Thursday, February 4, 2010

How did we do: Edge of Darkness

So, how were our predictions for Edge of Darkness?

First, the OK news.  Our statistical model predicted an IMDB score of 6.7 and as of this writing, Edge of Darkness has a rating of 7.4.  We were a little off, but keep in mind that these ratings are from the first people to rate the movie (or in other words, these are people who saw it and were likely excited about seeing it).  We expect the score to lower in the next couple months, and then when it gets released for rental, we might be right on the money!

For now though, we'll cheer ourselves up and look at what we got right.  You might know that IMDB gathers demographic info from people who rate their movies so that different statistics can be broken out.  From here, we can answer the question, who do we best represent?  Judging from the image below...

... we represent Females Aged 18-29.  Awesome.  Maybe this will turn out to be a pattern.  See if the pattern continues next week.


And in our own self-interest, we bury the bad news after the jump.



The bad news.  We predicted $35 million for their opening weekend, and it actually got $17 million.  What the hell happened?  Well, technically we used the largest weekend gross as our measure (some indie films had small opening weekends but then exploded), so it's possible that Edge of Darkness has yet to reach its peak, but we doubt it.

More than likely, our model for movie gross is just pretty bad.  We think we need better data on how extensively a movie is advertised.  Maybe we'll try to look at number of blog/twitter posts there are a week before release.  Getting historical data on this could be difficult though.  We'll think of something, and if you have any ideas on where to get better data feel free to suggest them.

No comments:

Post a Comment