned from the mistakes of St. Elsewhere, Game of Thrones, and other finale fiascos?

So today, we’ll investigate whether omniscient showrunner Tommy Westphall has gotten any better at sticking the landing, how finale quality has changed in recent decades, and whether finality is simply a structural weakness of television itself.

Do Most TV Shows Stick the Landing?

To assess the quality of television endings, we’ll compare each show’s IMDb finale rating with its average rating across earlier, non-final seasons.

For House of Cards, this calculation would go something like this:

• Average episode rating across first five seasons: 8.44

• Episode rating for the series finale: 2.6

• Finale’s difference from the show baseline: –69.2%

From there, we’ll sort each series into one of three buckets:

• Exceeds Expectations: The finale scores at least 10% higher than the show’s typical episode rating.

• Meets Expectations: The finale scores within 10% of the show’s typical episode rating, in either direction.

• Disaster: The finale scores at least 10% lower than the show’s typical episode rating.

To see this approach in action, consider the most popular “Exceeds Expectations” finales in recent memory, a list headlined by The Big Bang Theory, The Office, and Breaking Bad. Each of these shows ended on a high note, deepening viewer appreciation for the series as a whole.

And then there are the disaster endings: finales so deflating they make you question whether you should ever trust a TV show again. Somehow, I’ve been a fan of almost every series on this list, and can verify that these finales ruined my day, and sometimes my week. Game of Thrones, How I Met Your Mother, Dexter—I wish I had missed these endings because of a root canal or multi-day DMV appointment, or both. In fact, a root canal performed at the DMV could not surpass the ill will I have toward the finale of How I Met Your Mother.

Using our finale-quality metric, we can evaluate every series with a meaningful digital footprint to answer two key questions:

1. Do most shows end on a high note?

2. Have showrunners gotten better at sticking the landing?

The answers to these age-old conundrums are 1) “kind of” and 2) “yes.”

Across television history, 68% of finales meet expectations, 10% qualify as disasters, and 22% exceed viewer expectations. Better yet, finales appear to have improved over the past half-century, as disaster endings have become less common and exceptional conclusions have grown more frequent.

Television appears to have made meaningful progress in the art of endings, which raises a simple question: why?

The answer centers on program longevity, or rather, lack thereof.

How satisfying or enraging a finale feels depends heavily on when a show calls it quits. And paradoxically, the longer a show runs, the more likely its ending is to land at one of two extremes: exceeding expectations or imploding under the weight of them.

The greater fan investment—across years, sometimes decades—the stronger their reaction.

Before the streaming era, this narrative volatility was further compounded by the punishing pace of network television. Long-running hits from the 1980s and 1990s were mined for every last ounce of content: Cheers ran 11 seasons, ER ran 15, and M*A*S*H ran 11, each producing 22 episodes a year.

Week to week, episode quality could vary wildly. And nothing captured this uneven output better than “sweeps week.” In brief, “sweeps” were four Nielsen rating periods, traditionally held in November, February, May, and July, during which selected households recorded their viewing habits for Nielsen to tally. These results were then used to determine advertising rates.

As a result, showrunners were incentivized to create spectacle. They brought in buzzy guest stars, engineered cliffhangers, staged weddings and deaths, and produced novelty episodes that broke from a show’s normal rhythm.

What I always found strange about “sweeps week” was the implication that TV shows could simply choose to try harder. For a few weeks each year, every series became more dramatic, more expensive, and more desperate for attention, while the other 18 episodes received 75% of max effort.

Today, no such concept exists: shows run fewer seasons and feature fewer installments per season. Every episode matters.

And when a series does run longer, it has usually survived a much tougher gauntlet.

Among shows launched before 2022, we find that fewer programs make it to a second season, but those that do tend to produce better sophomore installments.

Taken together, these trends produce stronger finales: shows are less likely to linger for 11 seasons before limping toward an unsatisfying conclusion, and when a series does earn a second, third, or fourth season, its endurance is a testament to its quality.

Enjoying the article thus far and want more data-centric pop culture content?

Final Thoughts: What About The Pitt?

I have only participated in a weekly watch party for one show, and unfortunately, that show was Game of Thrones. By the time I exited adolescence, few series justified real-time engagement; most programs were consumed via streaming or TiVo. Game of Thrones remained the lone exception, for fear of spoilers ruining that week’s episode.

I was late to Game of Thrones, catching up through epic binge-watching sessions on a series of business trips. While I hate assigning emotional resonance to the act of sitting alone in a hotel room, these 5-hour watch sessions were exhilarating. And then the finale happened.

I still remember the collective groan from our watch party as the finale unveiled one of its many frustrating twists. All those hours binge-watching the Targaryens and Jon Snow were retroactively ruined. While I try my best to avoid sunk cost thinking, my Game of Thrones investment will forever feel like an absolute waste.

Such is the agony and ecstasy of television. The longer a show runs, the greater our emotional investment and the more likely it is to thrill or enrage us. A show can run years, sometimes decades, but an outsized burden is placed on its very last episode.

Which brings me to The Pitt, arguably the best show on television right now—or at least my favorite. For some reason, I find simulated leg amputations and mangled limbs deeply satisfying (and apparently I’m not alone). But how much can I give to this show if I have no control over its resolution?

What if:

• The series’ main character becomes a villain, moving from ER Doctor to hedge fund manager, determined to drive the hospital out of business.

• The show loses its control of tone, and the characters look into the camera and say, “It’s Pitt-ing time!” Whenever an emergency trauma arrives.

• The entire series takes place inside a child’s mind as they stare at a snow globe of Pittsburgh.

Maybe the real goal of this analysis was reassurance: a reminder that disastrous endings are relatively rare, even if the worst finales loom large in our collective imagination. More often, a show’s conclusion will exceed expectations, or at least meet them, unless prolific showrunner Tommy Westphall finds a new way to disappoint us.

Enjoyed the analysis? Support Stat Significant, and unlock the data behind the insights.

Become a paying subscriber for interactive dashboards, free datasets, and to support independent journalism.

Paying subscribers get:

• The Stat Significant Dataset Hub (Updated Weekly): a collection of free datasets across movies, music, TV, and economics.

• Interactive Data Hubs:

  • Streaming TV: what people are watching on television.

  • Movies: the films watched in theaters and on TV, plus historical trends.

