Who is the Most Important Marvel Movie Character?A Data Analysis of the MCU (or Why Iron Man May be Worth His Millions)Michael TaubergBlockedUnblockFollowFollowingMay 21Joe and Anthony Russo have acknowledged that it was Robert Downey Jr.
’s Iron Man who set the tone for all Marvel films.
After all, it was Downey who started and ended the Avengers saga with those same snarky words — “I am Iron Man”.
Given that Avengers Endgame has grossed almost 2 billion dollars worldwide, and that the other Marvel Movies also generated billions, I wondered, how much of this success is owed to Iron Man?To answer that question I’ve written code to parse all of the Marvel Cinematic Universe Movie scripts.
Although I was unable to get complete scripts for Dr Strange and Captain Marvel, I was still able to collect detailed data on 19 of the 21 Marvel movies.
The screenplay files and the computer scripts used to process them are available for others to look at here on github.
Armed with this raw data, I think it’s possible to see just how much of that superhero gold is due to Iron Man’s influence.
Actor SalariesThe easiest way to see how much value Tony Stark brings to the Marvel Cinematic Universe (MCU) is to see how much he gets paid.
While exact number are hard to come by, the Hollywood Reporter and other sources provide some estimates as to how much each actor in Marvel movies takes home.
I’ve focussed on just the biggest paycheck that each actor has received for a film thus far.
Results are shown below.
In terms of raw dollars, Robert Downey Jr.
is the biggest winner by far.
It’s rumored that he gets a percentage of the box office revenue of his films.
As such, he is estimated to have made $50 million for Avengers Infinity War and $75 million for Avengers Endgame.
In the next sections, we’ll see if he really earned it.
Movies RolesOne simple way to measure which character is the most important in the MCU is to see which one has been in the most movies.
If we plot the main Marvel characters in order of film appearances, we get the following.
When it comes to movie appearances, the technical winner is Steve Rogers (aka Captain America).
Of course, Cap only edges out the other Avengers because he has a brief post-credits cameo in Ant-Man.
If we ignore this one scene, then Thor, Captain America, Iron Man, and Black Widow all tie for the most film outings (with 8 credits each).
Nick Fury is just behind with 7 appearances.
This makes sense as he basically formed the Avengers.
So, in terms of movie roles, Iron Man doesn’t seem any more important than the other Avengers.
Lines of DialogueTo dig deeper, we can look not just at the number of movies a character is in, but also how many lines of dialogue they have.
This is a good proxy for screen-time and tells us how important a character is to the story.
Aggregating all available Marvel scripts and plotting the total lines of dialogue per character, the following picture emerges.
Here we see that Tony Stark has a wide lead.
Far behind him in a distant second and third are Thor and Steve Rogers who are also central to the MCU.
If we look at the total number of words spoken by a Marvel character instead of just lines of dialogue, then Tony Stark wins again.
Interestingly loquatious Loki now comes in fourth, while soft-spoken Steve Rogers falls well behind Thor.
Network AnalysisIt’s possible to argue that a character’s prominence on screen may not reflect their importance to the story.
Some character may be integral to the MCU without having many lines (I’m looking at you Groot!).
To get at the importance of the main Marvel characters, I decided to use some simple network analysis techniques.
Using the python NetworkX library I converted movie characters into nodes in a graph.
Then, I and added edges to represent character interactions onscreen.
So for example, if a line of dialogue from Thor follows a line of dialogue from Bruce Banner (the Hulk), we know that these nodes should be connected.
Using this approach, I was able to construct the following graph of major MCU characters.
Network Graph of the Marvel Universe based on Character InteractionsWe can see that that the original 6 Avengers are the heart of the Marvel Universe and form the center of the graph.
The outer sections relate to other “worlds” within the MCU.
These include the Ant-Man world, the Guardians of the Galaxy cluster, Dr Strange’s network and close by, Asgard.
Peter Parker also show up adjacent to the Avenger network and next to Iron Man characters.
This makes it clear that he is being positioned to replace Tony Stark as the heart of the MCU.
So, it’s clear that the Avengers form the core of the marvel universe, but which of them is the most central?.One basic measure of centrality is to count how many edges a node has.
By plotting the number of connections that each character has, we can get a measure of how connected they are in the graph.
Here again, Tony Stark is the clear winner.
In fact, Iron Man is direclty connected to almost every other Marvel character.
For completeness, we can look at other measures of network centrality.
These include betweenness centrality, eigenvector centrality, and closeness centrality.
Betweenness measures how often a node acts as a bridge to other nodes.
Eigenvector centrality attaches a numerical value to a node’s influence in the network.
And, closeness centrality represents the shortest path between a node and all other nodes in the network.
Results of these measures are below.
We can see that in terms of “betweeness”, Thor comes out on to.
This makes sense since he is directly connected to the Avengers, Captain Marvel, Steven Strange, and all Guardians of the Galaxy.
He’s a one man Bifrost between the worlds in the MCU.
In eigenvector centrality, Iron Man comes out on top, suggesting that his node is the most influential in the network.
Finally, in terms of closeness, Tony Stark again wins, since he is closely connected to most other characters.
ConclusionsBased on my analysis, Tony Stark, Thor and Steve Rogers are the obvious heart of the Marvel Cinematic Universe.
They are the most central to the MCU network, have the most dialogue and are thus paid the most.
Of course, Iron Man gets even more than his fellow Avengers.
Given that by most measures he is the most connected, the most influential and the most prominent character, I think it’s fair to say that the actor who plays him deserves his outlandish pay.
Bonus GraphsSentimentUsing the R sentimentR library, I calculated the average sentiment of the dialogue associated with each character.
It seems that most of the Avengers are pretty upbeat (unlike their DC superhero counterparts).
In fact, Peppily named Peter Parker and Pepper Potts come out on top in terms of dialogue valence.
The Guardians of the Galaxy on the other hand, especially Gamora, Rocket Racoon and Drax are considerably less positive.
Right in the middle is Nick Fury.
This befits a character who had both the paranoia to organize against intergalactic threats, and also the optimism to believe that they could be vanquishedAvengers WordcloudsFor fun I also made word clouds for all of the 6 core Avengers.
See if you can guess which cloud goes with which character.
Word Clouds for Iron Man, Thor and Captain AmericaWord Clouds for the Hulk, Black Widow, and Hawkeye (Clint Barton).. More details