When it comes to Oscar predictions, it might be helpful to consider some overlooked data

On Sunday night, the Academy Awards will finally put an end to the annual flurry of predictions that start each year when Oscar nominations are announced: who will win best picture?

But, despite people's best efforts, Oscar predictions are often a crapshoot. There are all sorts of different data points to reference. Two of the most accurate predictors of a win are the DGA awards (won this year by Birdman) and the total number of Oscar nominations the film has received. (This year's nomination totals are led by The Grand Budapest Hotel and Birdman, both with nine.)

Publicity and politics are also factors. Crash, for instance, won in 2006 based largely on its advertising campaign and on the refusal of various Oscar voters to consider its main competition, Brokeback Mountain, which many considered Hollywood's first mainstream gay love story.

there may be some clues in the historical trends of the Academy's decisionS

And, then for Jerry Vermanen, a data journalist for Dutch news site NU.nl and self-described "movie freak," there may be some clues in the historical trends of the Academy's decisions. Two years ago, he began compiling all the information he could on every Best Picture winner throughout Oscar history and turned it into a series of informative charts.

Although Vermanen cautions that "data on previous winners is no guarantee for the future," there do seem to be some meaningful patterns.

Here's what the Academy tends to like for Best Picture films and how that could influence this year's winner:

1) Release date

Is Selma's December release good or bad for its Oscar prospects? (Paramount Pictures)

As Vox's own Todd VanDerWerff has pointed out, some have argued that a December release date for Selma could account for its underwhelming nominations total (it received just two). Although the six nominations earned by American Sniper, which was released the same day as Selma, push against that theory.

In fact, films released in December have historically nabbed the most nominations and wins. According to Vermanen's data, about one out of five Oscar winners have had a December release date:

Oscar Best Picture winners by month. (Jerry Vermanen, Who won the Oscars?)

Vermanen also found that films released in October and November traditionally fare well in the Best Picture category and that November has the highest conversion rate from nominees to winners.

Of the eight films nominated for the 2015 Oscar for Best Picture, six were released between October and December. The only film released in November was The Imitation Game, which could give it an edge. And the outliers are The Grand Budapest Hotel, which was released in March, and Boyhood, released in July.

Not one film with a release date in July has ever taken home Best Picture. Ever. That makes sense when you think about the kinds of movies studios tend to promote during July: big-budget, action-packed blockbusters. As Vermanen says, "Summer movies aren't great Oscar movies." Though these films perform well at the box office, they usually aren't Oscar-worthy. But Boyhood is a considerable exception to the rule.

films released in December have historically nabbed the most wins

Also noteworthy is that February, March, and April aren't great release months if you want your film to nab an Oscar nomination or win. The last film released in February to be nominated for Best Picture was 1987's Hannah and Her Sisters. For February it was 2001's Erin Brockovich, and for March it was the 1990 smash hit Field of Dreams.

So why do movies from months later in the year do so well?

There are two ways of looking at this, says Vermanen. On one hand, maybe it's the case that most of the year's best films are released toward the end of that year.

On the other hand, he says, "maybe the Oscar judges have a short memory." That is, maybe they see really great films throughout the year, but tend to only remember the ones they see at the end of the eligibility period, which ends December 31. To be eligible for a nomination, a film must play the required one-week qualifying run in Los Angeles before the year's end.

2) Length

The Theory of Everything is one of the Best Picture nominees that falls into the expected length of a winner. (Focus Features)

As Vermanen notes, "The Academy really, really likes longer films." According to his analysis, the 10-year running average time for a winning movie has never fallen shorter than two hours. (To be nominated for Best Picture, a film has to be at least 40 minutes.)

And the pictures with the best shot of winning are between two and two-and-a-half hours. The three films that don't fall into this range are the exceptionally long Boyhood (165 minutes), The Grand Budapest Hotel (100), and Whiplash (106).

3) Genre

American Sniper is the right genre to have a good shot at winning. (Warner Bros.)

This should come as no surprise, but the vast majority of Best Picture winners are dramas, as Vermanen's data illustrates. This could work out well for American Sniper and Boyhood.

Comedies don't fare as well with the Academy and, on average, account for anywhere from 15 to 25 percent of the nominees (In the past two decades, only three have won: The Artist, Chicago, and Shakespeare in Love.) Thrillers and horror movies perform even worse.

Other trends? Movies about war typically do well, and Vermanen's data shows a notable and understandable winning streak for these kinds of pictures in the 1940s. Given that the Academy enjoys a good war flick, American Sniper just might win. It's worth noting, however, that Eastwood wasn't nominated for Best Director, and it's exceedingly rare for a movie to win Best Picture without a complementary Best Director nomination.

Which do you think will win the 2015 Oscar for Best Picture?

You can use either Vermanen's visualizations or a similar interactive at Bloomberg to help you come up with your guesses. Yes, there are lots of other factors to consider when making your picks. Still, these data points are worth considering.

Tell us your bet in the comments!

Explanatory journalism is a public good

At Vox, we believe that everyone deserves access to information that helps them understand and shape the world they live in. That's why we keep our work free. Support our mission and help keep Vox free for all by making a financial contribution to Vox today.

One-Time Monthly Annual

$95/year

$120/year

$250/year

$350/year

Other

$

Yes, I'll give $120/year

Yes, I'll give $120/year

We accept credit card, Apple Pay, and Google Pay. You can also contribute via