Subtitled: that there is minable data
We are no longer stuck relying on a single grumpy film reviewer in the local paper or the thumbs up thumbs down guys on TV. Film reviews are everywhere and a couple of smart visitors to rottentomatoes.com notice that there was minable data hiding behind those reviews, specifically mapping actors and directors to the Rotten Tomatoes rating for their movies.
It started with this observation on marginalrevolution.com:
Pulling a little more data in (and unfortunately without the trend line) Bing and Singer-Vine from Slate.com give us this:
… and this high level view:
source: Christopher Beam and Jeremy Singer-Vine, Slate.com
Why bother mentioning it here? It’s a great example of
- visualization of a complex topic
- empirically representing data based on fundamentally on judgement and opinion
- why trend lines/curve fitting/etc can be useful in interpreting data (I wish this tool had a button to do that)
- math and pop culture intersecting
- critics dont like Michael Bay very much and aren’t giving
I can see math teachers from elementary grades where the kids are just starting to create graphs to entry level statistics classes making some connections with their students and what they are working on.
Scroll down for the interactive bit and put in a couple of actors. What is interesting is that if you ignore the extremes (mentioned at the end of the article) the most prolific actors are all over the map and the lines normalize to be just about flat around 50%
It is also interesting to see the sawtooth pattern that emerges with so many prolific, especially comedy, actors. They produce a good one followed by a dud, almost as a rule.
On the other hand directors put out far fewer works so their graphs aren’t as compressed and really do produce more of an arc.