Could Algorithms Be Better at Picking the Next Big Blockbuster Than Studio Execs? (wired.com) 74
In a world where artificial intelligence is no longer just a Spielberg-Kubrick collaboration, could algorithms be better at picking the next big blockbuster than studio execs? From a report: "Filmmakers are getting closer to understanding what moviegoers go to theaters to see thanks to neural networks fed off of data from previous box office hits," says Landon Starr, the head of data science at Clearlink, which uses machine learning to help companies understand consumer behavior. "Although this technology isn't spot-on quite yet, AI-powered predictions are likely stronger than the human calculations used in the past." And they're advancing quickly.
Vault, an Israeli startup founded in 2015, is developing a neural-network algorithm based on 30 years of box office data, nearly 400,000 story features found in scripts, and data like film budgets and audience demographics to estimate a movie's opening weekend. The company is only a couple years in, but founder David Stiff recently said that roughly 75 percent of Vault's predictions "come 'pretty close'" to films' actual opening grosses.
Scriptbook takes a similar approach, using its own AI platform to predict a movie's success based on the screenplay only. The Antwerp startup's AI analyzed 62 movies from 2015 and 2016, and claims it was able to successfully predict the box office failure or success of 52 of them, judging 30 movies correctly as profitable and 22 movies correctly as not profitable.
Vault, an Israeli startup founded in 2015, is developing a neural-network algorithm based on 30 years of box office data, nearly 400,000 story features found in scripts, and data like film budgets and audience demographics to estimate a movie's opening weekend. The company is only a couple years in, but founder David Stiff recently said that roughly 75 percent of Vault's predictions "come 'pretty close'" to films' actual opening grosses.
Scriptbook takes a similar approach, using its own AI platform to predict a movie's success based on the screenplay only. The Antwerp startup's AI analyzed 62 movies from 2015 and 2016, and claims it was able to successfully predict the box office failure or success of 52 of them, judging 30 movies correctly as profitable and 22 movies correctly as not profitable.
Not difficult at all (Score:2)
Pick some random comic book characters, re-hash an old story line with one new twist, profit.
Re: Not difficult at all (Score:1)
I assumed that all superhero films were already written by Chinese AI.
Yeah, but it's not enough profit (Score:3)
I"m inclined to say yes. These aren't high concept art movies, they're popcorn flix. Like a pop song they follow a formula. Eventually that formula can be understood. Kinda like "PsychoHistory" from the Foundation novels. Eventually the math will be understood.
Re: (Score:3)
The question is what made Avengers clear 1.5 billion+ and Batman v Superman only do half that?
People are pretty good at finding patterns. In this particular case the pattern is that most Marvel universe movies are pretty good and will be enjoyable and that if a DC universe film is good, it was probably an accident. There isn't really a formula here, just a case of a shitty reputation. You can probably look for the underlying causes of that if you want to avoid such things in the future, or try to correct them now, but on the surface it's not difficult to reason out why one film did vastly more busin
We're talking about different patterns (Score:2)
The music equivalent is a four chord song. [youtube.com] And if you look at the history of these sorts of things we've always figured out sound first and video later. But so far video's always been figured out.
Re: (Score:2)
You can probably look for the underlying causes of that
Like having "Zack the Hack" for a director?
Re: (Score:2)
There was no executive involvement in Batman v Superman? One set of studio execs got it right, and one set got it wrong. So 50%. Still, neither knows until after the fact.
If you asked a third set to pick between the two, they would have 50% chance of being right.
If AI has a 51% ratio, it is better. I look forward to headlines predicting the end of studio executive jobs.
Re: (Score:3, Informative)
I do research in machine learning, literally designing the sort of predictors you're talking about, and I'd say no. Even given a lot of data (and even data on every movie in history wouldn't get near to what you'd call "big" data) you're still dealing with noisy samples - whether that noise comes from hidden variables or just dumb luck - and the system is fundamentally time-varying in either hard or impossible to predict ways (last years blockbuster could be this years flop as tastes evolve. If you don't
Re: (Score:2)
I think extraneous factors play heavily into it as well, and that's just more crap for machine learning to have to deal with.
