Questions about team rankings

[Matt Mills]

Hey everyone, 

  

          Lately I have been doing some analysis of 3rd downs, mainly what is the conversion % by distance and by play type. I want to put together a team ranking based on a team's 3rd down success rate compared to the type of third downs they faced. I am just unsure of the best way to go about that. At first I simply wanted to take a team's total 3rd down attempts by distance and multiply that by the expected conversion % based on all 3rd downs from 2008-2012. But then I figured that may give team's who just have more third downs a bias, if they are slightly better than average over more third downs they would "outrank" a team who was way above average but just didn't see that many 3rd down attempts. Have any of you guys encountered this problem before, and if so did you do anything about it? I am open to any suggestions or ideas, just trying to get some thought going on it. Oh and I have attached a graph of the summary of my 3rd down results. It shows 3rd down conversion % by play type and by distance to go. If you have any questions on how I got there feel free to ask over this thread or just email me personally. I plan on starting an article on Football Study Hall's web site tomorrow explaining how I got here as well. I wasn't sure if this has been done before, if any of you have then I would love to talk about it, I just wanted to get my hand's dirty in the data. 

Thanks, 

- Matt