Not all festivals charge a submission fee but most do. And are priced at more than $100 per submission. I looked at the cost of entering 4,631 festivals to find the average.

How long is too long for a festival film? When I interviewed 500+ film festival directors a few years back, the most consistent feedback was blunt: "Please, please, please make your films shorter."

YouGov turned my recent research on moral lessons in movies into a poll of US adults to see how those values land in real life. Some things that stood out to me... (Thread) @yougov.co.uk @today.yougov.com

How much control do a few major agencies really have over Hollywood’s directors? I tracked 1,000 top directors and linked each one to their current rep. The results are… lopsided.

Half of the directors behind the top 1,250 grossing films of the past 25 years went to film school and half did not. I looked at what happened next. Do film school graduates make better films? Are their films more profitable? Which schools have the most successful alumni?

I’ve just finished analysing the backgrounds of 682 directors behind 1,250 top-grossing films since 2000. What percentage do you think went to film school?

I researched who represents the 2,000 most famous actors to see where the power lies. Four companies represent more than half of top actors: Creative Artists Agency, United Talent Agency, The Gersh Agency, and William Morris Endeavor... (1/2)

Why do so few women become directors in Hollywood or partners in Silicon Valley? (Thread)

I wanted to do a data-led test into whether film and movie are interchangeable. The data shows they signal very different things about genre, quality and intent.

Happy 70 Year Anniversary to @guinnessworldrecords.com! I'm celebrating by having a dig around the archives to look for some movie facts around the number 7️⃣0️⃣. (Thread)

What counts as ‘great’ cinema in 2025? The Godfather Part II was once ranked 11th in British Film Institute (BFI)'s Sight and Sound's 'The Greatest Films of All Time' project, but in the latest poll didn’t even make the top 100.

What's the most common colour on movie posters? I analysed 58,687 movie posters from the past century to see which colours dominate, how their use has shifted over time and what each hue signals to audiences.

I just read a study which used data on 434 movies within 173 series released between 1998 and 2021, focusing on the effect of casting more actors of the global majority in leading roles before and after the emergence of the Black Lives Matter movement. Here's what they found. (Thread)

Part of the reason we enjoy movies is that we get to see ideas and situations played out to people who are not us. We can then learn from what they did, what they didn't do, and how it worked out for them. (Ted Hope talks about movies being "design for living", which sums it up perfectly).

I love the new Naked Gun (although it should have been called Son of A Gun). It made me think of the research Bruce Nash and I conducted for the American Film Market a few years ago. stephenfollows.com/p/patterns-among-successful-comedy-movies

It's out! The new @guinnessworldrecords.com book has been released and is packed with all manner of weird and wonderful records about movies* *(They tell me they cover other topics across its 250+ pages but I can say without any bias that the movie records are the best records of all the records).

Its so frustrating when names and faces on posters don't match up. And it seems to be everywhere - but how common is it really? I studied 4,000+ posters to find out.

There are loads of academic papers which do some form of prediction around movie economics. I've picked out a few I think are worth the read. First up, this one from 2020. (Thread)

The data from over 10,000 films across fifty years shows that some sub-genres actually perform best with lower budgets. Spending more can weaken the qualities audiences value most in these genres.

What even IS a superhero movie? Is 'Wicked' a superhero movie? What about 'Mortal Kombat' or 'Avatar'?

"Bad" movies can still make a lot of money. I looked at the detailed financials of almost 5,000 movies to find the ones which made at least 200% Producers Net Profit AND got a Metascore of under 50 out of 100.

Netflix has filed hundreds of trademarks, and they are quite revealing. Why does Netflix need a trademark for “Strong Black Lead”? What’s the long game behind “Netflix Bites”? And why didn’t they lock down “Netflix and Chill”?

Here's a great study by Thomas J. Billard from 2019 which tested how viewing habits affect prejudice. (Thread)

Avatar 5 was announced in Dec 2009 - over 8,000 days before current expected release date. I analysed 11,560 other movie release date announcements to see how typical that is.

Filmmakers dream of their film becoming a classic, a cult favourite, or part of the canon. No one dreams of it becoming a tax write-off. I took a look at why studios sometimes cancel completed films and why they’re forbidden from ever releasing them to the public.

