BK Smith at 2013-05-06 18:40:31:

An alternate argument completely discounting Bruzzese's formula is that there's *already* a formula out there which Hollywood uses to reintegrate successful components and minimize the creative gamble of big budget filmmaking: Sequels, and more sequels.

nayan1875 at 2013-05-06 19:08:30:

This is horse shit. Any exec who thinks they can research their way to a successful or unsuccessful film by focusing grouping a script is just doing the usual CYA. That's not to say there's no merit in trying to fix script issues, but this takes writing by committee to a whole new level. Movies succeed as a whole and fail as a whole. You can't pick apart what scene worked and what didn't work and try to stack the deck with out of context and stupid check lists like this.

Steve Turnbull at 2013-05-07 04:15:25:

Every now and then someone claims they've invented a system for picking a winner for horses; roulette; the lottery; or [insert favourite gamble here]. The only person who gets rich is the one claiming to have the system.

Dennis_Fischer at 2013-05-07 06:59:26:

Here is a similar approach by the UK-based company epagogix: "The Formula - What if you built a machine to predict hit movies? By Malcolm Gladwell": http://www.newyorker.com/archive/2006/10/16/061016fa_fact6?currentPage=1 What do you think about that?

pgronk at 2013-05-07 09:23:02:

Statistics don't lie, folks. But they are easily misused, a potent way to blow smoke. Hollyweird suits need to take a class in Statistics 101. All that statistical analysis shows is correlation. But correlation is NOT the same as causation. And every statistic carries with it associated quantifications of variation and confidence; IOW: numbers that inform how much faith and money can you bet on that number to be reliable and predictive. "If it’s a targeting demon, you are likely to have much higher opening-weekend sales than if it’s summoned." Hollyweird suits ought to be kicking the tires of that conclusion with questions like: How likely is "likely"? How much higher is "much higher?" What assumptions inform the conclusion? What's the null hypothesis and was the proper statistical methodology used against the null hypothesis? What is the underlying sample size that is the basis of that conclusion? What are the associate measures of variance to indicate how much confidence one can have in that statement being accurate and predictive? The New York Times reporter should have asked the questions, too.

Jeff Messerman at 2013-05-07 10:01:19:

This has the markings of one of Scott's 'Movie Idea a Day' entries. "A snake oil salesman pitches his tents in Hollywood and convinces all the power brokers that he has a surefire system to create box office gold." I'm left with the slightly bemusing image of "The Big Opening Weekend Machine" wherein Beelzebub, er, pardon me, Bruzzese simply feeds a draft of a script into one end of the machine and a guaranteed 1.5 Billion Dollar Worldwide Gross script pops out the other end.

Scott at 2013-05-07 13:02:06:

pgronk, thanks for this. I'm so not a numbers guy and your points drive home a point I implied, but didn't articulate. "Correlation is NOT the same as causation." Perhaps not sexy enough to be a bumper sticker, but one helluva truth when it comes to all things screenwriting related.

brad_sweet at 2013-05-07 15:58:23:

Agreed. Management consulting companies like Deloitte, Accenture and BCG have been getting rich off of Fortune 500 CEOs paying them fat fees to conduct studies confirming that the leader's strategy is sound. Standard CYA exercise so that if the strategy isn't sound, or if market conditions change unexpectedly, the CEO can shift blame somewhere else. Bruzzese is essentially charing $20k for studio execs to take out an insurance policy on their careers...

Dennis_Fischer at 2013-05-07 16:15:20:

Scott, I totally agree with you. There are only patterns and rules/techniques/methods used by the writer and combined/blended with her/his individual imagination to form a unique, universal and emotionally compelling story.

Jake MLB at 2013-05-07 20:58:12:

Those quotes by Bruzzese are indeed comical. What if a demon-targeting script has a bowling scene? What he's describing is often referred to as 'data mining' in the statistics world. It may be easy to describe what elements are present in past successful scripts but it's likely that none of these variables will prove predictive going forward. Beyond that, you need to control for all other variables (as hinted at in my question above). To create a predictive multivariable model would literally be impossible since there are so many variables that go into a script. But, that doesn't mean that there isn't value in this kind of work (if applied correctly). It may help to describe trends in current film. But it won't reveal the formula to future success.

JOÃO NUNES | Vinny Bruzzese | Será possível prever o sucesso de um filme? Vinny Bruzzese diz que sim. | cinema, Hollywood | at 2013-05-13 18:06:13:

[...] Craig Mazin foram dos primeiros a manifestar-se no podcast que publicam regularmente. Também Scott Myers, guionista e blogueiro, deixou a sua opinião no seu site. Mas uma busca na net encontra [...]