Thursday, July 22, 2010

Movie Finance's Holy Grail?


When deciding what movies to finance and produce, you’d think it would be useful for studios and producers to have a clearly structured way of assessing how they might perform at the box office. On the other hand you might think it’s a complete lottery and that movie-goers are just a fickle lot. However it would be the holy grail if a film-maker had a real formula to determine accurately the true revenue potential of his movie, and more importantly how best to tailor his work towards success during production.

One aspiring guru has come up with just such a formula and is hell-bent on applying his ground-up approach to identify success factors and weed out known mistakes in movie production. His revolutionary approach factors in certain features that he believes lead to a movie’s success such as character types and development, genres and storylines that resonate with audiences, it determines specific camera shots and angles and a myriad of other factors which together, he claims, can reliably generate profitable movies.

Would Hollywood listen to such a revolutionary concept that could potentially force it to change how it does business? Our friend Steve arrived in Hollywood last year to introduce his concepts but he found that few doors opened to allow him to get his message across. Many of you reading this won’t be surprised at the reluctance of insiders in any industry to be told what to do by a relative newcomer, and the thought of questioning the established beliefs in the movie business is certainly too much for many of its fragile egos. No producer wants to be told “you’ve got a pretty good concept there but step aside and let me rewrite your script and change a bunch of other stuff”. And, the thought of applying a scientific process with a checklist of criteria that need to be fulfilled probably sounds too much like regular corporate business practice, not something most Hollywood creative types are best at.

However, it’s a topical issue every weekend as the box-office numbers are pored over. This week, for example, the trades are focusing on the differing performances of Inception and The Sorcerer’s Apprentice. Inception, a thoughtful, high concept “thinking adult” kind of film they don’t really make any more, could have stumbled amid the summer popcorn audiences, but took in almost $63 million. And although The Sorcerer’s Apprentice looked like a slam-dunk for the summer crowd it stumbled at only $17 million. From a review of movies currently on release our friend Steve feels there are several obvious mistakes in some of the big movies out there which, by applying techniques determined by his model, would have been corrected as those movies were being made. And, although the studios’ current favourite survey partner CinemaScore does a remarkable job of predicting the performance and longevity of movies at the box office, their results are based purely on exit polls taken during the movies’ first-run weekend. Those movies aren’t going to get any better once they’ve been released, but Steve believes the studios ought to listen to him because his model will help them to make better movies and get them right before they’re released.

Many companies and consultants have tried to solve the problem of “how much will my movie make” with decidedly mixed results. Others have looked at the more oblique issue of “how can we present the best movie, with the right mix of talent, genre and time of release in the calendar, to ensure our best shot at a good result”? That’s daily business for the studios and some elevate it to a science. But they’re generally dealing with known elements: scripts rewritten to an existing formula and tailored to the whims of directors and stars, marketing departments doing what they know has worked before, and churning out genre fare to suit the perceived demands of the public. Maybe best known of the current number crunchers is Ryan Kavanaugh, whose Relativity group runs mind-numbing models to determine where to put its money. But is he beating the average, as you might expect with such an appliance of science? Quite possibly he is. But it’s going to take a strong will or a spectacular result to get a major studio to fundamentally change how they do things and apply Steve’s new techniques to their movie slate.

Steve believes that by applying real analytical techniques to movies at every stage of their creation he can ensure an absolute success rate. Of course this sounds fantastic and if he can prove his model’s capability he’ll see a rowdy line of Hollywood producers and studio exec’s develop outside his office. He has discussed his model with the few producers and studio exec’s who would agree to meet him, with similar results: they like his concept but aren't sufficiently convinced to put their money behind it, at least until it’s proven. He’s in a Catch-22 situation where he needs to prove his model but Hollywood isn’t listening to him or lining up to fund his efforts. He comments about a current studio blockbuster:
“It is clear they have no idea why "The Sorcerer's Apprentice" lost money. It is crystal clear to me. That means on the face of it I am better placed then Jerry Bruckheimer and Disney to understand what the movie-going audience is looking for. That is a sobering thought. What do I do with that?”
What he’s looking for is someone who can understand his modeling approach and the idea of a breakthrough concept, who’s prepared to back him to go make a movie that, in Steve's words, will "break the gut-feel mould of Hollywood movie selection and add some science to the process". We're watching eagerly.

