dcsimg
 

A brief analysis of superhero movies made in the past few decades.

Well, it is quite another superhero movie filled summer for Hollywood. Marvel Comics, owner of popular superheroes such as Spider-Man and Iron Man, is producing movies for each character that will appear in the upcoming 2013 vehicle entitled ‘The Avengers.’ DC Comics, most famous for their characters Batman and Superman, is trying to get more of their relatively unknown characters onto the big screen to match the filmography of its aforementioned competitor.  With the recent flop of DC’s Green Lantern and modest box office earnings of Marvel’s Thor, a few critics have mentioned the possibility of audiences getting tired of superhero flicks. Could Marvel be pushing their limits by trying to get the audience acclimated to all of those superheroes that will eventually appear in the Avengers? Should DC continue to focus on Batman and Superman, and wait until audiences recover from superhero burnout before releasing any other DC characters to the public?

Let’s look at the history of how well 46 superhero movies have fared across the last three decades.  After compiling data from www.imdb.com and www.rottentomatoes.com, I used Minitab's Scatterplot (Graph-->Scatterplot) to obtain the plot below.

Scatterplot1

DC Comics was the first to release big budget superhero movies, with the first four all being related to Superman. After 1998, however, Marvel Comics starts making their movies with consistency, and end up almost doubling the count of DC movies by 2011. As you can see, these movies clearly make a lot of money from ticket sales.  However, no Marvel movie has been able to reach that $500 million mark since Spider-Man, set back in 2003. DC’s lone two bright spots are Superman from 1978 and the Dark Knight from 2008.

Let’s look at the earners over the $300 million mark as well as the earners below $100 million.

By clicking on ‘Brush Mode’, I can highlight a group of points on my Scatterplot. I then go to Editor->Set ID variables, and choose what information I want shown for each brushed point. I decided to add the movie score as an ID variable as well. These movie scores are actually percentages of positive reviews out of all total number of reviews for a particular movie. These reviews are stored on rottentomatoes.com. Here is the end result:

brushing image

The graph on the left depicts earners over $300 million. Audiences must love Spider-Man. Even with a score of 63, Spider-Man 3 earned $365.5 million. For movies listed in the second table, the majority of them have low scores. The exceptions are Hell-Boy and Kick Ass.

Let’s see how the score relates to the earnings by going back to the Scatterplot. I can also add different regression fits by clicking on the Scatterplot after it’s created, and going to Add->Regression Fit.

Scatterplot

Upon visual inspection, it looks like earnings go up as movies obtain higher scores. I then decided to create 3 different fitted line plots (Stat->Regression->Fitted Line Plot). These plots essentially show the same information in the graph above, but with more statistics involving regression. R2 is one of these regression analysis statistics, and it is defined as the percentage of response variable variation (in this case, Earnings) that is explained by its relationship with one or more predictor variables (Score). In general, the higher the R2, the better the model fits your data. Unfortunately, R2 naturally increases as you add more predictors, which is going to happen when you have a quadratic or cubic model. Don’t worry folks! We can then use R2 (Adjusted) which compensates for the addition of extra predictors. After looking at the R2 (Adjusted) for linear, quadratic, and cubic models, it looks like cubic provides the best fit with value of 50.1%. The other 49.9% is defined as variation that is unexplained by the model.

From the graphs above, we see that domestic earnings are statistically affected by Score. From the Time Series Plot and its brushing capabilities, we notice that the higher grossing movies are left to 4 popular characters and their sequels. We’ll focus more on this in the next blog entry. We’ll also look at movie budgets and compare them to domestic earnings, which will help in answering whether the movies continue to remain profitable.

Stay tuned, same blog time, same blog channel.

*Adjusted for Inflation
**There is point on the Scatterplot that is missing for 1989’s The Punisher, with Dolph Lundgren. The movie was not released in the United States.

 

Fearless Data Analysis

Minitab 17 gives you the confidence you need to improve quality.

Download the Free Trial


 

Comments

blog comments powered by Disqus