  • Music: the songs, artists, and styles dominating Spotify and TikTok.

  • Podcasts: the most-listened to episodes and series.

  • Wikipedia: the top pages trending on Wikipedia.

Get access with:

• A $6/month Substack subscription

• A $60/year Substack subscription

Upgrade to Paid

*Note: For best results, open this link in a web browser while logged in to your Substack account.

Want to promote your data tool or media publication to Stat Significant’s 25,800 readers? Email daniel@statsignificant.com

Need help with a data problem? Book a free data consultation

Connect with me on LinkedIn if you want to chat about a data project.

Like movies? Follow me on Letterboxd

Intro: 19th-Century Brain Rot

For the chronically online, the term “brain rot” is an affliction that defines our present moment: the product of a media ecosystem built on low-effort, hyper-stimulating content. Given the concept’s ever-growing relevance, I assumed its origin was also fairly recent—coined offhandedly by some BuzzFeed writer. But upon further research, I was surprised to learn that humanity’s struggle with “brain rot” dates back to the 19th century.

Published in 1854, Henry David Thoreau’s Walden includes a passage in which the author laments a society-wide softening of the mind. Ever the snob, Thoreau compares the public’s appetite for trivial ideas to an English potato blight: “While England endeavors to cure the potato-rot,” he asks, “will not any endeavor to cure the brain-rot, which prevails so much more widely and fatally?”

One hundred and seventy years later, Oxford named brain rot its 2024 Word of the Year, narrowly beating out demure and romantasy. Well done, Henry David Thoreau. The term’s usage exploded in 2023, as large language models went mainstream and AI-generated content flooded the internet, with the worst of this garbage commonly dismissed as “AI slop.” The brain rot of 1854 concerned cheap literature and carnivalesque entertainment; the brain rot of 2026 is something newer, faster, and unnervingly synthetic. And as AI slop spreads across beloved social media platforms, so too does this metaphorical rot.

So today, we’ll investigate the prevalence of AI slop in modern media and the economic incentives that may help tame its spread.

How Much of the Internet Is AI Slop?

OpenAI launched ChatGPT on November 30, 2022. Prior to this milestone, the world’s understanding of machine intelligence was limited to AOL Instant Messenger bots, Microsoft’s Clippy, and 2001: A Space Odyssey. OpenAI’s major breakthrough was not simply building a powerful large language model, but making the technology accessible through a conversational interface. This overly agreeable bot could answer follow-up questions, admit mistakes, challenge faulty assumptions, and write a limerick about my dad’s favorite baseball player in the style of Dr. Seuss.

ChatGPT adeptly mimicked literature and speech, enabling its users to outsource one of humanity’s most annoying obligations: creating meaning by sequencing words into sentences, and then arranging those sentences into paragraphs.

Four years later, it appears human expression (or, rather, synthetic human expression) should require some additional friction.

According to a longitudinal study from the growth marketing agency Graphite, nearly half of newly published digital content appears to be AI-generated.

This chart comes with two caveats:

1. Graphite advertises itself as an AI-powered SEO agency, so its bleak portrait of content decay also supports its core business. Gross!

2. This graph would probably look different if weighted by web traffic—because human-created work still accounts for an outsized share of what people actually read.

Even if this chart is off by a factor of two, the underlying trend remains alarming.

This chart belongs to a canonical genre of data visualization, which I like to call “ChatGPT launched, and then a line goes up and to the right.”

Take literature, for example, where monthly Amazon e-book releases nearly tripled following ChatGPT’s introduction, a trend that roughly corroborates the Graphite study.

The reach of AI-produced text extends well beyond amateur authors peddling slop or SEO bloggers gaming search algorithms. Academia, a field tasked with expanding the boundaries of human knowledge, has also embraced machine intelligence. One Nature analysis measured the growing ratio of rejected to accepted scientific papers, offering another example of an LLM-induced supply shock: as writing becomes easier, submissions rise, and rejection rates follow. Here again, we see the same basic pattern—a near-tripling of productivity.

While the sheer volume of AI-generated text is concerning, it’s hard to tell how many of these synthetic articles, e-books, or papers are actually being consumed.

The same cannot be said of algorithmic social media feeds, where transitory engagement can snowball into widespread virality. According to a comprehensive study from video production firms Kapwing and NeoMam Studios, around 33% of all YouTube short-form videos qualify as brain rot (nonsensical, low-quality, but not necessarily AI-generated), while 21% are ostensibly AI-created.

This figure likely reflects the prevalence of slop across all short-form video platforms (including TikTok and Instagram), since these services have essentially converged on the same app design via ruthless copying.

At first glance, this study offers a glimpse into a post-truth media doom loop: Users fall prey to AI slop; social media platforms benefit from increased engagement, which, in turn, begets more slop. Ten years from now, every piece of online video content will be entirely synthetic, and digital life will be completely disconnected from our lived reality.

And that may well be true, though I have hope, and this hope comes in the form of an unlikely pro-truth ally: advertisers. Yes, as in the corporate marketers who gave us the GEICO gecko, Jake from State Farm, and Jared from Subway.

According to the same YouTube Shorts study, this nascent economy of video slop-spamming is already quite lucrative, with some channels collecting up to $4M in annual advertising revenue.

How could any of this be a good thing? If people are making millions of dollars from synthetically generated video, won’t that simply lead to more brain rot? Probably.

But these figures also mean that major advertisers like State Farm, GEICO, and Subway have allocated precious ad dollars to videos featuring seven-fingered humans and Komodo Dragons speaking Portuguese. At some point, these advertisers will require their promotions to air alongside non-GMO short-form video content.

A recent Digiday report captured this highly specific marketing anxiety: slop websites are increasingly appearing in programmatic ad networks, placing major brands alongside low-quality, AI-generated spam.

To be clear, I have little sympathy for programmatic ad buyers. But their demand for ROI and positive brand association does create an economic incentive for platforms to verify the humanity of larger accounts. Perhaps corporate marketers will induce some semblance of quality control, if not for consumers, then at least for the advertisers paying to reach them.

Enjoying the article thus far and want more data-centric pop culture content?

Final Thoughts: Atari Slop, Circa 1984

If left unchecked, substandard product quality can do more than damage a single company; it can undermine faith in an entire medium. Consider the video game market of the early 1980s, a fledgling industry that tested the limits of consumer tolerance.