What's the weather like on the first few weekends? (Might significantly impact attendance, and early attendance can make/break blockbusters.) What other movies are competing with it? When did they come out? (And how did the weather the first few weekends play into the timing of competing movies?) How similar/dissimilar to the hits of the last few years is this movie? (And is that good
Re: (Score:2)
Avengers were a team with male and female character. Superman and Batman are lone heroes.
Traditionally many of those adventure movies had a checklist of 12 plot features. It was describedf on this site many postings ago. Things like:
The Hero, The sidekick, The mentor, The foe, The girlfriend, The departure, The secret weapon, The first failed attempt, The lost buddy, The final success, The journey home, The celebration
You can fit movies like Star Wars and The Hobbit into this checklist.
Re: (Score:2)
Avengers were a team with male and female character. Superman and Batman are lone heroes.
Traditionally many of those adventure movies had a checklist of 12 plot features. It was describedf on this site many postings ago. Things like: The Hero, The sidekick, The mentor, The foe, The girlfriend, The departure, The secret weapon, The first failed attempt, The lost buddy, The final success, The journey home, The celebration
You can fit movies like Star Wars and The Hobbit into this checklist.
That sort of reductionist structuralism tells you nothing useful at all though. It's like all those "there are only X basic storylines" gags. They are unable to explain why one film or book is great, while another with a similar structure is utter crap.
You could map the broad outline of a Jane Austen novel against a generic Mills & Boon romance and say they are basically the same, but it would be meaningless.
Re: (Score:1)
By the rule of headlines (Score:4, Interesting)
The answer is of course "No".
The reason being is that it's mostly random. Kinda like the stock market. You can't make predictions based on past performance. Something or someone may fall out of favor in the public's eye. Or something or someone may be suddenly popular.
Re: (Score:3)
I was going to highlight this as a headline example where the obvious answer is "yes". Studio executives seem like they can be replaced by the AI of a magic eight ball with no loss in efficiency. (At least on the script picking side. I'm sure there's a lot of work that goes into Hollywood accounting.)
Re: (Score:3)
I'm sure there's a lot of work that goes into Hollywood accounting.)
It IS really hard to make a movie like Infinity War LOSE money on paper...
Re: (Score:2, Insightful)
We would assume it would be hard to fuck up star wars. Yet here we are.
re-hash/remake/horrifying future (Score:2)
That brings up a related topic of remakes. You can make a successful film and it can be remade with largely the same script but for it to bomb. This could be due to fondness for the original (Total Recall) or crapiness of the remake (Point Break).
Star Wars is not quite the same as the story is good but the need to introduce a more child-centric element (Jar-Jar Binks) caused some grief. Obviously the Ewoks got a similar reception in Return of the Jedi but as Tim in Spaced said "Yeah but Jar Jar Binks m
Re: (Score:2)
The reason being is that it's mostly random.
But it isn't random. Some movies are more likely to be profitable than others. For instance, sequels gross higher than new original movies, and they have more predictable costs since the director, writers, actors, efx guys, etc. have all worked together before. I can say with 99% certainty that the next "Fast and Furious" film will be profitable, as will the next 007 flick, and the next in the Star Wars series.
Kinda like the stock market. You can't make predictions based on past performance.
Making movies is nothing like picking stocks. For movies, past performance is a very good way
Re: (Score:2)
that's not the same thing as predicting the result of any single sample (movie) from that distribution.
Nobody is making that claim. The predictions only need to be probabilistically correct.
If Studio A makes a dozen movies that are all innovative new content, and Studio B makes a dozen movies that are all formulaic sequels or remakes of past hits, it is very very likely that Studio B will be more profitable, even though you can't predict the success of each individual film. Of course a few original new films will be hits.
