The Academy Awards: an awards show where every nominee wins. The publicity of a nomination in itself is desirable, regardless of the outcome. However, until the final announcements are made, revealing who will take the award home, fans and stars alike hold their breaths. To spare you the suspense, The Reckoner is using a combination of data analysis and reasoning to predict the Best Picture, Best Director, Best Actor, and Best Actress of the 89th Academy Awards.

Tl;dr: The winners will include La La Land, Damien Chazelle (La La Land), Casey Affleck (Manchester by the Sea), and Emma Stone (La La Land).

 

Best Picture: La La Land

Our selection for the winner of Best Picture is, perhaps unsurprisingly, La La Land. The musical romantic comedy has been incredibly successful since its release in December 2016, grossing nearly $300 million worldwide. The film has received a record of fourteen Oscar nominations, tying for the most nominations with All About Eve in 1950 and Titanic in 1997. In January, La La Land also won all seven Golden Globe awards it was nominated for, breaking the record for most Golden Globes won.

The winner of Best Picture can usually be predicted based on how many Golden Globe awards were won and how many Oscar nominations it received. Winners for Best Picture typically have more Golden Globe wins and Oscar nominations, indicative of the popularity and quality of the film. This corresponds with higher normal scores (explained in the methodology section below) allowing us to use the criteria of a positive normal score as a cutoff. We eliminated films with negative normal scores: Arrival, Hacksaw Ridge, Fences, Lion, Hell or High Water, and Moonlight. Of the remaining nominees, La La Land had the highest normal score of 2.06, making it the highest ranking candidate and our choice for Best Picture.

  

 

Best Director: Damien Chazelle

The Reckoner forecasts Damien Chazelle, the director of La La Land, as the most likely winner of Best Director. In the past seven of ten Oscars, the Best Director award has gone to the director of the winner of Best Picture. Good films need a good director. This has occurred frequently enough that Damien Chazelle would be a reasonable prediction for the award. He recently won two Golden Globe awards for Best Director and Best Screenplay for La La Land, beating Mel Gibson (Hacksaw Ridge), Barry Jenkins (Moonlight), Kenneth Lonergan (Manchester by the Sea), and Tom Ford (Nocturnal Animals). With those two awards under his belt, Chazelle’s normal score jumps above the rest. Although Mel Gibson is the only one of this year’s Best Director nominees to have previously won an Oscar, we decided that the La La Land hype outweighs Gibson’s past experience. Priority is given to recent data, and in the calculation of Golden Globe wins to nominations ratio, Damien Chazelle clearly comes out on top. Given the extent of La La Land’s success, Damien Chazelle is the best bet for Best Director.

 

Best Actor: Casey Affleck

Casey Affleck is ready to step out of his older brother’s shadow. As the leading actor in Manchester by the Sea, he won his first Golden Globe in January 2017. While the film was met with critical acclaim, the quality of the film itself has little to do with our prediction. Rarely is Best Actor awarded to the lead actor of the Best Picture. In fact, when analyzing the normal score of Golden Globes won by the actor as a function of critic rating, there is little correlation. The quality of a film, measured by Rotten Tomatoes, says little about the quality of an actor’s performance.

After last year’s #OscarsSoWhite fiasco, we hoped that Denzel Washington would win with his emotionally-charged performance in Fences. Unfortunately, the statistics say otherwise. Casey Affleck has a greater normal score for his Golden Globes wins to nominations ratio. Note that when tallying wins and nominations, we included the results of the 2017 Golden Globes ceremony. Thus, Casey Affleck, who won Best Actor of a Drama, has had his most recent win accounted for. While this bumps up his normal score, it is in itself a strong indication that Affleck will win again at the Academy. In the past ten years, only once has an Academy Best Actor not won a Golden Globe for the same film. As the nominee with the highest normal score, Casey Affleck will be the man of the hour at the 2017 Oscars ceremony.

 

Best Actress: Emma Stone

Previously, we foretold that La La Land would win Best Picture. If Emma Stone, the leading actress in La La Land, wins the Best Actress award, she will be the first in twelve years. The leading actress of the Best Picture has not won Best Actress since Hillary Swank in 2004. Emma Stone could be the actress to break the curse of this category.