This study proved what we already know (i.e. that sequels make money!) but it also revealed that the timing and length of the franchise matter more than we might think. They studied 167 films to measure what drives box office success in sequels. Here's what they found ↓

The busiest two days in the global film festival calendar aren’t when you might think. I gathered data on 50 major festivals, going back 25 years, to map the shape of the annual calendar. stephenfollows.com/p/what-is-the-busiest-day-for-film-festivals

The release of Eddington made me think back to this research I published a few years ago. It tracked the genre preferences of 117,000 fiction films produced in 151 countries over a 20 year period.

On a recent trip to the cinema I had to watch a run of really long trailers. This got me thinking - have movie trailers always been this long, or am I just becoming a grumpy old man?

Here's a fascinating study into patterns among successful movies. Researchers analysed 425 produced screenplays and matched them to box office results. They tracked characters, dialogue, pacing and emotional tone to see how well each script predicted commercial success. This is what they found...

Are movies with plot twists better than those without? I studied the plots of 26,710 movies to see what the data says.

This topic has come up twice in separate conversations this week so I think it's worth repeating. The vast majority of films do not have a long tail. Some do but most will earn the majority of their total lifetime income in the first few years of release.

Close-ups make us care... but only up to a point. This study tested how the number of close-ups affects how viewers describe a character’s inner life. Same film, different versions, each with a different number of face shots. Here's what they found... (Thread)

Here are the ten things cinemas want from the film industry... (Thread)

Movies with more positive messages are more likely to be profitable. Genres like sci-fi, adventure and comedy showed the strongest link between positivity and financial success. Drama, romance and fantasy were less consistent.

Germany makes wildly successful comedies and family films, but (unless you live in Germany) you’ve probably never seen them. Here's why German films look so different at home and abroad: stephenfollows.com/p/why-german-films-look-so-different-at-home-and-abroad

A strong network helps… but only if you're in the right role. This study analysed 2,110 films to see how a filmmaker’s professional network effects box office. It looked at over 15,000 people and mapped their collaboration ties across a decade of films. They found ↓

Most exhibitors agree that the next wave of cinema-goers is less loyal to tradition and more focused on the overall experience. But the industry is still figuring out how to adapt.

What do you prefer: 'cinematographer' or 'director of photography'?

What percentage of movies have a plot twist? I analysed 26,710 movies and found 984 that have a major twist (that's 3.7% of movies). But before you start reading, know that this fully spoils the endings to 59 great movies. You’ve been warned.

I've not seen M3GAN 2.0 yet but its upcoming release made me think of this great academic paper. Thread ↓

Can cinemas survive the next 20 years? Some exhibition execs aren't so sure. Screendollars and I spoke to over 200 of them about recovery, realities and the future: stephenfollows.com/p/can-cinemas-survive-the-next-20-years

28 Years Later has been unleashed! If you're wondering how the horror genre has evolved since the Rage Virus first hit, then here are a few relevant pages from my recent Horror Movie Report. (Thread)

People keep asking if AI will change filmmaking. It already has. Not the video generation stuff — I mean the plethora of "AI-powered" products and services already being pitched to filmmakers. Many fundamentally change jobs as we know it (some for the better, some for the worse IMO).

Stars sell tickets… but not all stars are equal. This study looked at 1,545 US films, and combined that with 61 earlier studies, to settle the question: do stars actually drive box office? Here's what they found ↓

Subgenres aren't equal partnerships. I analysed the dialogue in 65,000 films to see what happens when genres combine. Here's what came up: Sci-fi tends to overpower everything. Drama rarely dominates anything. And horror plays nicely... except when it doesn’t.

Sex is disappearing from movies. Why? My theory is that it has something to do with shifting audience tastes, greater awareness of actor consent, and the availability of spicy content elsewhere.

Cinema box offices are recovering... but not evenly. Recovery is connected to habit, affordability and the strength of national film cultures.

This study tested how new viewing habits change the way people respond to narrative messages. Researchers ran a field experiment with 273 viewers of East Los High, a Hulu series that built storylines around alcohol, abortion and immigration. Here's what they found ↓

Which movie genres share the same fans? I analysed 32 million film ratings to map out which genres attract the same audiences. This reveals ⤵️