When does a movie have box office ‘legs’?

Perhaps the most famous dictum in Hollywood production comes from the dean of American screenwriters, two time Oscar winner William Goldman:

No one knows anything

Not one person in the entire motion picture field knows for a certainty what’s going to work. Every time out it’s a guess—and, if you’re lucky, an educated one…. The initial preview of Star! was such a success that Richard Zanuck cancelled any further previews and sent a wire to his father, Darryl, that said, “We’re home. Better than Sound of Music.”

They don’t know when the movie is starting to shoot either. David Brown, Zanuck’s partner, has said, “We didn’t know whether Jaws would work, but we didn’t have any doubts about The Island. It had to be a smash. Everything worked. The screenplay worked. Every actor we sent it to said yes. I didn’t know until a few days after we opened and I was in a bookstore and I ran into Lew Wasserman and said ‘How’re we doing?’ and he said, ‘David, they don’t want to see the picture.’”

They don’t want to see the picture—maybe the most chilling phrase in the industry. Now, if the best people around don’t know at sneaks, and they don’t know during shooting, you better believe that executives don’t know when they’re trying to give a thumbs-up or-down; they’re trying to predict public taste three years ahead and it’s just not possible.

–William Goldman. Adventures in The Screen Trade (1982).

It’s pretty fascinating that a billion dollar enterprise like a new Disney movie will be known to succeed, or fail, in a period of 24 hours within the film’s opening. Big movies have large opening weekends, but most importantly–they have legs–ie they sustain their audiences over a longer period, with a more modest decline over subsequent weeks.

Consider an illustrative example, comparing SHANG CHI AND THE TEN RING’S fine legs versus BLACK WIDOWS paucity.

Black Window had a $10m larger opening weekend. At the end of their respective domestic runs, however, Shang Chi had grossed an extra $41m! People who study this area have a simple metric to measure this property–a multiple

\[ \text{Multiple} = \frac{\text{Domestic Total Box Office}}{\text{Opening Weekend Box Office}} \]

So the multiples for Black Widow and Shang Chi might be

Film Opening Weekend Domestic Total Multiple
Black Widow $80.4m $183.7m 2.28
Shang Chi $75.4m $224.5m 2.98

Predicting little differences in movie multiples is a rich opportunity for abitrage (Disney’s stock opened 2% points higher following Inside Out 2’s massive success last weekend, for example). How should we predict if a film will open well and sustain?

Here’s a data set for every movie released between 1986 and 2022 for which a Cinema Score was made publicly available.

library(tidyverse)
library(magrittr)

t1 <- "https://github.com/thomasjwood/code_lab/raw/main/data/boxoffice_multiples_86_22.rds" %>% 
  url %>%
  readRDS

What’s a CinemaScore? It’s a company that’s been doing exit polling of movie attendees on opening weekends, with the theory that a film which had systematically better audience evaluation will have better legs as a result of interpersonal recommendations.

Are these scores predictive of performance? Or are they merely tapping differences in film marketing budgets, or critical consensus?

To work!

  1. We currently have a concatenated string vector t1$genre which is compact but not fit for modelling. For each of the commonly used genres, generate a separate dummy variable. Please use a purrr::map() to this end

  2. Generate a categorical movie indicating if the movie was released on the big 8 holiday weekends–MLK, Easter, Memorial Day, Labor Day, President’s Day, Thanksgiving or Christmas

  3. Plot the mean multiple by each letter grade, and then plot the residual multiple by grade after separately adjusting for critical consensus (t1$metascore ) and production budget (t1$budget)

  4. Estimate a list of regression models, measuring robustness of Cinemascore grades as predictors of box office multiples.