Re: (Score:2)
Are you sure your signal stands up after normalizing for "contains Harrison Ford" and/or "original score by John Williams".
Surely the history (and demonstrated talents) of the cast and production team matter. Surely it matters if the original film was an epic blockbuster.
I almost wonder, though, once you subtract these terms, whether sequel-hood doesn't demonstrate a negative correlation with critical reception, alongside a mi
Easy (Score:2)
They could hardly be any worse.
Re: (Score:2)
Bravo.. You beat me to it..
Hollyweird has a abysmal track record on picking movies.
But the reasons are generally more self inflicted wounds than incompetence. They drink their own self serving cool-aid, believe only their views matter, then end up offending half their prospective customers pushing their left coast morality and ethics in movie scripts that have nothing to do with them. I don't know how many times I've said to myself that had they cut 20- seconds out of the love making sequence (The part
Could Algorithms Be Better? (Score:2)
Based on what's been released over the last 2 decades a brain damaged goat who shits on the movie titles would do a better job.
No. (Score:3, Insightful)
The goal is not to create good movies.
The goal is to create movies that sell.
That means pandering to the lowest common denominator of the movie going public.
Movies are not made to entertain people of above-average intelligence.
Re: (Score:2)
Some of what sells is good, and some of what is good doesn't sell.
Some of what's called, "good" that doesn't sell, isn't really good, either, though.
I'd say that "selling" is at worst orthogonal to "being good" but I think there is actually a component of "selling" that's on the "good" axis, since the whole aim of literary arts is permanent and universal interest, and by definition, universal interest will generate a few sales.
A.W.E.S.O.M.-O 4000 (Score:1)
I see THREE SHELLS! (Score:3)
in TEH FUTAR, all movies will be Pixar Marvel Wars.
It is an inescapable conclusion.
We've already seen entertainment by algorithm (Score:1)
When people call something 'formulaic' it isn't a compliment.
Anyone could pick better movies.. (Score:2)
not hard at all (Score:1)
SPACEBALLS 2 (Score:2)
The Search For More Money!
Not like it's hard... (Score:2)
... you could get better blockbuster predictions from a trained monkey than from studio execs!
Studio Executives? (Score:2)
Keep in mind that the basis of comparison here is studio executives. It seems like a pretty terrible way of picking blockbusters, so we should not be surprised if an AI outperforms that. The alpha male gets to green-light movies because as the alpha that is his prerogative, not because he has the best ability at that task.
Executives usually achieve and maintain their status by being the most successful at insider politicking and corporate infighting. That is not the same thing as being good at predicting
Futurama (Score:3)
Futurama did it with the Execubots.
Executive Alpha, programmed to like things it has seen before.
Executive Beta, programmed to roll dice to determine the fall schedule.
Executive Gamma, programmed to underestimate middle America.
Personal Experience (Score:4, Interesting)
A number of years ago I was excited to try a movie recommendation website. The premise was that you would rate a bunch of movies, and their algorithm would learn which ones you liked and which ones you didn't. It would then compare your results with other users. Suppose someone else rated a bunch of movies the same way you rated them, but in addition rated some movie very highly that you hadn't seen yet. That movie should make a great recommendation for you!
I started by spending a fair chunk of time rating a whole bunch of movies. I figured that the more I trained the algorithm on my tastes, the better results I would get. Finally, I decided to try for a recommendation. Lo and behold - up came some movie I had never even heard of, that was rated very highly by people who shared my taste.
Well, not only did I not enjoy the movie, I absolutely HATED it. I have distrusted recommendation systems ever since.
Re: (Score:2)
Did you only use it once? That seems a little premature. I mean, if 4 out of 5 suggestions are good, that'd be a decent track record. Particularly the movies we really hate, sometimes it's just one or two little things, which might be an issue for one but not others, like the way some people taste a weird flavor in cilantro and others don't.
