MPI system predicts pic's chance of success

Better box office through science?

That’s the contention of the Motion Picture Intelligencer’s Christopher Lanier, who claims to have developed an objective system to accurately predict a film’s chances of domestic theatrical success.

What’s more, the system makes its assessment while ignoring such factors as star value, release date and marketing campaign.

To back up his audacious assertion, in 1994 Lanier conducted a six-month, 81-film test of the system’s predictions. Each week, he faxed his projections about upcoming releases to studio execs, lawyers, agents and the press.

Remarkably, an independent UCLA Statistical Consulting Center study found MPI’s predictions, on the whole, to be extremely accurate.

“Actual theatrical rentals were not statistically different from what MPI predicted,” said UCLA professor Jan de Leeuw in summarizing the study. “MPI category predictions are a substantial improvement over the notion that box office success cannot be predicted rationally.”

But so far, Hollywood isn’t buying: Lanier has yet to convince a single Hollywood studio to sign up for his service.

One reason MGM/UA decided not to use MPI, according to Larry Gleason, MGM/UA president of worldwide distribution, was that by the time the system makes its predictions, it’s too late to be of value. “By then, experts within our own company can make a good guesstimate as to how a film will do,” he said.

Distribution execs at other studios expressed skepticism about the mysterious nature of MPI’s proprietary methodology and Lanier’s qualifications.

But Lanier, who holds a degree in economics and worked on MPI for nine years before going public, isn’t giving up. For one thing, he claims the system also can be applied to a rough cut, or even a shooting script, when creative changes still could be made.

MPI analyzes a film or script to determine its “entertainment quotient,” based exclusively on elements within the material itself, such as story, characters, acting and music.

Lanier doesn’t claim MPI can predict a film’s exact rentals. Rather, the system predicts a film’s probability of reaching certain rental milestones.

Lanier likens the information MPI provides to a baseball player’s batting average: While an average can’t tell you whether a batter is going to get a hit in any given at-bat, it does tell you that a .250 hitter will get a hit, on average, every four times at bat, whereas a .333 hitter will get one every three times at bit.

As in baseball, said Lanier, the information can be critical for making strategic decisions.

Lanier, who looks and speaks more like a college professor than a Hollywood exec, has analyzed every major film released since 1984 – more than 1,600 titles. While waiting for MPI to catch on, he runs Viagraph Prods., a high-end corporate video house.

Over the last few years, MPI analysis has shown a continued decline in entertainment value per production dollar spent, which Lanier said corresponds to the studio’s shrinking profit margins.

“The problem isn’t too many movies, it’s not that stars are being paid too much, it’s that the cost of the way we’re delivering entertainment value today is too high across the board. We have to get control of content,” Lanier said.

Lanier – who is big on sports analogies – compared a film’s theatrical release to a car race. While other factors such as driver, track, competition and weather may play a part in a race’s outcome, the car remains the biggest single determining factor, he said. “Even a great driver in a Volkswagen Beetle will lose to a mediocre driver in a Ferrari.”

Follow @Variety on Twitter for breaking news, reviews and more