In the 1970s, Atari brought video gaming to the mainstream, starting with arcade games and later transitioning to consumer hardware. Atari’s overnight success opened the door to a chaotic wave of third-party development, as its consoles were flooded with low-effort games from opportunistic publishers like Quaker Oats, Purina, Johnson & Johnson, and Kool-Aid. Soon, the Atari ecosystem became a dumping ground for rushed movie tie-ins, barely functional curiosities, and low-grade cash grabs. It was during this period that the term “shovelware” emerged, which, in many ways, was the 1980s gaming equivalent of brain rot: cheap, low-effort entertainment that eroded trust in the platform.

The most infamous Atari disaster was an E.T. the Extra-Terrestrial game developed in just a few weeks for the holiday season. The game sold so poorly that Atari buried truckloads of unsold inventory in a landfill in New Mexico.

By 1984, the North American console market collapsed, and Atari was sold off by its corporate overseer, setting the company on a path toward dissolution just five years later. Nintendo later revived the gaming industry by implementing strict quality standards for third-party developers, as evidenced by its “Nintendo Seal of Quality.”

Contained within this single parable is both a cautionary tale about the hazards of subpar quality control and a solution to widespread slop-ification. The medium may be different, but the principles of a functional consumer marketplace remain the same—whether the product is video games or social media feeds.

After 35 years of chaotic expansion, the internet still runs on two basic business models: direct payment (through purchases and subscriptions) and advertising. In both cases, low-effort products may flourish briefly, but they rarely win out in the long term. Consumers can only be tricked into purchasing low-grade spam so many times, and advertisers will eventually pull spending if their offerings are marketed alongside shrimp-Jesus hallucinations and an incoherent cat telenovela where the characters speak a nonexistent language.

Which brings us to the concept of Maximum Allowable Slop (MAS), a term I just coined and one that will hopefully be my great contribution to the internet. Platforms like YouTube and Instagram will never fully eliminate AI garbage; the genie is out of the bottle. However, there exists a hypothetical tipping point of slop pervasiveness at which users lose interest, advertisers lose confidence, and platforms begin to lose money. Social media sites will have to do everything they can to stay below that threshold.

To return to Henry David Thoreau’s 170-year-old question, “Will not any endeavor to cure the brain-rot?”—the answer, surprisingly, may be the free market. In the end, TikTok and YouTube will be forced to regulate their own ecosystems to ensure humanity does not exceed the maximum tolerable dose of brain rot. Which sounds like an exceptionally mediocre utopia.

Enjoyed the analysis? Support Stat Significant, and unlock the data behind the insights.

Become a paying subscriber for interactive dashboards, free datasets, and to support independent journalism.

Paying subscribers get:

• The Stat Significant Dataset Hub (Updated Weekly): a collection of free datasets across movies, music, TV, and economics.

• Interactive Data Hubs:

  • Streaming TV: what people are watching on television.

  • Movies: the films watched in theaters and on TV, plus historical trends.

  • Music: the songs, artists, and styles dominating Spotify and TikTok.

  • Social Media: what’s trending on Google, Reddit, TikTok, and YouTube.

  • Wikipedia: the top pages trending on Wikipedia.

Get access with:

• A $6/month Substack subscription

• A $60/year Substack subscription

Upgrade to Paid

*Note: For best results, open this link in a web browser while logged in to your Substack account.

Want to promote your data tool or media publication to Stat Significant’s 25,800 readers? Email daniel@statsignificant.com

Need help with a data problem? Book a free data consultation

Connect with me on LinkedIn if you want to chat about a data project.

Like movies? Follow me on Letterboxd

Intro: Cinema’s Totally-Real Revival

I’ve been writing this newsletter long enough that I can now revisit years-old essays with the benefit of hindsight. This is especially fun when an article ages poorly. Usually, it’s a stray sentence or chart that only I will notice. But every so often, subsequent real-world events render an entire piece meaningless.

And perhaps no essay has proven less predictive than my 2024 article “6 Reasons to Be (Cautiously) Optimistic About Movies in 2025.”

I would not recommend reading this piece, so I’ll provide a quick recap: I assessed the film industry’s post-pandemic woes, detailed six trends that signaled a return to pre-COVID moviegoing, and ended with an examination of the industry mantra “survive till ‘25”—a belief that if Hollywood could make it to 2025, some version of moviegoing equilibrium would be restored.

But 2025 came and went, and no great resurgence occurred. The box office was sluggish, Paramount imploded, Warner Bros. was put up for auction, and lizard king David Zaslav (Warner CEO) netted a cool $800M by running Hollywood’s oldest studio into the ground. The movies were not “so back.”

And now, four months into 2026, it seems like the movies may actually (kinda sorta) be “so back.” So here I am, once again, arguing that a theatrical resurgence is imminent—and once again, I have the data to prove it. This time, I’m right. No question.

So today, we’ll explore early signs of cinema’s totally-real revival, along with the persistent headwinds facing an industry that simply refuses to die—no matter what lizard king David Zaslav and lizard prince David Ellison throw its way.

Today’s newsletter is sponsored by The Stat Significant Dataset Hub

Access 235+ Curated Datasets to Elevate Your Analysis

Want the data behind the deep dives? The Stat Significant Data Hub gives you access to 235+ curated datasets across movies, music, TV, economics, and more.

New datasets are added weekly, with full archive access for Stat Significant paid subscribers. Since last week’s post, we’ve added 13 new datasets—spanning earthquakes, YouTube engagement, a global database of lotteries, and UFC fights.

Check Out The Dataset Hub

1. The Box Office Is Up (Finally) and Less Franchise-Dependent

Box office is an imperfect metric for evaluating the health of the film industry, just as GDP is an imperfect measure of human flourishing. But it’s the best metric we have for assessing the health of a capital-intensive business—and by that measure, things are moving in the right direction.

Year-to-date domestic box office is up after the down years of 2024 and 2025, which were dampened by protracted writers’ and actors’ strikes.

Better yet, of the 12 highest-grossing movies of 2026, 8 are unaffiliated with an existing franchise.

It pains me to include Michael among the standout “original” films on this list because:

1. It’s a music biopic, which is the lowest form of non-franchise filmmaking.

2. Who the subject is.

3. There will be a sequel about the second part of his life, which will probably skip some important sections of his Wikipedia page.

