It’s that time of the year when the anticipation and speculation about the upcoming Academy Awards fills the airwaves and makes for juicy internet chatter. Unlike last year’s heavyweight nominees, Barbie and Oppenheimer (aka Barbenheimer), this year we have more of a level playing field potentially delivering surprise winners. This Sunday’s 97th Academy Awards will be hosted by Conan O’Brien from the Dolby Theatre at Ovation Hollywood. Once again, we are happy to do our part at BigML and publish our Machine Learning predictions as they have become an annual tradition.
Emilia Pérez, the polarizing French-directed, Spanish-language musical about a trans Mexican gangster, leads the 2025 Oscars with 13 nominations, which is not only a record for a non-English-language film but also is just one shy of the all-time record for any film! Previous non-English-language record holders were Crouching Tiger, Hidden Dragon and Roma with 10 nominations. The Brutalist and Wicked, tied with 10 noms each complete this year’s top three designated movies. It is also interesting that among the studios, Netflix leads with 16 nominations. How times change!
Other interesting tidbits from the nominee field include the Brazilian entry I’m Still Here‘s star Fernanda Torres, who repeated her mother Fernanda Montenegro’s Best Actress category nomination from 1999 for Central Station. Meanwhile, Timothée Chalamet, followed in the footsteps of James Dean as the youngest two-time Best Actor nominee. This year also marks the first time two musicals, in Wicked and Emilia Pérez, have been nominated for best picture since 1968.
The Data and Models
As we have done since 2018, our team has collected the relevant data points about this year’s nominees and generated a new set of Machine Learning models for each award category. This makes your job much easier should you decide to make your own ML-powered predictions. Simply clone the Movies 2001-2025 public dataset from the BigML gallery and use it in conjunction with the wide variety of modeling resources of the BigML platform. Don’t have a BigML account yet? You can easily create a free one here in less than a minute!
The latest version of our dataset covers 1,377 movies nominated for various awards from 2001 to 2025 and 299 features for each one including:
- Film characteristics such as synopsis, duration, budget, and genre.
- Film critic and audience reception measures like ratings and box office performance.
- Nominations and winners for key industry awards such as Golden Globes, BAFTA, Screen Actors Guild, and Critics Choice.
As in the last year, we updated the betting odds from various online sites for additional predictive power. This data point applies only for the major award categories: Best Picture, Best Director, Best Actor, Best Actress, Best Supporting Actor and Best Supporting Actress. Our OptiML models from last year showed that historical odds are pretty important in predicting Oscar winners. Of course, the data fields capturing the results of top movie industry awards for the year to be predicted continue to reign supreme in terms of signal to noise ratio.
As a reminder, OptiML (our popular AutoML feature) automatically finds the best-performing supervised models for you. When the individual OptiML model search executions finished for each award category, our engineering team built Fusions, by combining the top performing models from each OptiML search based on the phi-coefficient metric. This means different modeling techniques may partake in the final predictions, which in turn can make the resulting predictions more robust and arguably less noisy due to equal weight averaging. This year we chose to make use of the top 10 models (vs. last year’s top 5 or the prior year’s top 20) to find a happy medium between tightening the predictions and avoiding fewer potentially overfitted models.
Once our Fusion models were created, we made Batch Predictions against the movies produced in the year 2024 (slated for the 2025 Oscars), which we had split aside into a separate dataset from our main dataset mentioned above.
As usual, given BigML’s emphasis on white-box models with full explainability, we can dig deeper into models and predictions for added introspection as needed. For example, you can navigate to any Fusion model’s partial dependence plot to see how various data fields interplay in determining whether a given movie or individual nominee will win the Oscar.
The 2025 Predictions
Without further ado, we move on to this years predictions with the most likely winner along with other nominees in the same category sorted by decreasing scores. Keep in mind that these scores are not supposed to add up to 100. The models are telling us how a movie/artist with a given set of characteristics will likely do in a given award category based on 25 years of historical data on that particular award. That assessment is made independent of the other competing nominees for the same award this year. In other words, a high score can be interpreted as that nominee’s overall profile looking quite like the old winners of that category.
- Anora is the clear favorite to win the biggest prize of the night.
- This prediction usually goes hand-in-hand with the Best Picture and that points out to Sean Baker as the victor.
- Get ready to see real competition here. No one should be shocked if Mikey Madison pulls it off with the Anora momentum from other categories behind her, but our Fusion models put Demi Moore a step ahead for what could be a career moment for her.
- Adrian Brody is expected to walk away with the golden statue potentially marking the peak moment of the night for The Brutalist fans.
- Zoe Saldana is as solid as favorites come. Overall, our models predict Emilia Perez won’t have as much to show for its big tally of 13 nominations, but this category is the exception.
- We think Kieran Culkin will be crowned in this category with a comfortable margin and it won’t be anything near A Real Pain for him!
- Prior years have shown screenplay awards can be tricky to predict. Anora seems to be in the lead here but don’t be shocked if The Brutalist or The Substance end up edging out.
- Nickel Boys is our dark horse pick of the night and we’re sticking to it even tough Conclave and Emilia Perez make for serious competition.
In addition to the popular awards above, three years ago we added 11 more categories that are more technical in nature. So far we have a pretty good track record with those so we continue predicting them for you as follows.
- Best Cinematography: THE BRUTALIST (NOTE: DUNE: PART TWO is a close 2nd.)
- Best Costume Design: WICKED
- Best Film Editing: WICKED (NOTE: Although THE BRUTALIST and CONCLAVE are practically in a dead heat with WICKED.)
- Best Sound: DUNE: PART TWO
- Best Visual Effects: DUNE: PART TWO
- Best Makeup and Hairstyling: THE SUBSTANCE (NOTE: EMILIA PEREZ is a close 2nd here, and WICKED a close 3rd.)
- Best Music, Original Song: “EL MAL”, EMILIA PEREZ
- Best Music, Original Score: EMILIA PEREZ (NOTE: CONCLAVE is a close 2nd here, and WICKED a close 3rd.)
- Best Production Design: WICKED (NOTE: CONCLAVE is a somewhat competitive 2nd.)
- Best International Feature Film: EMILIA PEREZ (NOTE: I’M STILL HERE is a close 2nd here.)
- Best Animated Feature Film: THE WILD ROBOT
This concludes our 2025 Oscars predictions. Now it’s your turn to follow in our footsteps building your models as you get ready for Sunday’s ceremony, which has a promising lineup of prime time entertainment. As usual, we’ll follow up with a post early next week to grade our predictions. Until then, good luck to all the nominees!