The Out Of Obscurity team.

11 comments:

Tini said...

I am, as any scientist would be, willing to be proven wrong by hard evidence. However I do NOT believe it to be possible to predict which movies will succeed and not by applying hard analytical techniques.

For two connected reasons. Firstly it is simply that people are fickle. What's 'good' today is not necessarily so tomorrow and for NO DEFINE-able reason.

Second the number of variables is too huge and the process is subject to Chaos theory. Most particularly the concept of 'Sensitive Dependence on Initial Conditions" or the butterfly effect as it is better known.

I would bet actual "Martin" money that the models being discussed will always be able to explain after the event why a movie flops or flies but never successfully predict it reliably.

Steve said...

My model predicted Avatar to the 98% accuracy based on factors developed before that movie came out. Hence a foresight prediction. Audiences are not fickle. They are predictable on mass. I have looked at movies that span 70 years and found the common factors in ALL of them. These are therefore the non variable factors. As far as how many? You suggest too many? I have 800 factors and more are being discovered. Your lack of a true understanding of the process is what drives these comments. I have4 years invested in this. It works.

Steve said...
This comment has been removed by the author.
Steve said...

My foster brother worked for 30 years in aerodynamics for formula one teams. For years they thought certain elements of airflow were confined to the systems of chaos theory and would be never "understandable". Now they use some of the most powerful super computers in the world and predict these behaviors with perfect accuracy. There is a man who predicts wave swells for the top surfing companies who schedule big surfing competitions around his predictions. The waves are supposed to be a chaotic unpredictable system like the weather. He is 95% accurate and has been for years. As an investment banker I learned that the markets, like people, are unpredictable. I then learned that people like George Soros and Paul Tudor Jones and the mathematician James Harris Simons have become billionaires doing just that. I know that many hundreds of years ago our planets were seen as random dots of light who's movements we would never predict. Now we can predict their movements for millions of years into the future. Maybe mankind does not want to believe it is predictable. That does not change that fact that it is. I hope almost everyone else in the movie industry believes like you do, then, like the Billionaire fund managers I will remain a valuable unique asset. I am looking for just one person, with money to invest, that is willing to test my theory in a real movie.

Tini said...

As I said in my post I'm willing to be proven wrong...keen even, for I would love to see you become rich and famous.

You've waited a good long while to get this far. So what will it take to devise a test? One Avatar prediction won't do it. That's no improvement on anecdotal evidence.

A *set* of predictions would be a start. Can you not persuade twenty or so film makers to let you analyse what they are doing and make a prediction?

When you are proven right and I wrong I will happily buy the first crate of Champagne - though I'll let you buy the Scotch ;-)

Tini said...

You'll need to build in the appropriate safeguards so people can't accuse you of cheating. Maybe you can get James Randi to devise the protocols - and gain some publicity in the process.

Martin

Steve said...
This comment has been removed by the author.
Steve said...
This comment has been removed by the author.
Steve said...

I assess billion dollar grossing movies and movie franchises. I find common creative factors in all of them. I currently have 800 factors. I then compare movies to that list of factors. The movies that grossed over $2 billion score over 95%. A billion dollar grossing movie scores somewhere between 90% and 95%. Avatar scored 98% indicating over a $2 billion box office. People were not even sure if it would break even when it came out. There is some correlation between the score and the returns for lesser movies. I cannot tell you if a movie will make $200m or $250m. What I can do is judge in terms of bands like <$100m, $100m to $400m, $400m to $800m, $800m to $2B. That might not seem like a lot yet when the studios spend $150m on a movie + $50m in advertising it has to make $400m to break even. I can tell them if that will happen or not. My focus is billion dollar grossing movies.

