[EDIT: here’s a more specifically dishonest and high profile example of someone claiming a graph to show something that it doesn’t]
So, I saw this chart…

… on video games (and stuff) weblog Wonderland where it’s used to support the notion that

collectively, movie companies really have no clue about what works

Alice got the graph from The Long Tail weblog where Chris Anderson opines

This chaotic mess of a chart below says it all. Like the wise man said, in Hollywood, despite the bluster about track records and taste, “nobody knows anything.” It’s all a crapshoot.

Chris got the graph from this report (pdf) which I haven’t read yet but looks quite interesting.

Anyway, I don’t disagree with Alice and Chris’ point that “Quality is also an issue in bigmedia-generated content”, hell, of the billion or so screens at my local Vue cinema not one of them is showing a film that I have any interest in seeing, especially not for £9 (maybe Die Hard 4 for old times sake but I’m prepared for it to be dire). What I do disagree with is the idea that this chart supports that.

First off the chart doesn’t include any information about the average cost of the films. I mean these Hollywood guys are presumably making their decisions based on a cost benefit analysis of some kind: “OK, we’ll probably make $80 million off this so I’ll spend $15 million to make it, wheras this one’s only likely to bring in $10 million so I’m not parting with any more than $1 million”. The chart tells us nothing about profitability of the films a studio makes. Disney’s apparently stellar performance in 2006 could really be nothing of the sort if the films the made cost on average $90 million each.

Second, the chart shows average box office returns for each year. i.e. total box office returns / number of films. It’s entirely possible that in 2006 Universal chose to make more films with lower budgets than they had the previous year, the chart gives us no information about their total earnings only that their films on average made less money. Their overall box office income may have remained static. From this chart there’s just no way of telling.

Third, people only have finite amount of time and money to spend on going to the cinema. If the six studios shown on the chart released between them 100 absolutely brilliant Citzen Kane beating films next year I just wouldn’t have enough time or money to see them all, no matter how much I wanted to. Looking at the average of each years averages i.e. the average box office for all the big studio films that year you get a much steadier picture. Reverse engineering the figures out of the original chart and then doing the maths I get each years mean lying between about $41 million and $47 million with a standard deviation of about 2.5. Quite stable really, so maybe another reading of the chart might be that all the people commissioning films for these companies are brilliant geniuses who know exactly what the public want and deliver pretty consistently but because of the aforementioned time and money constraints the public can only see a pretty much random selection of the films on offer. Hmmm…

OK I admit that last one’s not likely the explanation either but you see what I’m getting at: That the chart doesn’t support the argument that the bloggers are making. Even though I happen to agree with them, this annoys me as it does a disservice to the position that big business has no monopoly on quality creative content.

OK, I think that went fairly well, I was worried it was gogin to turn into one of those massive un-finishable posts that just sits in the drafts folder for 3 years. Here are some of the tangents I have successfully avoided going off on during the course of this post:

• The graph itself is bloody awful, a good potion of its chaotic appearance comes from its poor design.
• Something about false origins/ scale on graphs and how it can change your perception of the figures.
• Covert indie elitism.
• Covert hip-hop hatin’.
• Post-rock Taleban.
• People who don’t understand significance testing.
• DJ Dangermouse’s Grey Album, seriously it’s 4 years ago now and no one’s managed to com up with anything better than that.
• Inherent problems with predictability of quality of creative output c.f. Ang Lee.
• How effin amazing it’s going to be when I’ve spent a months salary on ATP tickets.
• Michael Moore and his tricksy editing and out of context imagery.
• Michael Moore and his covert xenophobia in Fahrenheit 9/11.

Please feel free to ask about them or talk about them in the comments.