Our prediction of Emma Stone being the best actress is founded on her normal score for her Golden Globe wins to nominations ratio. Using just the normal score for Golden Globe wins, actresses like Meryl Streep that have been active since the late 1960s should have swept the awards in the last decade. To give young actresses a statistical fighting chance, we based our predictions on their proven probability of winning. Just observing the normal scores of this year’s nominees, it seems that the winner should be Isabelle Huppert, not Stone. But the fact is, Huppert is a 63-year-old French actress who has never broken into the American big leagues until this year. Although a hotshot on the French circuit, she has never even been nominated for a Golden Globe before Elle. Yes, she may have won Best Actress of a Drama, but Stone won Best Actress of a Musical or Comedy. So we disregarded Huppert’s 100% Globes success rate as well as her normal score. And so, as the nominee with the next highest normal score, Emma Stone will be taking home Best Actress.

 

And there you have our predicted winners for the 2017 Oscars. Results will be revealed 26 February – may the best nominee win!

 

Methodology

We analyzed data from the 2007 – 2016 Academy Awards. In each category of Best Picture, Best Director, Best Actor, and Best Actress, we compiled data such as Golden Globe nominations, Golden Globe wins, Oscar nominations, and critic and viewer ratings. Award records were gathered from the Internet Movie Database (IMDb) [1]. Critic ratings and viewer ratings were gathered from Rotten Tomatoes [2].

In the data analysis of the nominees, one-variable statistics were used to compare candidates of the same year. We took the arithmetic mean and standard deviation of the categories’ number of Oscar nominations, number of Golden Globe nominations, and number of Golden Globe wins to compare the nominees. These categories were selected as they were indicative of a nominee’s success and would help predict the winner of this year’s Oscars. To calculate relative success, a normal score was calculated using the following formula:

z = (x – μ) / σ.

Where x represents the value of the data set, is the mean of the data set, and is the standard deviation. This results in a normalized score in which higher values indicate better standing among the nominees of the year.

To verify that it was valid for us to make predictions founded on normal scores, we compiled the normal scores of the last ten years of nominees. In general, the highest normal score guaranteed a win. All our compiled data is available for viewing online [3].

In all four categories that we analyzed, we based our predictions on another awards show: the Golden Globes. The Globes are typically held a month before the Academy Awards and thus are a strong indicator of who the outstanding Academy nominees are. For the categories of Best Director, Best Actor, and Best Actress, we also studied each nominee’s performance in previous Oscars. However, subsequent analysis proved that there was a weak link between having received previous awards and being awarded again. The Golden Globes being a better predictor of Academy winners than the actual Academy can be explained by the sheer number of Golden Globe categories. Instead of just a Best Actor category, the Globes offer both a Best Actor of a Musical or Comedy award and a Best Actor of a Drama award. It gives a more general idea of who the better actors are. Based on performance record, we can predict the best actor, the winner of the Academy.

It should also be noted by anyone who follows the Academy winners that the awards do not necessarily indicate timelessness. The voting system, instant runoff, favours nominees with broad support. While we at The Reckoner support ranked ballot voting for governmental elections, the system is not appropriate for all cases [4]. As Vox Media has discussed, this system has placed nominees with “broad support rather than passionate support” in first place [5]. This means polarizing films like Moonlight, which may be ahead of their time, are not recognized with the Best Picture award. So even if La La Land wins this year, it may not be the best movie of 2017 in 50 years.

Before finalizing our forecasts, we had considered many avenues through which we could determine the quality of a film. Intuitively, we first turned to film budgets as a possible indicator of quality. However, the problem did not come from the fact that budgets are seldom released. Movies, especially those that are produced by major studios, recalculate their budgets after the release. The problem lies in Hollywood accounting; this is possibly the shadiest practice of the movie industry. In order for unsuccessful releases to appear less prominent, studios siphon the loss into the budget of a successful film. Infamously, Darth Vader’s actor, James Earl Jones, has still not received the residuals from Return of the Jedi, despite the fact that it mysteriously ranks 16th in inflation-adjusted top box office earnings.

 

Sources

[1] http://www.imdb.com/

[2] https://www.rottentomatoes.com/

[3] https://docs.google.com/spreadsheets/d/1Hr6N62-W8Tlf780VN_68IK6PfP4ZTU4m1wrRVmSkcaE/edit?usp=sharing  

[4] http://thereckoner.ca/op-ed-winner-takes-all-means-canada-loses/

[5] http://www.vox.com/2016/1/14/10767586/oscars-nomination-process

[6] http://www.boxofficemojo.com/alltime/adjusted.htm