Is This The Most Stupid Of The Stupid White Papers On Film This Summer?
It’s early… and no doubt, Deadline and The Wrap will publish stories all August long giving credence to a parade of moronic insights from the industry analysts working for the money movers. But this could be the early winner, with such a lead on stupidity that it can’t be caught from behind.
The Hollywood Reporter published this doozy on Monday without a hint of disbelief and it escaped my notice until AV Club decided to mock it, but not clearly debunk it. Japanese Scientists Produce Mathematical Model to Predict Box Office From Social Media Activity
Every year, someone comes along to claim to be able to predict box office based on pulling a feather out of their ass in some way. But this one is misses on its fundamental conceit… the idea that heightened awareness, whether an increase in ads (which 98% of wide releases choose in the week of release) or Social Media buzz, can be used to predict box office results in anything but the broadest ways. (And there is plenty of proof that even broad assumptions based on Social Media are inaccurate most of the time.)
Should we be taking an analysis on Social Media and box office popularity that is using movies from 2006, pre-Twitter and very early in the history of Facebook, seriously? (Were they studying AOL chat rooms?) And then they leap to Asian openings from this year, where the marketing and publicity strategies can be quite different.
It’s really simple. Garbage in, garbage out. Lack of insight into the market they claim they have a new tool to analyze combined with limited, mismatched info = stupid.
I know the news cycle is relentless. These white papers can be news since they are being offered by brokerage houses and might influence how people invest. But why any journalist or journalistic outlet would take the provocative finding from these papers and turn them into headlines without analyzing what’s been said is just beyond me. Shocking.
Does the Hollywood Reporter or any reporter there really believe for a second that advertising only matters the week before opening and that Social Media buzz would – with very few exceptions – exist to be measured without the studio marketing departments… must less the silly suggestions in this paper? Show of hands!
It’s a rhetorical question. Everyone at THR is smarter than that.
And yet, they printed it.
Oy to the 8th power.
Ugh… the reporting here is abyssmal.
The paper does not “predict box office” – Hollywood Reporter and Deadline got that wrong. What it DOES do is model a curve of blog postings regarding a movie – note that this requires BOTH pre-release information AND post-release information. Similar to the also-badly-reported “Twitter Predicts Box Office” paper, both give correlation to social media activity that is useless for anything more than a week out from collection. Standard tracking methods are better at long-term prediction than either of these methods.
Note that in this latest paper, the modellers state that only the SHAPE of the curve is similar to box office – 100 blog posts for Movie A might translate to $Xmillion within the day or two target window, while 100 blog posts for Movie B might translate to $Ymillion. Also note that correlation is not causation – it may be possible that people could read these blog posts and decide to go see the movie and that accounts for the change in box office, or it may be that these people have paid money to change the box office and are now blogging about it.
What the researchers specifically state that this model is useful for (and it is NOT “predicting box office”) is that knowing the shape of the curve gives you an indication of inflection points – places where a hit of advertising dollars may affect blog posting/attendance the most. This is a SPENDING model, not a REVENUE model.