UFR Database

This has been an extremely long offseason for me up to this point, and I've been struggling to find ways to get my Wolverine Football fix, as I'm sure just about everyone that frequents MGoBlog has. My internship has also been extremely slow, so I've had a lot of desk time to sit and do nothing.

While perusing the site I began to wonder if there was some sort of UFR database that could be queried in order to research different trends, etc. I'm a computer engineering major with a math minor and statistics are not exactly my cup of tea, but I would at least like to poke through that information if it was all in one place.

After emailing Brian and finding that there was no such database, I found myself pondering how one might go about making such a database. I started writing an application in C++ that would take the HTML table and convert it into XML that could be imported into Access. I was relatively close to getting that to work, but it was a bit of a hassle, so I decided to try something else.

I decided to write a Perl script that would parse the HTML, clean out all the tags (<table>, <tr>, <td>, etc), and put in their place more useful XML tags (<line>, <player>, <analysis>, etc). The result was a pretty nifty script that does almost all the work for me. After converting each UFR table into XML, I needed to add some extra information like drive IDs, play IDs, and things of that nature so that the database would be easily queryable. I did some of this by hand (lots of search and replace), and wrote another script to take care of a bunch of the nit-picky stuff that would have taken hours. I didn't really want to spend the rest of the day adding <PLAYID>some number</PLAYID> to over a thousand plays... I decided to be an engineer because I'm lazy and there's ALWAYS an easier way to do something.

If you're interested in seeing the Perl scripts, especially if you're proficient in Perl, let me know and I'll send them to you. This was my first useful Perl script (I taught myself, and hadn't ever done anything more significant than Hello World!) and a lot of my techniques came straight off a Google search and probably aren't following best practices. Suggestions/critiques would be greatly appreciated.

After some trial and error, I ended up with what I think is a pretty nifty database that includes information about every game, drive, and play that was charted over the course of last season. OSU is missing because Brian didn't do a UFR, along with halves of some games (a couple were only charted for offense or defense, but not both), and some drives (I believe a couple games were so painful that the final drives were non-UFR-able).

I've hosted the database file on FileFactory, along with the final XML file for the complete season. Hopefully that works...

If I get really bored in the future, I'd ideally like to find a way to parse the text to record all the +/- information for each player, but I haven't thought about that much... seems like it would be quite a bit of work, and I might not have time for it now that my boss and my manager are back from vacation.

So anyway, feel free to download and look through that. Do whatever you'd like with it. I didn't really do this for any reason other than I wanted something to do, and it seemed more useful than Facebooking all of our potential recruits for the next five years. Sorry about all the parentheses... no idea how they've become so prevalent in my writing over the years, but everything sounds choppy when I write without them.

Go Blue!

---

EDIT: As was mentioned in the comments, the Miami game, the Northwestern game, and the other half of the PSU and Illinois games were actually available. These have been added to the database and the final file is located HERE. (The corresponding XML has been updated as well.)