By the rules I’ve created, Michael technically counts.

This list will likely change over the coming months as the summer season delivers its usual wave of franchise entertainment. But for now, it offers a nice change of pace and a little hope.

2. More Movies Are Being Released in Theaters

So far, 2026 has delivered the most theatrical releases of the post-pandemic era, outpacing every post-COVID year by at least 13%. The industry is adapting to shifting moviegoer tastes and learning how to sell films outside the Marvel universe, with recent box office successes in horror, animation, Colleen Hoover adaptations, other smut-adjacent literary phenomena, and video game movies. The result is a faster pace of content development and a broader definition of what works in theaters.

Meanwhile, television production has slowed. English-language series output has declined every year since its 2022 peak, and the buzzy “prestige” shows that once drove subscriber growth have become far less plentiful.

To add insult to injury, streamer subscription prices have skyrocketed just as content output has declined.

Streaming’s golden age is most likely over, and audiences seem increasingly willing to return to theaters.

3. Online Interest in Movies Is Rising

This next point is somewhat abstract, so bear with me. You probably noticed that I cited streaming data as evidence of a moviegoing resurgence, suggesting a zero-sum struggle between film and television. And while this concept is rather squishy, I do believe this tug-of-war exists, at least to some extent.

In a crowded attention economy, TikTok, YouTube, AMC theaters, Netflix, and other platforms all compete for our time. To reduce this phenomenon to a single corporate buzzword, these pastimes are battling for mindshare.

So, in this hypothetical mind-battle between film and television, which medium occupies a greater portion of the collective consciousness? To quantify the zeitgeist, I looked at Wikipedia traffic—specifically, the share of movie- and TV-related pages that appear among the site’s 100 most-searched entries.

Over the past five years, film pages have captured an increasing share of Wikipedia search traffic, while TV’s share has grown more slowly.

One explanation for TV’s relative decline is that there are simply more TV series than there were 10 years ago, making it harder for any one show to break through. Or perhaps movies have always been better suited to capturing collective attention, while TV caters to fragmented niches. In the streaming era, hardly anyone watches the same thing at the same time, except for The Pitt and the NFL.

4. Netflix Is Warming Up to Theatrical Releases

You read that correctly: Netflix is giving Greta Gerwig’s upcoming Narnia adaptation a legitimate theatrical release. The film will play in theaters for at least 45 days before moving to streaming, and unlike a typical Netflix big-screen release, Gerwig’s movie will play in easy-to-find multiplexes at normal showtimes (rather than a single 12-seat theater on the outskirts of town at 1 p.m. on a Tuesday).

So why is Netflix backtracking after years of shunning brick-and-mortar exhibition? A few reasons:

1. Money: Netflix is a public company tasked with satisfying Wall Street’s demand for infinite revenue growth. Getting into theatrical exhibition is a net new revenue stream that will grow faster than its legacy subscription business.

2. Insider knowledge of Warner Bros. finances: For a few days in 2025, Netflix successfully purchased Warner Brothers, before lizard prince David Ellison usurped their bid using Saudi oil money. During this strange liminal period when they didn’t-not-own Warner Brothers, Netflix got a look into the finances of theatrical exhibition, and presumably liked what they saw. Before losing the acquisition to Paramount, Netflix promised to release Warner Brothers movies in theaters and touted the strength of the theatrical business model.

3. Improving filmmaking craft: A “Netflix movie” has become shorthand for subpar filmmaking. The streamer has excelled at replicating the low end of the market, such as Hallmark movies and middling romcoms, but has minted few breakthrough hits beyond the occasional KPop Demon Hunters. Arguably, the biggest driver of this filmmaking deficit is directorial talent.

   

   Established filmmakers like Greta Gerwig want to see their movies on the big screen and have avoided the service to secure theatrical commitments. The upcoming Narnia adaptation is the first time that the streamer has acquiesced to a director’s demand, and there is reason to believe that it may not be the last.

Enjoying the article thus far and want more data-centric pop culture content?

5. The Oscars Have a Path to Cultural Relevancy (in 2029)!

As of this moment, the Oscars are down bad. Viewership has fallen precipitously, and the broadcast remains tied to a legacy cable model that makes the ceremony hard to watch for the average cord-cutter.

But that’s about to change! I think.

This past year, the Academy announced that the Oscars would move to YouTube in 2029, a far more accessible platform than your local ABC affiliate. YouTube is the sneaky winner of the streaming wars, capturing the largest share of TV viewing time by Nielsen estimates—and those figures don’t even include desktop viewing.

For a culturally diminished awards show, the move offers a much-needed lifeline: a chance to reinsert itself into the cultural conversation.

And then there’s gambling, which I don't approve of but which may prove beneficial to Oscar viewership. According to recent reports, the 2026 Oscars generated more than $120 million in bets on prediction markets like Kalshi and Polymarket, and that number will likely grow as these platforms see increased adoption.

Is it bad that some people will find their way into movie culture through gambling? Absolutely. Has sports betting helped pro leagues like the NBA and NFL deepen fan engagement? Also yes. The Academy Awards have always possessed a horse-race quality, and in a world where people will gamble on almost anything, someone will inevitably commoditize this metaphorical horse race.

6. The Physical Media Revival Is Upon Us

In writing this newsletter, I have continually returned to the topic of physical media because every time I revisit the data, I find another proof point for its continued resurgence. Take this latest example: for the first time in 20 years, search traffic for DVD players has started to grow.

Obviously, this is still orders of magnitude removed from the DVD’s heyday, but it represents a minor reclamation for film nerds.

This revival is driven by demand for a more tangible movie culture—one where fans can own the things they love.

Boutique labels like Vinegar Syndrome, Arrow Video, and—most prominently—Criterion have turned Blu-ray collecting into a hobby for a certain kind of film obsessive, and rising search traffic suggests these brands are finding a wider audience.

DVDs will never return to the $16 billion market they were in 2008, but they can remain a thriving niche for cinephiles seeking cultural ownership.

