The success of comic book movies: an interactive chart.

Blade movie poster. Stephen Norrigton (1998).

Blade movie poster. Stephen Norrigton (1998).

There is little question that comic book movies have become one of the pillars of contemporary Hollywood. About 150 films adapting comics characters have been produced in the USA since Richard Donner’s Superman in 1978, with a notable concentration in the past 20 years, following the success of Blade (Norrigton, 1998) or X-Men (Singer, 2000) depending on which precursor you want to emphasize. These films have also become the focus of sustained of sustained academic interest in the past few years, with books such as Liam Burke’s The Comic Book Film Adaptation and Drew Morton’s Panel to the Screen, to name but two recent examples.

In teaching a class on the contemporary comic book films1 I was led to produce a visual representation of this corpus, or more accurately, several representations. The aim was to prompt the students to think about the relationship between critical and commercial success of these films. What started as a single slide quickly became a more interesting project, which led to a series of usable charts. This article contains a description of the results, an explanation of the methodological problems encountered in the course of this work and finally a short history of the process which led from the slide to this article.

 

Domestic budget multiplier/Imdb rating

Domestic budget multiplier/Imdb rating

On this chart, the green lines indicate the median values for the chosen datasets: Imdb ratings (all films – median value 6,6/10) and budget multipliers (all the film referenced at TheNumbers.com – median value 0,9). The comic book films are clearly located in a majority in the most favorable quadrant : the top-right one, with a few notable critical and commercial outliers. Another key finding here is the very imperfect correlation between critical appraisal and commercial success. While the movies do tend to cluster around the expected diagonal line, the dispersion is significant. It is worth noting that other researchers have found even starker gaps between critical and commercial success, even suggesting there might be a negative correlation between the two factors!

A more interesting and teachable version of this mapping appears below :

Budget/rating, with labels - Clicking on the image takes you to an interactive version

Budget/rating, with labels – Clicking on the image takes you to an interactive version

This version makes it possible to identify the films – though you may want to click and visit the interactive version to do that, and perhaps choose a more legible version, which excludes the extremes – and it also distinguishes between pre-Blade movies (in red) and post-Blade ones (in blue). This visual distribution makes it clear that pre-Blade films are clustered on the left of the chart, with quite a few of them in the less desirable quadrant; thus confirming the validity of the chosen breakpoint. It also calls our attention to spectacular pre-Blade outliers, such as The Mask (Russell, 1994), Teenage Mutant Ninja Turtles (Barron, 1990) and Batman (Burton, 1989). The budget multiplier for these films put to shame the domestic results of such blockbusters as The Dark Knight (Nolan, 2008) and The Avengers (Whedon, 2008), though the net gross of these two films easily surpass those of their predecessors.

These results call our attention to some of the methodological choices underlying these charts:

  • budget multiplier: this is an unusual choice, in that the scholarly work I am aware of relies most often on net results. This simply divides the box-office results by the production budget, to generate a multiplying factor. It quantifies success not in terms of absolute results, but as the potential for a film to exceed expectations; it rewards breakaway successes more than careful bets. It is also interesting in that it is not affected by inflation. One possible issue: it factors in production budgets, figures provided by the studios themselves, which can be underestimated for PR purposes.
  • Domestic, not worldwide gross : a crucial choice, as contemporary superhero films increasingly play to an international audience; the choice to focus on domestic gross reflected my intent to adequately compare films over 40 years on a steady basis. A version of the chart taking into accounts worldwide results can be found here, in which Deadpool stands out among recent films ; one of the problems with worldwide budget multipliers is that production budgets do not change when a film is released abroad, but publicity budgets – which I do not take into account – do. In other words, a film may actually lose money by being distributed abroad if its communication budget overtakes its local gross, but its budget multiplier will still go up.
  • Imdb ratings as a proxy for « critical » success : this is really a measure of audience reactions. While correlated to critical evaluations, they can differ significantly in a number of cases, such as

 

A quick note on what led to this interactive chart

While preparing a class dedicated to contemporary comic book films (using, among other things, Liam Burke’s Comic book film adaptation: exploring modern hollywood’s leading genre), I tried to come up with a convincing representation of the variety of profiles these films possess. I came up with my trusted tool, the semiotic square

Critical/commercial success (manual)

Critical/commercial success (manual)

In this representation, the top right quadrant includes films which were both critically and commercially successful. Scott Pilgrim vs. the World (Wright, 2010) is an example of a critically succesful but financially problematic film, while X-Men Origins: Wolverine (Hood, 2009) presents the opposite profile. Catwoman (Pitoff, 2004) is a fine example of a film which failed on both levels.

