The 93rd Academy Awards ceremony is taking place this weekend against the background of the ongoing global pandemic, which has caused the event to be postponed for several months. For the most part, movie and festival-goers around the world had to stay put in their homes and rely on streaming services for their viewing this past year. Depending on whether the nominees were available on your favorite streaming service, many of you may not have had a chance to get familiar with the nominees. Nevertheless, we have to trust that the distinguished members of “The Academy” have done their due diligence before the envelopes will be opened and the winners announced by a string of celebrities on Sunday. We hope that you use this post as your guide as you get in front of your TV among friends and family munching on your favorite appetizers. As they say, come rain or shine, the show must go on!
As usual, in case you’re up for some DIY Machine Learning fun, you can find our updated Movies dataset on the BigML Gallery and build your own models to actively join in the fun after you create a free account.
Machine Learning models typically improve with more data instances so we have kept all the previous data and features we considered for previous years’ predictions, and we added data from 2020. The dataset covers 1,345 movies nominated for various awards from 2000 to 2020 with 100+ features including:
- Film characteristics such as synopsis, duration, budget, and genre.
- Film evaluation measures from IMDB such as viewer votes, ratings, and Metascore.
- This year’s nominations and winners for 20 key industry awards including Golden Globes, BAFTA, Screen Actors Guild, and Critics Choice.
We’ve tested multiple modeling approaches such as OptiML, the popular AutoML feature on BigML that automatically finds the best performing supervised models, individual Deepnets, and some Fusions, which combine multiple supervised models to potentially add to the robustness of predictions. For each award category, eight in total, we trained separate models to see how the predictions compared and which approach gave the best results.
Once our candidate models were created, we made Batch Predictions against the movies produced in 2020 that we had set aside in a separate dataset. As was the case last year, all approaches yielded more or less similar predictions supporting each other’s constructs. The individual Deepnets models configured with the Automatic Network Search option were chosen in the end.
For example, the field importance report below is that of the Best Picture Oscar and it shows that fields like user reviews, synopsis, votes, and wins in Critic’s Choice plus nominations for Online Film Critics Society, People’s Choice, Hollywood Film, and BAFTA awards all factored in strongly in the final scores for each Best Picture nominee.
Having briefly touched on what’s under the hood, let’s go ahead and predict the 2021 winners! For each category, we predict the most likely winner along with other nominees sorted by decreasing scores. Keep in mind that these scores aren’t supposed to add up to 100. Rather, they are “points” given to the nominee by the underlying model on a scale of 0 to 100. Another way to look at this is that the model is telling us how a movie/artist with a given set of characteristics will do in a given award based on 20 years of historical data on that award AND independent of the other nominees for the same award this year.
Kicking off with the most anticipated category, Best Picture, our models (Deepnets, OptiML) consistently gave the best scores to Nomadland, which many experts agree is a deserving frontrunner. There’s a big gap down to the second-placed Minari and the remaining productions. If Nomadland wins this will be one of those special years the same entry wins both the Golden Globe and the Academy Award.
This year’s Best Director award is by far the most lopsided one according to our models. Chloé Zhao of Nomadland seems to have a virtual lock on the Oscar with Minari‘s Lee Isaac Chung showing up as the next best choice representing South Korea.
The Best Actress award seems a two-way race between Carey Mulligan of Promising Young Woman and the timeless Frances McDormand of the multi-category threat, Nomadland. The other nominees need a strong spell to get to hold the statue, but we wouldn’t be surprised to see either of the two walk away with it.
One of the consensus winners throughout this unusual award season has been the late, talented Chadwick Boseman and his performance in Ma Rainey’s Black Bottom, which otherwise got overshadowed by the likes of Nomadland. Our model has made Mr. Black Panther the clear favorite with Riz Ahmed‘s strong performance in Sound of Metal behind with a lukewarm score of 51.
The Best Supporting Actress category seems competitive with Yuh-Jung Youn appearing as the top choice followed by an up-and-coming Maria Bakalova. Olivia Colman also sports a respectable score in the hunt for an upset in this category.
As was the case last year between Joe Pesci and Al Pacino both of whom were nominated with The Irishman, the Best Supporting Actor award presents us with the dilemma of Daniel Kaluuya and Lakeith Stanfield this time. However, that’s where the similarities end because unlike Pesci and Pacino getting a similar number of nominations, Daniel Kaluuya (our pick) has swept the majority of the nominations for Judas and The Black Messiah. Sasha Baron Cohen‘s performance in The Trial of the Chicago 7 is the clear second choice.
Expected to pick up multiple awards by our models like its South Korean predecessor, Parasite, Minari is the heavy favorite to receive this weekend’s Best Original Screenplay award.
And last but not least, our model once again predicts a banner night for Nomadland as the winner of Best Adapted Screenplay as well and it’s not even close.
This concludes our 2021 Oscars predictions. As you get ready for the ceremony on Sunday, April 25th, you can now make some educated predictions with Machine Learning and perhaps impress friends. Best of luck to all the nominees and a big thanks to the Academy of Motion Picture Arts and Sciences (AMPAS) for putting on a show despite challenging times!