The Sorcerer's Apprentice was supposed to be the next Pirates of the Caribbean. So was The Prince of Persia. They were expected to make close to $1 billion. I could have told the studio and Bruckheimer that those movies were going to be in the sub $400m category. It is worth a lot of money to studios to know that. What is most important is I could have told them why. I could have determined the factors that their movies lacked. I could have fixed the problems of those movies. I could have helped both of those movies into the $400m to $800m revenue band.

I have found clear patterns in my 800 creative factors that point to a few basic theories of what billion dollar grossing movies are telling us and why we go in large numbers to see them. These theories can then be applied to new movies. How much of the core theories does a movie concept express? I can look at a trailer and see that. I can estimate how much initial box office interest a movie will generate in its opening week without even seeing the movie. Or I can put a movie project through the entire 800 factor model. That takes up to a week to do that. I have put 19 billion dollar grossing movies, and 20 non billion dollar grossing movies through that process. That is 39 weeks of work. I have another 8 non billion dollar grossing movies that I will soon put through that process.

I have trailers for over 300 recent movies that I assess against the core theories of the model. I assess them on the Movie Concept, the Character Representation and the Visual Expressions. I can then predicted outcomes like the success of Inception at over $500m. I can predict the higher gross of The Other Guys versus the similar Cop Out. I can also predict that The Other Guys will struggle to make a profit on a $100m budget. I can explain the lower revenue of the just released Salt compared to the similar movie Wanted. I can explain the success of How to Train Your Dragon, a movie the media was predicting would fail. I can explain the failures of Prince of Persia and Sorcerer’s Apprentice and The Last Airbender. This is only from studying the trailers. The unexpected success of The Hangover is no surprise to me. That movie scored very high on my model. It fits closely the basic theories of billion dollar grossing movies.

Take the Land of the Lost. Budget $100m. World Wide Gross $69m. Compare that result to a similar type of movie called Journey to the Center of the Earth. Budget $60m. World Wide Gross $242m. The reason one movie failed and one movie succeeded is crystal clear in their trailers. I have not seen either movie, yet I know why this outcome occurred. I read a Hollywood journalist that said we all suddenly collectively got tired of Will Ferrell yelling at us. No such thing happened. It was a case of not understanding two key elements of a trailer: Audience Collective Comfort and Target Audience Identification. I don’t claim to be able to predict every single cent of revenue for every movie. What I can do is tell a studio how close their planned movie will score against the highest revenue producing movies of all time and then improve it to bring it as close to 100% as possible.

Steve.

Tini said...

Steve: Whatever my opinion about this it isn't people like me you need to convince. And I suspect that to convince the money men in the industry you will need to do something that is slow and even harder.

You need to analyse BEFORE the event a range of movies.

If I understood right - and I'll apologise if I haven't - your analyses that you describe are all so far post event. IE you have looked at movies in all their myriad glory and explained why they did what they did. That's interesting but it doesn't mean you can predict. It might mean that but it doesn't demonstrate it is what I'm getting at.

Did I misunderstand? Have you been able to demonstrate this before a movie has been made, reliably in the bands you describe?

Cheers//Tini

Steve said...

In a sense I have made pre movie release predictions, as described here, by my correct assessment of movie trailers. Trailers are all the information I have before a movie is released. For example James Cameron did not send me a copy of Avatar to assess before it was released. Never the less the 800 factors were determined before Avatar came out and when I put that movie through the model it scored 98% out of 800 factors. What you are looking for is a public forum where I state movie predictions prior to their release and then in time I am seen to be right. I don't have access to the movies prior to release to do that. I recently spent 9 months in Hollywood and I was not able to get a meeting with anyone at any studio. They were not interested in even talking to me. That is what the article here is about. It does not matter how accurate my work it. So far major people in the industry have shown no interest in even looking at it let alone testing it.