This was not meant as a carefully constructed chart – Powerpoint is not the right tool to do that – rather as a general guideline, based on my current readings and my memories of the critical reception of these films. As such, it misrepresents some publicly available data (Cow-boys vs. Aliens should have been much further on the left of the graph) and fails to answer basic questions, such as the nature of the central point: is there an example of a film falling squarely in the middle of that graph? Maybe the original Blade (Norrington, 1998)?

Of course, the chart was meant as non-exhaustive inventory, meant to present the students with a useful tool and to suggest the existence of well-known examples in each quadrant, though a majority of comic book film adaptations appear on the right-side of the graph (hence the role of franchises and the popularity of the genre2. Still, it was not entirely satisfying.

I took to Twitter to ask about a way to reconstruct that graph using an actual dataset rather than just a few examples, and colleagues from the Université Bordeaux Montaigne library pointed out that Imdb makes it possible to access its data through downloadable data sets : http://www.imdb.com/interfaces

I used this dataset, the list of « comic book movies » provided by Boxofficemojo (minus a few international releases with limited distribution) and the budgets indicated on The Numbers to assemble my own dataset. Unsuprisingly, it did not work out that way:

  • Movie titles are not consistent from one database to the other: X2: X-Men United become X-Men 2 or X-Men 2: X-Men United, for instance. The increasing number of superhero movies with overlong titles does not help
  • Movie titles are often reused. Imdb makes it necessary first to remove TV shows and video games, but even then, manual verifications are needed to filter out homonymous movies. Steel, for instance, is both a horrendous Shaquille O’Neal superhero vehicle from 1997 and a well-liked 1979 movie about steel workers.
  • Excel does not know how to add labels to dot-cluster graph: that was the biggest surprise of all. Fortunately, a move to Google docs solved that problem.

 

  1. As Blair Davis has shown, there was a prior intense period of interplay between movies and comics, in the 30s to 50s, but that was the subject of a previous class.

    Davis, Blair. Movie comics: page to screen/screen to page. New Brunswick, New Jersey: Rutgers University Press, 2017. Print.

    []

  2. Burke convincingly argues that in spite of their apparent variety, comic book films from Ghost World (Zwigoff, 2001) to The Avengers (Whedon, 2010) are often marketed and received a belonging to a single genre []

Vous aimerez aussi...

6 réponses

  1. Marion Wesely dit :

    Dear Hypotheses blogger,
    We found your article particularly interesting. To increase its visibility so the community can more easily appreciate it, we made it a headline article on the hypotheses.org slider.
    Best regards,
    The Hypotheses.org team

  2. Cristiano F. C. dit :

    « Imdb ratings as a proxy for « critical » success (…) »
    What? Why not Metacritic or Rotten Tomatoes?

    • Nicolas Labarre dit :

      That’s a great question, of course.

      The short answer is that I plan to use both sets of numbers, time permitting.

      The longer answer is threefold:
      – the Imdb data was readily available, not so much for Rotten Tomato and Metacritic
      – I am not familiar enough with Rottentomatoes to be confident in their weighted average, and I distrust their main RT score
      – I wasn’t sure the data would be coherent enough for older films (Howard the Duck is missing from Metacritic, for instance)

      In short, I went with data I was familiar with, estimating that Imdb provided a valid if incomplete entry point into the issue of critical reception. I suspect using RT as well will not drastically change the picture, but I may be entirely wrong about that.

      NL

      • Cristiano F Cangucu dit :

        Hey,
        thanks for the answer.

        « The longer answer is threefold:
        – the Imdb data was readily available, not so much for Rotten Tomato and Metacritic
        – I am not familiar enough with Rottentomatoes to be confident in their weighted average, and I distrust their main RT score
        – I wasn’t sure the data would be coherent enough for older films (Howard the Duck is missing from Metacritic, for instance) »

        Fair enough.

        « In short, I went with data I was familiar with, estimating that Imdb provided a valid if incomplete entry point into the issue of critical reception »

        Well, that’s where we still disagree. IMDB is possibly the best proxy for viewers’ opinion, yes, but there is a nontrivial disparity between audience and critics appraise of a given film (compare audience score vs critical score on Rotten Tomatoes).

        That’s why I still think using IMDB ratings to measure critical success is not the best option, when there are two alternatives designed for the problem at hand. IMDB ratings are, of course, valuable to estimate audience satisfaction.

  1. 15/04/2017

    […] The success of comic book movies: an interactive chart Nicolas Labarre, Picturing it!, 7 avril 2017 > http://picturing.hypotheses.org/477 […]

Répondre à Nicolas Labarre Annuler la réponse.

Votre adresse de messagerie ne sera pas publiée. Les champs obligatoires sont indiqués avec *