This Jewish Data Whiz Uses Math to Predict the Oscar Winners

Ben Zauzmer says that the Jewish love of calculating odds is no coincidence: 'that same spirit with which we approach religion is also I think a very useful one to approach statistics with'

South Korean screenwriter Han Jin-won, South Korean film director Bong Joon Ho and producer Kwak Sin-ae pose with their awards at the 92nd Oscars Governors Ball in Hollywood, California, February 9, 2020.
AFP

In 2012, the first year Ben Zauzmer made Oscar predictions based on mathematical modeling, he received an email from a member of the Academy of Motion Picture Arts and Sciences.

Zauzmer was wrong to predict that the silent French film “The Artist” would win best picture, the member said: He and some of his friends in the academy had heard that “Hugo,” Martin Scorsese’s steampunk adventure, had a better chance.

Zauzmer “politely thanked him” and was proved right — “The Artist” took five Oscars, including best picture.

Since 2011, as a freshman at Harvard, Zauzmer has put his data skills to work predicting which films have the best chances of winning in most categories.

Zauzmer, 26, is a baseball analyst for the Los Angeles Dodgers, using math to identify candidates for the team.

Ben Zauzmer works during the day as a baseball analyst for the Los Angeles Dodgers.
Courtesy of Zauzmer

He continues a long trend of Jews who have been successful at forecasting — see Nate Silver, Nate Cohn and Harry Enten in the political realm. He’s also far from the first Jewish baseball analyst — the Dodgers’ president of baseball operations, Andrew Friedman, and the new “chief baseball officer” for the Boston Red Sox, Chaim Bloom, are two other prominent examples known for heavily using analytics to make baseball decisions.

Zauzmer, who hails from a suburb of Philadelphia and says his Reform synagogue was his family’s “home away from home,” doesn’t see Jews’ love for statistics as too coincidental.

“One thing I love about the Reform Judaism that I grew up with, and of Judaism in general, is that we’re always taught to ask why. … And that same spirit with which we approach religion is also I think a very useful one to approach statistics with: to not just accept anything, but to actually dive into the data and prove it or disprove it for yourself” he said.

Zauzmer says his dataset includes anything about movies he can put a concrete number on. The best predictors usually turn out to be previous award shows, but it also includes critic ratings or “scores” on aggregation sites like Rotten Tomatoes.

His success — in 2018, he correctly predicted 20 of the 21 categories he put into his models — has earned him an annual column in The Hollywood Reporter, as well as freelance pieces for The New York Times, among others. He documented his approach in a 2019 book, “Oscarmetics: The Math Behind the Biggest Night in Hollywood.”

Of course, each year brings surprises. Zauzmer’s predictions are never perfect — they just represent the nominees with the best statistical chances of winning.

Even he was surprised when “Moonlight” won best picture over “La La Land” in 2017, and not because of the infamous envelope mix-up. “La La Land” had dominated awards season and was therefore a heavy favorite in Zauzmer’s model.

“It’s why my math doesn’t present 100% or 0% for any nominee,” he said. “Probability is only probability, and not a guarantee.”

Here are Zauzmer’s main predictions for this year. Was he right?

Best Picture: “1917”; Best Director: Sam Mendes, “1917”; Best Actor: Joaquin Phoenix, “Joker”; Best Actress: Renee Zellweger, “Judy”; Best Supporting Actor: Brad Pitt, “Once Upon a Time in Hollywood”; Best Supporting Actress: Laura Dern, “Marriage Story”; Best Original Screenplay: Bong Joon Ho and Han Jin-won, “Parasite”; Best Adapted Screenplay: Taika Waititi, “Jojo Rabbit”; Best International Feature: “Parasite”; Best Documentary Feature: “American Factory.”