After reading "The Importance of the 30 minute Population Radius on MLB Attendance" on Fangraphs - http://www.fangraphs.com/community/the-importance-of-the-30-minute-population-radius-on-mlb-attendance/ I decided 2 things:
1. I would use Geographic Information Systems (GIS) to run this process using more current US Census Bureau data (2010 as opposed to 2000).
2. I decided to create a blog to share this information. I plan on running similar analyses in the future and will take a deeper look at demographics and the relationships between Geography and Baseball.
An interactive map displaying the results can be found here: http://bit.ly/1EOz6B5
Table with results: https://docs.google.com/spreadsheets/d/1oI-uAU0CjxNgtAEieNOUmayXJemIw3e_WXdivojQ2IU/edit#gid=0
The numbers are quite a bit higher than the original study referenced in the Fangraphs article. Part of this might be due to the fact that I did not use rush hour traffic as a variable. I might take that into consideration the next time I run this analysis. Using ArcGIS and Network Analyst I used the process below:
1. Identified the locations of all 29 ball parks.
2. Using Network Analyst, I used the Generate Service Areas tool to create a 30 minute drive time polygon to the identified ballparks. For the most part I used the default settings, and these polygons do not account for rush hour traffic.
3. Using the Union tool, I joined the 2010 Census data with the polygon drive time polygons.
4. I then subtracted out the FIDs with -1 to remove the polygons outside of the 30 minute distance.
5. Used the Field Calculator to obtain population totals.
