Baseball Wins vs. Attendance, which fans are loyal? fair weather?
I discovered the Baseball-Databank, a database of wonderful baseball stats that you can download. So now my math+data geek and baseball geek can merge until an uber dork. So don’t be surprised over the next several weeks as the baseball season is about to get underway and hope is in the air, to see plenty of stats about our national pastime.
First up, in today’s post I plotted a Wins vs. Attendance correlation for each team. This allows us to answer a few interesting questions: do wins contribute to attendance? who has the most fair weather fans? most loyal? Does a new ballpark have a bigger impact on attendance than wins?
How to read the graphs
The x-axis is wins and the y-xaxis is attendance. So the top-right corner is a high attendance and a high number of wins. The bottom left low attendance and low number of wins. A loyal team plot would look like high horizontal line on the graph regardless of number of wins attendance stays high. A fair weather team would have a diagonal line, as wins increase, so does attendance. A disloyal fan base would have a low plot regardless of the number of wins.What stories does the data tell
A couple of interesting things I gleamed out of the data:
- The San Francisco Giants fan base was a pretty fair weather bunch until the new stadium in 2000 and then became one of the most loyal.
- It appears the Minnesota Twins are the most fair weather fans, which seems appropriate for the Minnesota weather. The plot for the Twins is almost a perfect diagonal line, high correlation between wins and attendance.
- The Baltimore Orioles graph looks to me as the most optimistic fans, there is a extremely step diagonal line, it is almost like they are saying, please just give us a few wins and we are there.
- A few plots that are confusing to me; the Texas Rangers, Seattle Mariners and Colorado Rockies both have a surprisingly consistent number of wins, but attendance varies greatly.
- As far as most loyal, tough to call but it might be the Atlanta Braves, it looks like they have a good base of attendance at 2.5m. The least loyal is fairly obvious, the Florida Marlins, regardless of some fairly high win seasons, the attendance does not go up.
Take a look, what stories can you tell?
About
- Data source: Baseball-Databank.org, Seasons 1990-2010
- Processed and graphed using R Language
- Thanks to Michael Driscoll’s Dataspora Blog for inspiration, in particular his article on Sabermetrics with R