Friday, January 04, 2008

Predicting the Visual Effects Oscar, Part 1

When I was much younger, I naively thought that the honor of winning an Academy Award was synonymous with being 'the best.' I even thought of the honor as equivalent to having the highest batting average in Major League Baseball for a particular year. There was no doubt that in 1984, the best batter in baseball was Tony Gwynn with a .351 average. In that same year, there was no doubt that "Amadeus" was the best film made that year... because it won "Best Picture" at the Academy Awards... right?

Of course, since then, my eyes have been opened and my outlook matured. Film is an art form, and by definition, not quantifiable. And any attempt to quantify the subjective nature of the appreciation of art is a ridiculous endeavor.

Unfortunately, the Academy Awards are still looked upon as a horse race, as if winning the award will somehow magically launch one's career into the stratosphere, and 'losing' the award should cause one great heartache and agony. The frenzy around awards season will, as it always does, reach a fever pitch, as we argue who 'should' win the Oscar, who 'deserves' it for their performance, or who 'deserves' it for their body of work.

Which reminds me of a pre-"The Departed" Onion article about Martin Scorsese's lack of Academy Awards...
Achieving the honor of an Academy Award nomination is an important milestone for a film, or an individual. In the visual effects world, like other branches of the Academy, getting a nomination for Achievement in Visual Effects is highly significant, because the nominees are determined by the branch itself, meaning visual effects professionals have decided that your film deserves the recognition of a nomination. Being honored by ones peers is quite meaningful. Although politics certainly come to play, the sometimes overwhelming force that is general popularity is diminished, and the art and innovation play a significant role.

As we all know, the entire Academy votes for the winner, and the force of popularity comes back into play. And since the Academy membership largely consists of retired actors, you can be assured that the film with the hottest buzz, or the one with the most aggressive ad campaign will win.

For branches of filmmaking such as visual effects, cinematography, sound design, editing, and the other non-acting branches, one should look at the nomination as a wonderful acknowledgment of your hard work and talents, and an actual statuette as the cherry on top of an already fulfilling chocolate sundae.

I have written about this before on this blog; Industrial Light & Magic got a bum rap for 'losing' the Academy awards for eleven years straight (after winning 14 visual effects awards in 18 years) . Of course, its achievements in those eleven years were extraordinary, and ILM (like Martin Scorsese) had nothing for which to be ashamed.

Over the years, I have come to believe that the winner of the Academy Award can be predicted fairly well, even without seeing the nominated films (much like the Academy voters, themselves). Anyone who has been to one of my Oscar parties knows that I'm a pretty good predictor of who will take home a statuette. It really has nothing to do with the most amazing cinematography, the most realistic visual effects, the most moving score, or the gutsiest performance. It all comes down to how these retired actors vote.
However, I've always wondered if there exists a statistical relationship between the winner of the visual effects Academy Award and some sort of quantifiable criteria which, of course, informs the Academy's voting. After heated debates with co-workers, I decided to hunker down and do some real statistical analysis. Using the quantifiable criteria of North American box office receipts and acclaim by film critics, I wondered if one could accurately predict which nominee wins the award. And, most importantly, which is the more accurate predictor of the visual effects Oscar - critical acclaim or box office success?

In conducting my research, I've laid down these statements:
  • Popularity for a film would be rated by gross box office revenue in North America, as a proxy for ticket sales. Data was provided by Box Office Mojo.
  • Critical acclaim for a film would be rated by the Rotten Tomatoes' Tomatometer rating, a website that tracks accredited film critics' reviews of Hollywood movies. The percentage rating indicates the percentage of approved critics who recommend a certain movie. More info on the Tomatometer can be found here.
The added benefit of using Tomatometer data to represent critical acclaim is the fact that is is measured in a percentage, which helps us fairly compare critical acclaim between years.

And now, some notes on the research and data:
  • The statistical dataset begins in 1984 and runs to 2006, because I wanted to track modern visual effects films. In addition, the Tomatometer ratings, although still accurate, pull from a smaller set of critics as one goes backwards in time.
  • Even though I'm using box office revenue as a proxy for ticket sales, we don't need to do any messy conversions to account for inflation, since we're only interested in intra-year comparisons.
  • However, even though we won't be comparing inter-year box office, we will need to compare the relative size of box office margins, to help compare box office between years. (more on that later)
  • Academy Award nominees and winners come from the website Visual Effects Headquarters, which was a pretty darn good website, if I do say so myself.
  • Unfortunately, there exist three instances where the data is skewed between 1984-2006. Two occur in 1987 and 1995, when, due to a rare voting anomaly amongst the visual effects branch, only two films received nominations ("Predator" and Innerspace"; and "Babe" and "Apollo 13", respectively), instead of the usual three. In 1990, after the votes were tabulated, only one film qualified to receive a nomination ("Total Recall"). After the 1995 incident, the voting rules were changed to better guarantee three nominees.
Of course, it's utterly ridiculous that the Academy doesn't allow five nominees for Achievement in Visual Effects. With the amount of quality work being produced these days, there are certainly enough quality films to fill five measly nomination slots. For example, the competition was so fierce in 2003, neither of "The Matrix" sequels made the long list of seven films that presented their work at the bake-off.

Does this all sound exciting? It better, because in Part 2, you'll be inundated with charts galore. Finally, in Part 3, we'll summarize the data and see if we can answer the question - is there a way to predict the winner of the visual effects Oscar?

No comments: