Thursday, March 04, 2010

The VFX Predictinator, Part 4

Just joining us? Please read "The VFX Predictinator Part 1", "Part 2" and "Part 3."

Finally, it's time to answer the two burning questions that result from the creation of the 100% accurate-for-20-years VFX Predictinator. First, can the Predictinator be tweaked and used to predict other categories? And second, which film will bring home Oscar for visual effects this year according to the Predictinator?

Could this formula, or at least the ideas behind the formula, work for predicting other categories? In fact, I think this formula, with some tweaking, could work very well in the other 'craftsman'-type categories, particularly the sound and sound effects editing Oscar, since expensive, well-made, and popular films are usually nominated in the sound categories.

However, I cannot say the same for acting categories, writing or directing, or even cinematography and editing. There is no way to predict the randomness of a breakthrough performance or a particular film. I mean, Marisa Tomei won an Oscar for a Joe Pesci comedy. Who could have predicted that? And a non-winning "American Idol" contestant (Jennifer Hudson) won an Oscar for "Dreamgirls" in her very first film performance. What quantifiable data could possibly support that prediction? On the same wavelength, who could have predicted the awards success of 2004's "Crash," directed by Paul Haggis? Certainly when a filmmaker like Eastwood, Spielberg or Scorsese is in the running, their quantifiable chances certainly improve, since they have a significant body of work that can be numerically tracked (for acclaim, box office, etc). There simply exist far too many chances for breakthrough Academy Award victories, which makes numerically predicting them virtually impossible.

And now, it's time to reveal the Predictinator values of the 2009 contenders for the Academy Award for visual effects. And the nominees are "Avatar," "District 9," and "Star Trek." Let's see what the Predictinator thinks:

As you can see, "Avatar" ended up with a Predictinator score of 8.03, with "District 9" geetting 6.36, and "Star Trek" earning 4.61.

The 2009 films share, unlike most years, near-universal love from the critics. All three films earned over 82% on the Tomatometer, which hasn't happened since 1995, when both "Babe" and "Apollo 13" earned rave reviews. But that's where the similarities between this year's three nominees end. The box office tally was overwhelmingly dominated by the cultural phenomenon that was "Avatar." It also earned heavy points in the Month of Release criteria (December) and additional Oscar nominations (9 total).

"District 9" puts up a good fight against "Avatar" on two significant fronts, since the films both shared organic characters as their primary visual effects, and both featured facial acting amongst those characters. But the other important criteria gives "Avatar" a Predictinator-predicted victory.

Of course, we'll revisit the Predictinator's success after the Oscars on March 7, and see if its streak of accuracy can be continued into its 21st consecutive year. And here it is.


Blake Tucker said...

Do an experimental predictination on the sound or editing categories! DO IT. COME ON, DO IT!

Akira said...

The VFX Predictinator works!!!