Final Thoughts: Karma

When I was nine years old, I won an iPod in a school raffle. Being nine, I didn’t know much about MP3 players, Steve Jobs, or the sentimental value of physical media. I owned two CDs—Barenaked Ladies and Smash Mouth—which I played once a day. That was the full extent of my music-listening life.

iTunes expanded my cultural horizons to something bordering on infinite. I could sample nearly every song in existence for 30 seconds at a time, buy the tracks I liked, and carry them around in my pocket. Suddenly, my Barenaked Ladies CD was a relic—a reminder of the “before time,” when you had to buy 12 songs to hear a single hit track.

Music had migrated online, turning the industry’s remaining physical spaces into artifacts from a pre-digital world. Every Friday, my mom and I would go to the mall, and on these trips we developed a strange little pastime: marveling at the ever-decreasing foot traffic inside Tower Records. It became a recurring game between the two of us: “Does Tower Records still exist?” Each trip we’d bet on its continued presence, and each trip we’d marvel at its protracted demise.

At the time, I never considered the subculture that might disappear with the record store, or the people who would lose one of the places they loved most. Decades later, I’m on the other side of this equation, where the game is something akin to “Does film culture still exist?” It’s not a perfect comparison, but the broader anxiety is familiar. Moviegoing is no longer at the center of culture, and the rituals around it are increasingly fragile.

For now, film culture persists, sustained by a handful of benevolent upstarts who may, or may not, be on the precipice of enshittification: Letterboxd, A24, The Criterion Collection, Mubi, etc. Yet this remains a delicate balance. Each year feels like a referendum on the continued existence of a certain subculture: one that believes cinematic craft can be commercially viable. And as long as that’s the case, I’ll keep writing these articles, like a violinist going down with the Titanic.

A broken clock is right twice a day, so at some point I’ll be rewarded for my optimism—even if just for a little while. That is, until the lizards of Hollywood muck things up beyond recognition.

Enjoyed the analysis? Support Stat Significant, and unlock the data behind the insights.

Become a paying subscriber for interactive dashboards, free datasets, and to support independent journalism.

Paying subscribers get:

• The Stat Significant Dataset Hub (Updated Weekly): a collection of free datasets across movies, music, TV, and economics.

• Interactive Data Hubs:

  • Streaming TV: what people are watching on television.

  • Movies: the films watched in theaters and on TV, plus historical trends.

  • Music: the songs, artists, and styles dominating Spotify and TikTok.

  • Social Media: what’s trending on Google, Reddit, TikTok, and YouTube.

  • Wikipedia: the top pages trending on Wikipedia.

Get access with:

• A $6/month Substack subscription

• A $60/year Substack subscription

Upgrade to Paid

*Note: For best results, open this link in a web browser while logged in to your Substack account.

Want to promote your data tool or media publication to Stat Significant’s 25,800 readers? Email daniel@statsignificant.com

Need help with a data problem? Book a free data consultation

Connect with me on LinkedIn if you want to chat about a data project.

Like movies? Follow me on Letterboxd

Become a paying Stat Significant subscriber for interactive dashboards, free datasets, and to support independent journalism.

Paying subscribers get:

• The Public Dataset Directory (Updated Weekly): a curated directory of public datasets from around the web.

• The Stat Significant Data Catalog: Downloadable datasets assembled, cleaned, and updated weekly by Stat Significant, covering movies, TV, music, podcasts, and online trends.

• Interactive Dashboards: interactive charts, updated daily, with the underlying data available for download.

  • What People Watch:

    • Streaming TV: real-time tracking of the shows dominating streaming platforms.

    • Movies: box office performance, streaming popularity, audience ratings, and historical film trends.

  • What People Listen to and Read:

    • Spotify & TikTok: the songs, artists, and styles dominating Spotify and TikTok.

    • Pop Music History: a deep dive into music history, Billboard chart performance, genre trends, and song composition.

    • Podcasts: the podcast shows, episodes, and categories capturing listener attention.

    • Book Bestseller Trends: The books and authors topping the bestseller lists.

  • What People Search:

    • Wikipedia Traffic: the topics capturing global attention, as told by Wikipedia.

    • Online Trends & Social Media: what’s trending on Google, Reddit, TikTok, and YouTube.

    • Prediction Markets: the cultural events, celebrities, and internet moments people are betting on.

Get access with:

• A $6/month Substack subscription

• A $60/year Substack subscription

Upgrade to Paid

*Note: For best results, open this link in a web browser while logged in to your Substack account.

Stat Significant’s Podcast Report 🎙️

This page offers a real-time view of which podcast shows, episodes, and categories are capturing listener attention. The interactive charts and tables below highlight the episodes with the most staying power, the podcasts ranking highest on the Top 100 charts, and the genres driving the biggest share of podcast consumption.

All visuals are updated regularly, with new data added daily and weekly. When available, click “Get Data” at the bottom of a chart to download the underlying dataset.

Because many of the graphics below are interactive, this page is best viewed on the web rather than in an email client or the Substack app.

Table of Contents

For paid subscribers — click a link below to jump straight to that section:

1. Section 1: The Top Podcast Shows and Episodes

   1. Top Podcast Episodes: Last 14 Days

   2. Top Podcasts Over The Last 14 Days

   3. Top Podcasts By Episode “Stickiness”

2. Section 2: Top Podcast Categories

   1. Weekly Share of Top 100 Podcast Episodes by Category

   2. Weekly Share of Top 100 Podcast Shows by Category

Section 1: The Top Podcast Shows and Episodes

Top Podcast Episodes: Last 14 Days

• What This Chart Shows: The podcast episodes that spent the most time in Apple Podcasts’ Top 100 over the past 14 days.

• Data Source: Apple Podcasts API

• Data Update Frequency: Daily || Data Can Be Downloaded: ✅

Top Podcasts Over The Last 14 Days

• What This Chart Shows: The podcasts with the best average ranking on Apple Podcasts’ Top 100 chart over the past two weeks.

• Data Source: Apple Podcasts API

• Data Update Frequency: Daily || Data Can Be Downloaded: ✅

Read more

Intro: “If It’s in the Top 10, It Must Be Good.”

When choosing a TV show or movie, I tend to make the process harder than it needs to be. I’m burdened by the irrational belief that I should always be watching the best possible thing, which inevitably leads to endless research on Letterboxd and Metacritic. Quite frequently, I spend so much time searching for the perfect program that I end up watching nothing.

My wife, however, is unburdened by this self-imposed standard. She’ll gladly watch whatever Netflix places in front of her: Trainwreck: Poop Cruise, Squid Game: The Challenge, Too Hot to Handle, and other streaming curiosities. When I ask how she ultimately chose, say, a 6-part crime docuseries about the Amish, she usually offers the same rationale: it was in the Netflix Top 10. And when I tell her that this movie or show looks bad, she’ll shrug and say something like, “If it’s in the Top 10, it must be good.”

The conversation usually ends there because I can’t argue with the vague credibility of Netflix’s top ten titles—until now.

So today, we’ll examine whether Netflix’s Top 10 actually signals quality, how these hallowed content options stack up against other services, and whether you’re better off choosing a program at random.

Should You Trust the Netflix Top 10?

Digital marketplaces like Netflix, DoorDash, and Amazon often struggle with problems of abundance. Given a near-limitless catalog of restaurants or movies, consumers can suffer from analysis paralysis.

To address this champagne problem, tech companies employ hundreds of workers dedicated to “search and discovery.” The mandate of these teams is decidedly straightforward: get people to watch, click, or buy more stuff. Over time, these companies have converged on a familiar toolkit: personalized recommendations, improved search results, optimized thumbnails, and other subtle nudges designed to reduce friction.

In February 2020, Netflix added a new tool to this playbook: a daily list of the most popular shows and movies on its platform, displayed directly inside the app. The feature quickly became a hit, generating ongoing media coverage of its 10 trending titles and prompting other streaming services to create their own Top 10 lists.

But helping people choose faster is not the same as helping them choose better. So does the Netflix Top 10 actually lead viewers to better recommendations? Or does it simply make popular titles feel more worthwhile?

To answer one of streaming’s great conundrums, I pulled every Netflix Top 10 entry since the feature’s inception, along with its corresponding user rating on TMDB. By this measure, the average TV series trending in the Top 10 is approximately 1% more enjoyable than a non-trending Netflix title, yet still scores below the average series available on Paramount+, HBO, Disney+, or Hulu.

Within Netflix’s own ecosystem, the Top 10 offers a modestly useful signal of TV show quality. But once you compare Netflix against the broader streaming landscape, its trending titles are no better than an HBO or Paramount+ series chosen at random.

That said, this calculation only covers television. What about da’ movies—a business Netflix seems determined to drive into the ground?

Using the same methodology, we find that the average Netflix Top 10 movie is about 0.5% worse than a Netflix film that never earns this distinction. In fact, Netflix’s Top 10 films rank among the weakest content bundles in streaming, outperforming only Peacock, Prime Video, and AMC+.

Sometimes data can be beautiful, and in this case, quite amusing. For movies, landing in Netflix’s Top 10 seems to carry the cultural prestige of a middling Rotten Tomatoes score. Yikes.

This finding hints at an unflattering truth: Netflix heavily promotes its Originals across the platform, muscling them into the Top 10—and most of the streamer’s movies are quite bad.

Between 60% and 80% of Netflix’s Top 10 titles are Netflix Originals, meaning the chart effectively showcases the streamer’s best work. By that measure, Netflix’s output is getting worse: over the past six years, the average trending show has become less acclaimed, while its trending movies have remained consistently below average.

My best guess is that this pattern reflects Netflix’s pullback in spending as it pivots from growth to profitability, a process internet theorist Cory Doctorow has dubbed “enshittification.” The Top 10 no longer signals the best of streaming so much as the easiest content to manufacture: dating shows, celebrity documentaries, cooking contests, and numerous docuseries about Ted Bundy or the Menendez brothers.

Yet there’s also another possibility: maybe Netflix is still producing prestige films and shows, but they’re underrepresented on its best-of charts. So is the Top 10 actively boosting lower-quality titles at the expense of acclaimed programming? To test this, I flipped my analysis: instead of asking whether Top 10 titles are well regarded, I asked whether highly rated titles tend to perform better on the Top 10 charts.

By this measure, the more acclaimed a show or movie is, the worse it ranks within the Top 10.

Taken to its logical extreme, you’re better off choosing a show situated at #9 than one sitting at #2. If that conclusion sounds implausible, it should: our metric is being distorted by a major confounding variable—time.

Netflix’s biggest hits—Stranger Things, Adolescence, and Wednesday—all make the Top 10 and then stay there for weeks. As of this writing, KPop Demon Hunters has remained within the Top 10 for 46 straight weeks and currently sits at the #9 spot.

Which means that perhaps the most valuable indicator of content quality is not whether a show reaches the Top 10 or where it sits on the chart, but rather how long it remains there.

By this marker, the relationship between program acclaim and platform visibility is clear: higher-quality shows stay on the charts for longer.

Here lies a bizarre quirk of the Netflix Top 10 as a search-and-discovery tool: arguably its most useful piece of information—the amount of time a title persists on the chart—is not displayed in the app. Instead, viewers are shown a ranking that reflects a single moment in time, making the Top 10 less a marker of quality than a shortcut for decision-making.

Functionally speaking, the feature is no different from a promo code on Booking.com or an average star rating on Amazon: it reduces the friction of content discovery, making it easier to pick something and press play.

Enjoying the article thus far and want more data-centric pop culture content?

Final Thoughts: All The Good Movies

When I was 11, I stumbled upon a TV special chronicling the American Film Institute’s 100 Greatest Movies. Enthralled, I watched it again the following week, printed out the full list, and vowed to watch every title before I turned 13. This proved somewhat problematic, since it meant seeing A Clockwork Orange and Silence of the Lambs well before my teenage years. But I persevered nonetheless. On a random Thursday, you’d find me watching Citizen Kane or One Flew Over the Cuckoo’s Nest. I was a pretty cool 12-year-old.

After working through the age-appropriate titles on AFI’s list, I moved on to Roger Ebert’s Great Movies book series, which eventually led me to critics like Pauline Kael and Richard Roeper. There was an endless archive of movies to discover, and these critics—with their well-argued recommendations—became guides in my quest to see All The Good Movies.

By today’s standards, however, this discovery process seems byzantine. You’d read the opinions of a few professional critics, ask friends what they thought, and then make an active effort to track the movie down. So much effort!

When Netflix arrived, it promised something more dynamic: personalized recommendations, tailored thumbnails, and a vast library organized around your preferences. It was Roger Ebert, AFI, IMDb, and Letterboxd folded into one product.

