Jack Valenti, longtime president and CEO of the Motion Picture Association of America, once said “No one, absolutely no one, can tell you what a movie is going to do in the marketplace… Not until that film opens in a darkened theater, and sparks fly up between the screen and the audience can you say this film is right.”

So how can a Hollywood outsider who admittedly sees very few movies each year predict whether or not a movie will be successful before it hits theaters? Using analytics, of course!

In a breakout session at the Analytics 2011 conference, Dursun Delen, Professor in the Spears School of Business at Oklahoma State University, shared a system that he and colleagues developed that predicts if a film will be a blockbuster or a flop.

Using seven different parameters of data, such as genre, the actors’ star power, special effects and the film’s rating, Delen determines a revenue range for a film before its release. After  that range is determined, Delen classifies the movie into one of nine categories ranging from flop (a film projected to make less than $1 million) to blockbusters (films that gross more than $200 million).

From 1998 to 2006, Delen’s model accurately predicted the amount of money a movie would make 75 percent of the time, and projected the exact range of revenue for films 37 percent of the time. For 2006 alone, the system accurately predicted movie revenue more than 90 percent of the time and the exact amount 56 percent of the time.

Despite Delen’s predictions, Hollywood executives haven’t yet tapped into his analytic approach to predicting the financial success of movies. Delen said he and his colleagues are having discussions with studio execs about using his model.

“More movies lose money than make money in Hollywood,” Delen said. “They need all the help they can get.”