Computer Program Can Predict a Hit Movie
#1
Computers can do anything.


LONDON (Reuters) - Hollywood producers fretting over this year's box office downturn should take heart.

A scientist in the United States says he has come up with a computer program that helps predict whether a film will be a hit or a miss at the box office long before it is even made.

"Our goal is to try to find oil in a way," said Professor Ramesh Sharda of the Oklahoma State University on Wednesday.


http://today.reuters.com/news/newsarticl...xml&rpc=22
Reply
#2
Absolutely ridiculous. I would think that in researching this individual would have found that there's no constant to any of these things. Half of the factors he's plugging in are completely unreliable. I think it would have a 50/50 chance at best, and the 50 percent that it did get right would more often than not be due to dumb luck.

I could go through the next three months releases and fire off figures, and I'd put good money down that his computer program would be no better than me.
Reply
#3
Well, he did use 800+ movies, and the things the variables do make sense. For instance, sequels have built in audiences, and the bigger the previous movie, the more likely the next will be as big if not bigger: Star Wars and Harry Potter certainly prove this rule. Star power also drives a movie for the same reason: built in audiences who follows stars. More and bigger stars, the more likely (though not guarenteed) your audience is. The article makes clear that his success is certainly far from perfect. But his rate is, what, nearly 40% dead on and 75% (operating from memory) within one of his five assigned categories? That's enough to make it more than reasonably worthwhile to further develop, and the more data you enter, the more accurate the program would become.

I mean overall, a "big flop" is almost as rare as a "Cinderella hit". My Big, Fat, Greek Wedding or The Blair Witch Project are exceptions that prove the rule. But did anyone truly think that Gigli was going to be a hit? Even a "surprise" hit? Recently, on the Star Trek forum, we've been talking about the last Trek movie Nemesis and how it was released right before Lord of the Rings. Is it any surprise, therefor, that Trek did poorly when Rings was released the next week.

I agree that the program certainly can't predict with certainty, but if a studio can predict 3 out of 4 times, that's not bad. It would certainly be worthwhile for that same studio to increase the accuracy further with more criteria and deeper data.
All your base are belong to us.

It could be that the purpose of my life is only to serve as a warning to others.
Reply
#4
What is scary about that is that studios will lean even more into the "formula" of sequels, remakes, and standard movies than they do now, if they can "run the numbers" and generate a percenatge chance of a hit or flop.

This makes me fear even more bland sameness from the big studios.
Wrestling Darwin on a daily basis.

"Question boldly even the existence of a God; because, if there be one, He must more approve of the homage of reason than that of a blindfolded fear." -Thomas Jefferson
Reply
#5
Mike of Quantum Muse Wrote:What is scary about that is that studios will lean even more into the "formula" of sequels, remakes, and standard movies than they do now, if they can "run the numbers" and generate a percenatge chance of a hit or flop.

This makes me fear even more bland sameness from the big studios.

But isn't this, in a more primitive fashion, what studios do right now? They get a decently large named actor, pair him with a decently young and known actress, slap the currently large action director and group of writers in there, and throw together a loose plot and a wrap it in a big check with a bow of marketing on top and hope it swims. If they're right 6 times out of ten, then they make money.

The worthwhile movies have generally come from smaller budgets, more serious, but less agressive actors, actresses and directors, and with one or perhaps two writers at the most.

A computer program will just do faster and more efficiently, what big studio execs have been doing for years.
All your base are belong to us.

It could be that the purpose of my life is only to serve as a warning to others.
Reply
#6
I still don't buy it. One could dig up as many examples of the factors failing the model as them supporting it.
Reply
#7
Boomstick Wrote:I still don't buy it. One could dig up as many examples of the factors failing the model as them supporting it.

To me it just makes sense. The data shows that he has about a 40% dead one rate, and 75% one up or one down. That's about as good as one could hope for in the industry. <shrug>
All your base are belong to us.

It could be that the purpose of my life is only to serve as a warning to others.
Reply
#8
RobRoy Wrote:To me it just makes sense. The data shows that he has about a 40% dead one rate, and 75% one up or one down. That's about as good as one could hope for in the industry. <shrug>

I guess. I still don't think a computer program was necessary, though. I doubt if studios are just going to take a print out and go with it, there's too much money involved. They're still going to pour over it themselves, so it's sort of an unnecessary redundancy, if you ask me.
Reply
#9
I think this is an example of what we've seen for 50 or more years - computers can do the same things that humans can do, if the humans program the info into them. And they can do it faster. But they don't come up with original ideas - they just process data.

So if a computer is able to analyze the fact that a sequel to a hit science fiction movie will probably make a lot of money... well, who don't know that? Or that a big blockbuster movie will do better if it's released around a holiday... and if it doesn't have any strong competition the same day.... well, a computer can analyze the box office data from the last decade in a fraction of a second, while it would take any of us hours if not days to come up with all the numbers. But we know it already, based on past observations.

So if you assign numerical values to things like a) popular series of children's fantasy books, b) stodgy old Oxbridge don as the author, c) success of the three LotR movies, d) success of the Harry Potter movies about British schoolchildren, e) drawing power of the name Disney, and f) previous box office success by the director of the Shrek movies.... then a computer can confidently predict that a big-budget adaptation of Narnia released close to Christmas is likely to make several hundred million dollars.