And yet people still struggled to find something to watch. They always have, and I guess they always will. It was at this point that some product manager figured they could increase platform conversion if they created a “trending” section, and thus the Netflix Top 10 was born.

This new feature promoted populism at the expense of personalization. All you really know is that other people are watching that thing—never mind if those people have completely different tastes. The end result is an amorphous recommendations tool that appears democratic, while remaining extremely easy for Netflix to shape.

But is there anything wrong with that? Depends on what you want from television.

Best I can tell, there is little difference between picking a show at random and choosing from the Netflix Top 10. If your goal is simply to watch something and pass the time, then great—this feature gets you there faster.

But if you believe, as I do, that you should always be watching the best possible thing, then “because it’s in the Top 10” is not much of a justification.

Enjoyed the analysis? Support Stat Significant, and unlock the data behind the insights.

Become a paying subscriber for interactive dashboards, free datasets, and to support independent journalism.

Paying subscribers get:

• The Stat Significant Dataset Hub (Updated Weekly): a collection of free datasets across movies, music, TV, and economics.

• Interactive Data Hubs:

  • Streaming TV: what people are watching on television.

  • Movies: the films watched in theaters and on TV, plus historical trends.

  • Music: the songs, artists, and styles dominating Spotify and TikTok.

  • Social Media: what’s trending on Google, Reddit, TikTok, and YouTube.

  • Wikipedia: the top pages trending on Wikipedia.

Get access with:

• A $6/month Substack subscription

• A $72/year Substack subscription

Upgrade to Paid

*Note: For best results, open this link in a web browser while logged in to your Substack account.

Want to promote your data tool or media publication to Stat Significant’s 25,800 readers? Email daniel@statsignificant.com

Need help with a data problem? Book a free data consultation

Connect with me on LinkedIn if you want to chat about a data project.

Like movies? Follow me on Letterboxd

Intro: The Imprecision of Genre

Technically speaking, the films Dumb and Dumber, White Chicks, Deuce Bigalow: European Gigolo, Groundhog Day, Airplane!, and Dr. Strangelove all qualify as “comedies.” Taken in one sitting, however, they would make for an incomprehensible movie marathon. These films have almost nothing in common tonally or intellectually, except for one basic ambition: to make people laugh.

Such is the strange imprecision of genre. A label like “comedy” or “thriller” isn’t a strict definition so much as it is a signal—shorthand for a bundle of storytelling tropes that audiences recognize and actively seek out. A rom-com promises the thrill of a meet-cute and final kiss, a horror film provides scares in a controlled setting, and a musical offers spectacle and sensory pleasure. These categorizations serve as a heuristic, guiding moviegoers toward a desired emotional experience.

So what does it mean when the world’s oldest and most prolific genre enters a period of sustained decline?

For years, Film Twitter and Letterboxd devotees have lamented the death of the movie comedy. According to these online critics, laughter has been canceled, replaced by quippy Marvel slop and self-serious Vin Diesel car garbage. Yet while building a recent dashboard, I noticed an even sharper decline among dramas—one that significantly outpaces the demise of movie comedies.

After confirming that this dip was, indeed, legit, I set off on a new quest: to uncover the dramatic collapse of the studio drama.

So today, we’ll examine the case for and against the commercial viability of dramatic films, uncover the storytelling formats that have taken their place, and identify where the genre has found a new home.

The Case for Drama Films

The 1990s marked the peak of the modestly budgeted adult drama. Indie darling Good Will Hunting grossed $225 million on a $10 million budget, Sex, Lies, and Videotape earned $36 million on a $1.2 million budget, and The Brothers McMullen brought in $10 million on a budget of roughly $25,000. Distributors regularly scooped up these films at festivals like Sundance and successfully marketed them to moviegoers seeking small-scale, character-driven stories.

Which brings us to the first commercial advantage of the drama format: these films are inexpensive to produce.

Adjusted for inflation, dramas rank among the lowest-cost genres, surpassed only by ultra-low-budget horror.

The appeal of these films is not epic scale or intricately choreographed action. In the case of Good Will Hunting, it’s the charm of a janitor solving difficult math problems. In 1997, that concept drew thousands of people from their homes.

Good Will Hunting was eventually nominated for nine Oscars, winning two, and continues to rank among the top 250 films on sites like IMDb and Letterboxd.

Which brings us to the next artistic advantage of the movie drama: audiences tend to like them. The genre regularly outperforms other formats in online user ratings, and historically, 60% of Best Picture nominees have been primarily classified as dramas.

If you’re in search of prestige, because who needs money, then a low-budget drama is your best bet.

The genre’s precipitous decline does not square with its reputation. In a survey of 65,000 consumers that asked which genres they “want to see more of,” dramas consistently landed in the middle of the pack across all age groups. That’s not a sign of breakout demand, but it hardly suggests a format worthy of wholesale abandonment.

So, if these films are inexpensive, well-liked by viewers, and remain reasonably popular with general audiences, why are they disappearing from movie theaters and streaming services?

The Case Against Dramas Films

Over the last decade, studios have gravitated toward “genre filmmaking.” This label is confusing—because every film belongs to a genre—but in practice it’s shorthand for everything outside traditional comedies and dramas: action, horror, sci-fi, even rom-coms.

Historically, these “genre” categories were treated as niche, with limited audience potential. But not anymore.

When we look at the formats gaining ground in theaters and on streaming, the list is dominated by quintessential “genre” categories—horror, thriller, and sci-fi.

“Genre” films have long offered greater commercial upside, delivering a return on investment that far exceeds that of the prototypical drama, which ranks near the bottom of all formats.

And it gets worse. The comparison above is a relative percentage calculation, while studios ultimately care about absolute dollars. By that measure, dramas have the lowest box office ceiling, consistently grossing less than any other major format. Making $15M on a $10M budget sounds good on paper (50% ROI!!), but grossing $400M on a $225M budget will get you promoted ($175M ROI!).

A major driver of these commercial limitations is international box office. Dramas and comedies are dialogue-heavy and rely on cultural nuance that doesn’t always translate well abroad. As a result, these formats earn the majority of their box office revenue domestically.

You know what does translate well across all languages and continents? A gigantic dinosaur. Which is why we have seven Jurassic Park films and only one Manchester by the Sea or Requiem for a Dream.

