Previously, I explored how to find the Golden Age of television by analyzing IMDb data. After I wrote this, Gabriel Rossman, an assistant of sociology at UCLA, noted that it would be interesting to see what this chart would look like if broken down by broadcast network. With Rossmans help (he also had been playing with the IMDb data), I was able to examine the average television show length over time, in number of episodes, for the different networks. A chart with the results is below:

Clearly, in the past decade or so, the number of episodes per show has calmed down quite a bit. My hunch is that this is due to the larger number of shows that are currently being aired, of which many get canceled, lowering the mean number of episodes. But if you look further back in time, you can see that network show longevity has hints of being a zero-sum game. When one network has long-lived shows during a time period, such as ABC in the mid-1980s, another network might have a series of duds, such as NBC during the same time. This is reasonable, as there is only a finite amount of collective attention that we can lavish on television watching. Therefore if one network is doing well, it is not surprising that others do more poorly. Of course, this metric is far from perfect, but it can be used to test further hypotheses.

You can also notice other things in this chart, such as even historical events. For example Rossman has pointed out to me that there are hints of the 1988 writers strike, displayed as a precipitous drop in show length, though the 2007 strike is invisible.

As Ive already noted, we are in the middle of a great time for television. Maybe its also a great time for the applied mathematics of television as well.

Top image: urbanora/Flickr/CC-licensed

Visit link:
Television Show Longevity by Network