Warner Bros.’ controversial Joker amassed nearly 75 million views on YouTube for both its official teaser trailer and its final trailer pages in the months ahead of the film’s Oct. 4 release and after. And since its launch, Joker has gone on to dominate the October box office and break records globally, becoming the top-grossing R-rated film of all time and set to pass the $1 billion mark any day now.
Are these viewership stats related? While studios regularly tout when their film trailers break records — earlier this year Marvel Studios said Avengers: Endgame became “the most viewed trailer in history with 289M views in 24 hours!” — using YouTube views as a proxy for box office interest has taken a back seat to studios’ tracking surveys.
Related Stories
But YouTube — part of a Google suite that reached 258 million unique visitors in the U.S. in September, per Comscore — may be as close to a standard for evaluating how well trailers perform as any site in a fractured media environment. (And major Hollywood studio films often have their trailers re-uploaded by numerous third parties on the platform.)
While historical YouTube data isn’t available, independent tracking website Box Office Report has been collecting data on official YouTube trailer views, counting “views from official trailer pages and Fandango Movieclips” until the Saturday before release. Using this data set — an inexact measurement — on 421 major releases since the start of 2017, there can be, at least, a rough determination as to how strong of a relationship there is between interest in a trailer and in the film itself.
To begin, here is a chart showing the relationship between trailer views and opening weekend box office revenue. The further to the right a point is, the more views it received online. The higher up a point is, the more money it made during its initial weekend run.
Two points stand way out from the crowd. Avengers: Endgame (which grossed $2.8 billion globally in total) and Avengers: Infinity War ($2.04 billion) are a good deal ahead of all other 2017-19 movies in terms of both opening weekend revenue and YouTube trailer views. They flip-flop their rankings in the two measures, as Endgame wins on dollars but Infinity War wins on the trailer.
Both Avengers films stand out in both of these metrics, while other films are more impressive in one than the other. Venom ($856 million total global gross) and Spider-Man: Far From Home ($1.1 billion) are the two points far to the right but pretty low down, meaning they did even better in trailer views than opening weekend results. Star Wars: The Last Jedi ($1.3 billion) — the point sitting atop the cluster of dots on the left — had the opposite outcome.
Across all points, there does appear to be an upward trend, meaning a positive relationship between YouTube hits and opening weekend ticket sales. Rather than just eyeballing it, though, we can test the correlation between these two metrics. Correlation is a statistical metric which tells us how related two things are: A correlation of –1 means they go in the opposite direction as each other, 0 means no relationship, and +1 means the two are perfectly aligned. For the points on the above chart, the correlation is +0.86, a very high mark indicating a strong relationship between the two.
This is hardly surprising. For one thing, films with a bigger marketing budget and studio backing tend to drive up both trailer views (through social media and online advertising) and box office numbers. For another, the trailer itself serves its main purpose as a form of advertising to go see the movie, and like any other type of advertisement, the more people who see it, the more people who will buy the product.
To state this relationship another way, for every additional 1 million views a trailer on YouTube receives, the corresponding movie earns an addition $1.14 million dollars in its first weekend, on average.
One caveat is that many films these days have multiple trailers, often starting with a teaser before building up to one or two full trailers, and perhaps even a final trailer on top of that. As an example, take Incredibles 2 ($1.2 billion total gross) — the most viewed non-Avengers trailer of the past three years, at least per Box Office Report’s tally. Pixar released a teaser trailer (currently at 88 million views) in November 2017, followed by an “Olympics Sneak Peek” in February 2018 (55 million) and then an official trailer in April of that year (128 million).
Using the data included by Box Office Report, they also calculate the ratios of opening weekend revenue to trailer views, and there is a bit of a spectrum:
On the left side, we have movies with much less box office success than we would have expected based on their trailers’ online performances. Examples from this year include the smaller-grossing The Art of Self-Defense ($2.4 million total gross), Brian Banks ($4.3 million) and The Prodigy ($21.1 million). On the right side, we have movies that sold more tickets than their slighter YouTube numbers would have predicted. A few 2019 examples are Penguins ($7.6 million), Run the Race ($6.4 million), and Unplanned ($19.3 million). Most movies fall somewhere in the middle, with 59 percent of this 421-movie dataset represented by those two middle bars, right around the average relationship.
That still leaves plenty of room for outliers. Assassination Nation’s bloody trailer soared to 19.3 million views, yet only managed $1.05 million on its opening weekend for the worst ratio of the last three years. In the other extreme, Boo 2! A Madea Halloween only drew 1.1 million fans to its trailer yet scared up $21.23 million at the box office in its first few days, perhaps thanks to facing weak competition that October weekend.
So, that +0.86 correlation still leaves room for some movies that significantly over- and underperform what we would expect based on their trailers, since a myriad of other factors go into predicting box office success. Much like the trailer gives us a good but not perfect sense of what the movie will be about, it seems that the trailer’s success provides a good but not perfect predictor for the popularity of the film.
Ben Zauzmer (@BensOscarMath) writes about the intersection of entertainment and data for The Hollywood Reporter. He works as a baseball analyst for the Los Angeles Dodgers.
