Recruiting footprints - how far away do women's BB teams recruit from their home base?

I like to say that the problem with women's basketball statistics isn't doing the analysis, it's getting the data.  Women's Basketball State (wbbstate.com) is a big help, and they'll even allow you to download their information in a much more workable format...if you're willing to pony up $100+/year.  Either way, for a lot of their data you'll have to figure out a way to cut and paste, and that introduces errors in whatever you're planning on doing.

Take, for example, the question of how far women's basketball teams go into the field to recruit players - do certain teams recruit close to home, and if so, which teams are those?  You need a lot of information to solve this problem:

1)  A list of every player associated with a team, and their hometowns
2)  A way to measure distance from their hometown to the colleges they attend
3)  A way to assign some sort of average distance without skewing the results

Even with the help of a website like WBBState.com, each of these steps can introduce errors. 

Roster is missing players? Results skewed.
Distances calculated incorrectly? Results skewed.
Don't do the math right?  Results skewed.

Even so, I thought this was worth doing.  I started this with 2012-13 data, which I painstakingly copied from the WBBState.com website.  Then, I had to find some way to assign recruit hometowns to a latitude and a longitude from which I could make calculations.  (You need decimal latitudes and longitudes, by the way.)  Then I needed to find an online source that had this information, then I needed to find a formula that could calculate it.  Then I needed to figure out a way to solve the errors that were generated.

You get the point.  The 2013-14 data is finished...and I finally have data based on the 2012-13 rosters.

So how do we determine the average distance? 

1.  Take the distance from the hometown of every player on the roster to that team's home school.
2.  Find the median distance, i. e. the distance that divides the roster into two halves.  If you have an odd number of players, it's the player in the middle, else, you interpolate between the two closest players on either side of the dividing line.

Even with super-duper power math, there are still some problems.  Take a city like Los Angeles, California.  Los Angeles includes a lot of area.  However, in order to calculate, you have to find some point in Los Angeles which we can say is the imaginary center of the town.  As you can see, another degree of error might be introduced.

So let's assume as a thought exercise that the results are correct.  What can one conclude?

Well...it depends.  The data could be interpreted any number of ways.  Suppose Big Time State recruits most of its players within 50 miles away from Metropolis, USA (home of Superman).  It could be that Big Time State has a lock on all the big name recruits in the city and doesn't need to recruit from far away if the talent is rich there.  Or it could mean the reverse - Big Time State can't get anyone to come from their city that lives there.  The same kind of argments could be made about schools that recruit far and wide.

So based on the 2012-13 rosters, here are the various schools and an indication of their recruiting footprint, in miles away from the school.  You might not draw any conclusions, but you might have some fun with the discussion.