I can also think of a couple of movies that I liked on one viewing and hated on another, or vice versa. Mood and circumstance can always play a role.
That said, some reco
When Summaries Ask a Question (Score:2)
Maybe.
Random Number Generator Is More Accurate (Score:1)
Grace Randolph has explained really well in the past how studio execs are completely out of touch with the audiences. They live in their own little world. There are movies that most people figured would be bombs, but studio execs thought would be top blockbusters.
It isn't like they don't have the data telling them the fact either. Take Ghostbuster for example. The original was made for
They could hardly be any *worse* (Score:2)
Cough. Firefly. Cough, cough.
Don't doubt algorithms could improve the median. (Score:5, Interesting)
Every person in charge of acquiring new material for a big media company is always on the lookout for the same thing, only different. That's because the financial backers have two, largely mutually exclusive goals: guaranteed audiences and a runaway hit. In economic terms they're looking for an investment with higher than normal returns for its risk.
That's why pop culture is so clogged with retreads. It's only a matter of time before we see Star Trek: With Tits.
Now I happen to know more about publishing than movie making, so I'll focus on that for a moment. New authors submit their manuscripts on spec to agents and publisher acquisition editors. These agents and editors are usually pretty sharp, but that makes their time valuable. So someone like an intern has to wade through the "slush pile". It's a horrible job because 99.9% of the slushpile is pure rubbish.
What the slushpile reader does for hours on end is skim the first page, and toss, skim the first page, and toss. The first page is about ten lines of text in standard manuscript format. But if an algorithm could make the first cut, it would be able to examine entire manuscripts for the desired combination of (a) resemblance to past hits and (b) differences from recently published books, winnowing hundreds or thousands of manuscripts down to a couple dozen candidates fit for human eyes.
The exact same process could be used for movie or TV spec scripts. American broadcast TV shows often have a problem ginning up enough story ideas to fill an entire season, but accepting spec scripts means someone has to deal with the slushpile. So there's usually a couple of writers-don't-have-any-ideas episodes each season. If you could process a couple of thousand spec scripts and pick a dozen candidates that fit the show, you might find an idea you could use.
The real problem... (Score:2)
To use this profitably, you need to toss in tons of dirt to get the few nuggets buried in the mountain of crap. The problem is that the algorithm can't tell you why a movie will be successful. If it could do that, you could skip all the crap generation process and just write a hit. Who is the genius out there who is going to figure out how to interpret neural network weights and features to build human understandable models?
SJW settings? (Score:2)
Pass the not stale test and the SJW AI approves the project.
The death of innovation ... (Score:2)
would be the result of AI choosing what the film studios make next. AI would work by rating what has been produced in the past. Someone with an idea for something completely new would not be highly rated by AI ... the risk is that they would not get the support to do it, so we all loose.
It would take true AI to do a good job (Score:2)
It's hard for me to imagine how an algorithm could really do a good job of "picking the next big blockbuster". On paper, the Justice League movie looks like it ought to be about as good as the latest Avengers movie, yet the Avengers movie is way better (compare IMDB ratings histograms: Justice League [imdb.com] Infinity War [imdb.com]). Zack Snyder's movies have made a lot of money. Joss Whedon has been associated with very successful movies like the first two Avengers movies. The actors have made very successful movies. Th
AI block (Score:1)
I have an idea (Score:1)
How about AI to replace humans to give viewers of movies original concepts that are not rehashed crap?
Make a movie about talking monkeys on another planet to symbolize the existentialism of ..... oh screw it. Give me explosions.
Well, the algorithms say so. (Score:2)
Could Algorithms Be Better at Picking the Next Big Blockbuster Than Studio Execs?
I wasn't aware they were even working on algorithms to pick studio executives.
Films already created by formula (Score:1)
Same old, same old (Score:2)
founder David Stiff recently said that roughly 75 percent of Vault's
predictions "come 'pretty close'" to films' actual opening grosses.
In other words, it doesn't fucking work.