Skip to Content
Humans and technology

Why Avengers: Infinity War was so successful—according to … ecology

Mathematical tools developed to analyze interactions in natural ecosystems help explain which films in the Marvel Universe make the most money.
A movie still from Marvel's Avengers: Infinity War
A movie still from Marvel's Avengers: Infinity WarMarvel Studios' AVENGERS: INFINITY WAR ©Marvel Studios 2019

The Marvel Cinematic Universe is arguably the most successful film franchise in history. It is based on the characters from the Marvel comic book franchise and the shared universe they inhabit. In the last decade, these stories have been told in more than 20 movies so far. And together they have generated well over $6 billion in revenue.

What makes the Marvel Cinematic Universe so successful? There are clearly many factors, but an important one is the nature of the characters and complexity of the interactions between them. That’s difficult to study because there have been few recognized tools for characterizing these relationships.

Today that looks set to change thanks to the work of Matthew Roughan and colleagues from the University of Adelaide in Australia. They point out that ecologists have long had the mathematical tools to study the interaction between species. Now these guys have used the same tools to provide a unique insight into the Marvel Cinematic Universe.

Marvel Cinematic Universe

First some background. One of the main problems for ecologists is to determine the roles of different species in the complex web of interactions that exist in any ecosystem. A simple reckoning of the number of species does little to reveal this complexity. Indeed, it can mask it if the counting process somehow misses important actors or inflates the value of others.

Instead, ecologists focus on the interactions between species. The number of interactions can then be used to determine the importance of each species using well-known mathematical tools based on Shannon entropy. These essentially determine the information content of each interaction as a proportion of the whole.

A similar problem exists in analyzing a film universe. At first blush, the size of the cast must be equal to the actors listed in the credits, but there is no standard for creating such a list. Some films list only the main speaking parts, while others include every extra. For example, the 2003 film Closer and the 2004 film Stuck on You have similar numbers of meaningful characters, and yet the former has a cast list of six while the latter credits 328 actors.

The Marvel Cinematic Universe has a similar problem, particularly when big-name celebrities have cameo appearances. For example, Marvel creator Stan Lee has a cameo in every film but does not generally contribute much to the plot.

So Roughan and co develop a different way to assess meaningful characters. An important element of any plot is the conflicts each character has with other characters. The researchers assume that the number of these conflicts is a measure of the character’s importance.

They also assess the number of lines each character has as a proportion of the total script. However, they consider this a less reliable indicator of importance, since some characters can be loquacious or reticent in a way that does not reflect their significance.

So Roughan and co simply count the conflicts between characters and then study the Marvel Cinematic Universe using the same mathematical tools developed by ecologists.  That allows them to study the films from an entirely new perspective.

One thing they examine is the relationship between the number of conflicts in a movie and the total amount of dialogue. This immediately identifies films that are driven by dialogue rather than conflict (and vice versa).

By this measure, so-called “origin” movies stand out because they tend to use dialogue to develop the character from zero to hero. Spider-Man and Captain America: The First Avenger are good examples. By contrast, sequels and team-up movies are often driven by conflict rather than dialogue.

Another discovery is that the cast size is strongly correlated with a movie’s profitability. The only film in the series to take less revenue than it cost to make is The Incredible Hulk, and this also has the fewest characters. By contrast, the film with the largest cast size—Avengers: Infinity War—took in twice as much revenue as it cost to make.

That seems to imply that audiences prefer larger cast sizes. But Roughan and co caution against this assumption. Instead, they point out that movies with larger casts tend to be team-ups. These attract fans of several different origin movies, which is why they are more profitable.

An interesting exception is Guardians of the Galaxy, which is ostensibly a team-up movie but lacks the usual prequel origin films. Because of this, it could not have drawn in multiple existing fan bases.

That is reflected in its profitability. “Although the movie was widely praised, it wasn’t a profit generating engine compared to several other team-up movies,” say Roughan and co.

That’s interesting work that pioneers a new kind of analysis for movies and has other applications, too. “The presented metric and its generalisation could be explored in relation to other cinematic universes, or TV shows,” Roughan and co. It could also be used for recommender systems to show viewers movies with similar features to the ones they’ve previously enjoyed.

Either way, the mathematical tools developed by ecologists look set to have a fruitful future in the movie world.

Ref: : How The Avengers Assemble: Ecological Modelling Of Effective Cast Sizes For Movies

Deep Dive

Humans and technology

Human-plus-AI solutions mitigate security threats

With the right human oversight, emerging technologies like artificial intelligence can help keep business and customer data secure

Merging physical and digital tools to build resilient supply chains

Using unique product identifiers and universal standards in the supply chain journey, the whole enterprise can unlock extended value

Unlocking the value of supply chain data across industries

How global standards and unique identifiers are turning supply chain data into a game-changer

Transformation requires companywide engagement

Employees need to be heard for leaders to overcome the hurdles of organizational change

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at with a list of newsletters you’d like to receive.