Posted by Every year, the Academy Awards captivate audiences worldwide, celebrating the best achievements in filmmaking. However, predicting the winners remains a difficult task—even for industry insiders. Awards season involves complex dynamics: critical acclaim, festival momentum, guild awards, box office performance, and sometimes pure narrative momentum.
At BigML, we like to approach the problem from a different perspective: using machine learning.
Using historical data from previous Academy Awards along with key indicators from the current awards season, we trained models to estimate the probability of each nominee winning their category. By learning patterns from past Oscar events, the ML models can detect patterns that often precede a victory. In this post, we present our machine learning predictions for the 2026 Oscars (the 98th Academy Awards).
Data and Methodology
To build our predictive models, we collected structured data for nominees across the major Oscar categories. The latest version of our dataset covers 1,427 movies nominated for various awards from 2001 to 2026 and 299 features for each one, such as:
- Nominations and wins from precursor awards (Golden Globes, BAFTA, SAG, Critics Choice, etc.)
- Total nominations received by the film
- Historical performance of the director or actors
- Film release timing and festival reception
- Genre and production characteristics
These features were used to train classification models capable of predicting the likelihood that a nominee wins the Oscar. Our models were trained using data from previous Academy Awards ceremonies, and once trained, we applied them to the 2026 Oscar nominees to estimate their probabilities of winning.
The 2026 Predictions
The Best Picture is often the most complex category to predict. Many factors influence the final outcome, including industry momentum, ensemble recognition, and the overall narrative surrounding the film.
Among this year’s contenders, One Battle After Another, directed by Paul Thomas Anderson, has emerged as a strong frontrunner. The film has already accumulated several major precursor awards and nominations, giving it a strong statistical profile.
Another major contender is Sinners, directed by Ryan Coogler, which leads the nominations tally with 16 nominations and strong support across multiple categories.
Other notable nominees include Hamnet, Frankenstein, and Avatar: Fire and Ash, each with varying levels of awards-season momentum.
This year, One Battle After Another is the clear favorite to win the biggest prize of the night!
As for the more technical categories, we would also like to share our predictions:
- Best Cinematography: Sinners
- Best Costume Design: Frankenstein
- Best Film Editing: One Battle After Another
- Best Sound: F1: The Movie
- Best Visual Effects: Avatar: Fire and Ash
- Best Makeup and Hairstyling: Frankenstein (Although Sinners is a pretty close second position)
- Best Music, Original Song: Sinners (Although both KPop Demon Hunters and Train Dreams got practically the same score as Sinners)
- Best Music, Original Score: Sinners
- Best Production Design: One Battle After Another (Sinners is a close second)
- Best International Feature Film: The Secret Agent
- Best Animated Feature Film: KPop Demon Hunters
Final Thoughts
Predicting the Oscars is always a challenge. Even the most sophisticated models cannot fully capture the human factors that influence Academy voters: personal taste, industry narratives, and cultural context. Nevertheless, machine learning provides a powerful way to analyze historical patterns and quantify the signals that often precede Oscar victories.
Will the Academy follow the data this year? Or will there be surprises? We’ll find out when the winners are announced at the 98th Academy Awards. Good luck to all the nominees!
The Official Blog of BigML.com 