But we could have predicted that all along. Smile

I think accountants and industry forecasters have been doing this for quite some time now, just without one program. Like whoever realized that it was worth paying Jim Carrey $20 million dollars for a film, since his presence alone would boost revenue by more than $20 million, whether or not the film was any good.

What the program can't do is predict things that have never been done before. Like "Greek Wedding" for example - a family film derived from a stand-up comic's routine on her family, produced independently? No one connected with the film had any previous indication in their careers that they could make such a huge hit (producer Tom Hanks for example had only a modest success with the much-bigger-budget "That Thing You Do.") And hundreds of independent, low-budget horror films bomb - there was no way to predict "Blair Witch" would do so well, other than by the reaction of people when they first saw it.
August  - Jack's Pack Fan # 1, Keeper of the List, 3-Time Speaker of the JoAT Fan Quote of the Week, and the only person ever to have Back 2 Back Jack and Cleo fan quotes !
Reply
#10
I would have thought that Serenity would do better. I wonder why it didn't. When the DVD's of the series came out they sold well. The movie was pulled before I even got a chance to see it on the big screen.
Reply
#11
I don't really know, since I never saw it either.

But my guess is that it was really simple - the million or more hardcore fans who all went out and bought dvd sets all went to see it the opening weekend... and the movie made something like $9-10 million. It came in at #2 that weekend, but then took a steep nose-dive the following week. And for it to generate any large-scale publicity, it would have had to come in at #1, causing people (who are generally sheep in their movie-going habits) to say "Wow - it must be good if it's #1.")

But it had minimal advertising, at least that I saw (and I should think I'd be the sort of viewer they'd want to target) plus it had no name-brand stars... and the commercials gave no clue as to what it was about (except that it was set in space.) The advance reviews that they quoted were about the sfx being mind-boggling... but what I saw in the commercials was no more than a basic episode of "Star Trek." (Now the sfx probably were incredible...but that does no good if a channel-surfing, hard-core genre fan such as me isn't made to think they are.) Plus the title itself was not exactly enticing. Teen-age boys want titles like "Fast and Furious," not a word they can't even pronounce. Plus the lure of "...from the producer of 'Buffy' " causes the typical teenage boy to assume it must be a chick flic about valley girls and lesbians.

So that's my guess as to why it tanked. It certainly got good reviews.... although again the critics most commonly mentioned characterization and themes and deeper meaning...things that cause the average movie-goer to go "Snoooooozzzzzzz..."
August  - Jack's Pack Fan # 1, Keeper of the List, 3-Time Speaker of the JoAT Fan Quote of the Week, and the only person ever to have Back 2 Back Jack and Cleo fan quotes !
Reply
#12
Arcadia Wrote:I would have thought that Serenity would do better. I wonder why it didn't. When the DVD's of the series came out they sold well. The movie was pulled before I even got a chance to see it on the big screen.

The discussion on Serenity was actually quite revealing. I believe Mike of Quantum Muse stated that while he enjoyed the movie himself, it wouldn't/couldn't appeal to a new set of fans. The underlying problem was that while accessible as a fun scifi/western adventure romp, there was no connection with the characters. If the audience saw the television series and/or bought the DVDs, then sure, you knew why it was so emotionally devastating in parts.
To view, use your mouse to select the text:
Like when Wash was suddenly killed!

But other than that, you would just say, "Gee, that's sad." There isn't time in a movie to backtrack and show us why the characters are so cool and fun. That kind of development required investment that two hours of run time just wasn't going to acheive.

Entertainment Weekly did a very small (four paragraphs) on Joss Whedon's next move. He says Firefly/Serenity is effectively done, and he has closure, that he's working on other projects and he'll go in whatever direction those take him. It's unfortunate, from my view, because the series had great potential and clear depth. But there it is.
All your base are belong to us.

It could be that the purpose of my life is only to serve as a warning to others.
Reply

MYCode Guide

Possibly Related Threads...
Thread Author Replies Views Last Post
Star Disney’s The Lone Ranger movie starring Depp badlands 1 40 November 22nd, 2021, 02:07 PM
Last Post: Michael
  New C.S Lewis Movie bilbobaggins764 2 111 November 22nd, 2021, 02:00 PM
Last Post: Michael
  Remember the Don Bluth movie : The Secret of NIMH badlands 0 110 July 27th, 2021, 05:15 PM
Last Post: badlands
  This civil rights leader’s story should be a movie badlands 1 106 July 25th, 2021, 11:28 PM
Last Post: Alvin Eriol
  Ali Pourahmad directed an amazing sci fi movie Aliens Reaction scifilover 0 152 July 12th, 2021, 04:05 AM
Last Post: scifilover
  Does anyone recognize this movie? FilmGuy 0 96 July 4th, 2021, 12:45 PM
Last Post: FilmGuy
  Reboot of this old classic B movie The Last dinosaur badlands 1 174 March 28th, 2021, 12:18 PM
Last Post: Michael
  Edelweiss pirates movie badlands 1 199 March 21st, 2021, 10:58 PM
Last Post: Michael
  Colorized Movie - "The Phantom Planet" Michael 0 597 November 26th, 2020, 05:09 PM
Last Post: Michael
  Dungeon and Dragons movie again badlands 1 462 November 4th, 2020, 12:11 PM
Last Post: Michael

Forum Jump: