When Ben was studying applied mathematics at Harvard, he decided to use statistical methods and data to handicap the Oscars. Since around 2011, he has been using his models to make predictions with relatively good results.
It started back in 2011, when Ben watched a film called Moneyball. The movie was about how the Oakland A's baseball team used math to pick a winning team. He was really interested in the topic and it lead him to looking for the same type of prediction system for the Oscars but there wasn't anything.
He says he uses anything he can that can be quantified such as Rotten Tomatoes scores, previous award shows such as the Golden Globes and BAFTA's, and any other aggregate data bases such as Metacritic. In the end, he chooses the films that have the best statistical chance of winning. Since some of the voting is based on human nature, his predictions are never 100 percent but he has done reasonably well. Furthermore, he only makes predictions for 21 of the 24 categories and never makes any predictions for the three short film categories. In 2018, he predicted 20 of the 21 winners but he is not usually that accurate.
He also looks at the previous year's Oscars to see how well each of the different factors effected the accuracy of the results and uses this to adjust the weights of the predictions. Every year, he looks at this and readjusts the weights for his current calculations. Most people think that box office numbers would be a good indicator but it turns out there is really no correlation. Since he began doing this in 2011, he has had a 77 percent success rate.
He has also released a book explaining the whole process. His book, Oscarmetrics:The Math Behind The Biggest Night In Hollywood, talks about the process of how his models came about, the history of the Oscars, and he tried explaining everything so you don't need a degree in math to understand it. So check it out if you want. Let me know what you think, I'd love to hear. Have a great day.
No comments:
Post a Comment