In fact, among all sequels produced over the last 50 years, dramas consistently rank among the least represented—which is great because I don’t need films like Hamnet or The Brutalist to become a franchise.

Part of what makes dramas effective is their sense of finality. I love the final scene of The Social Network because it perfectly captures Mark Zuckerberg’s self-imposed isolation: he has created the world’s largest online community, yet remains profoundly alone, waiting for his ex-girlfriend to accept his friend request. Great ending! Story over, right? Nope! Unfortunately, a sequel called The Social Reckoning is set for release in late 2026. I will be obligated to see this movie, and it will almost certainly leave me bummed. Can’t wait!

And while The Social Network has aged well enough to justify a sequel 16 years later, most dramas tend to lose their cultural standing over time.

To be honest, I don’t have a perfect explanation for this next finding. I’m mostly beholden to the data, and the data suggests that dramas do not age well.

When we compare a movie’s average online rating across two separate periods—1995 to 2004 vs. 2018 to 2023—we find that comedy and drama films see a marked decline in appraisal over time.

My best guess is that these formats are tightly bound to contemporary taste. When I was a kid, I would watch Turner Classic Movies and try to appreciate films from the 1940s, only to find the exercise strangely difficult. I could admire them—in theory—but I struggled to experience these stories the way their original audiences did.

I feel similarly about many Oscar-bait dramas of the 1990s, including but not limited to: Chocolat, American Beauty, Shakespeare in Love, Scent of a Woman, The English Patient, and Life Is Beautiful. I simply don’t understand what contemporary audiences saw in these films.

Action and horror, meanwhile, have visceral elements that translate across generations: big dinosaurs, jump scares, campy set pieces, and other straightforward pleasures. The first ten minutes of Raiders of the Lost Ark are timeless and feature almost no dialogue.

So if dramas age poorly, earn less, and rarely spawn sequels, do those limitations also carry over to other mediums? Are TV showrunners and playwrights abandoning the form as well? Not at all. If anything, theatrical dramas are simply migrating elsewhere.

Perhaps the most significant driver of the studio drama’s decline is the rise of prestige television. The format has exploded on streaming, where dramas now account for ~30% of all new series, well ahead of any other category.

Think of the most acclaimed dramatic series from the past five years—Adolescence, Beef, Mare of Easttown, Task. With a few tweaks, each could just as easily be a feature-length film. The increasingly blurry line between TV and film often comes down to a concept no one can quite define: theatricality.

A “theatrical” story is one that justifies leaving home for an overpriced ticket and tub of popcorn. Small-scale dramas rarely clear that bar anymore, and streamers like Netflix have capitalized on the gap. In the process, the genre’s perceived commercial viability has collapsed.

Enjoying the article thus far and want more data-centric pop culture content?

Final Thoughts: Little Sads

During this year’s Oscars ceremony, host Conan O’Brien did an extended bit about “Little Sads,” his tongue-in-cheek term for short documentary films that tend to leave audiences “a little bit sad.” As a former film student, I found the label pretty accurate, though it applies just as neatly to plenty of feature-length dramas. Hamnet, Sentimental Value, Train Dreams, Nickel Boys, Past Lives, CODA, The Banshees of Inisherin—all of these films did, in fact, make me a little sad.

For the first 80 years of feature filmmaking, these “Little Sads” had a reliable theatrical audience. If you had asked me two decades ago, I would have assumed demand for the genre was evergreen. Then I went to the epicenter of Little Sadness: the Sundance Film Festival.

Once a hub of American movie culture, Sundance struck me as a film festival on its last legs. Most of the movies headlining the program had not been acquired for distribution, and the general moviegoing atmosphere was oddly subdued.

Then there were the movies themselves. Screening after screening, my dad and I sat through one Little Sad after another, each smaller and more melancholic than the last. By the final day, I couldn’t stomach another story about a fractured family or a middle-aged man processing his emotions. So we skipped our last Little Sad and saw a horror movie instead.

One of my biggest pet peeves is when people invoke the word objective in matters of taste. And yet—objectively—the experience of seeing that horror film in a packed theater was 800 times more enjoyable than our other screenings. Objectively!

Halfway through the movie, a jump scare caused someone in the audience to scream at the top of their lungs, prompting a burst of communal laughter from the rest of the theater.

If you asked which of these films I’d pay $17 to see in theaters, plus another $4 for Cookie Dough Bites, I’d pick the lone horror movie and nothing else. The rest I’d skip—or perhaps invoke one of the most dreaded phrases in modern entertainment: “I’ll wait for streaming.”

Am I drawing sweeping conclusions from a single film festival experience? Absolutely. But my takeaway seems to align with broader audience behavior. On the other hand, do I want theatrical dramas to go extinct? Not at all.

Some of my best movie-watching experiences involve being a little sad while seeing Lady Bird, Past Lives, Do the Right Thing, Boyhood, and Aftersun, among others. But at some point, audiences—myself included—decided this particular brand of sadness no longer needed to be experienced exclusively in movie theaters. We could be a little sad on the subway, at the gym, or while folding laundry, with this melancholy now spread across six to eight installments and multiple seasons.

Those seeking small doses of sadness can simply peruse the Netflix Top 10—or, on occasion, head to theaters for the next installment of The Social Network franchise.

Enjoyed the analysis? Support Stat Significant, and unlock the data behind the insights.

Become a paying subscriber for interactive dashboards, free datasets, and to support independent journalism.

Paying subscribers get:

• The Stat Significant Dataset Hub (Updated Weekly): a collection of free datasets across movies, music, TV, and economics.

• Interactive Data Hubs:

  • Streaming TV: what people are watching on television.

  • Movies: the films watched in theaters and on TV, plus historical trends.

  • Music: the songs, artists, and styles dominating Spotify and TikTok.

  • Social Media: what’s trending on Google, Reddit, TikTok, and YouTube.

  • Wikipedia: the top pages trending on Wikipedia.

Get access with:

• A $6/month Substack subscription

• A $72/year Substack subscription

Upgrade to Paid

*Note: For best results, open this link in a web browser while logged in to your Substack account.

Want to promote your data tool or media publication to Stat Significant’s 25,200 readers? Email daniel@statsignificant.com

Need help with a data problem? Book a free data consultation

Connect with me on LinkedIn if you want to chat about a data project.

Like movies? Follow me on Letterboxd