#3 on Lindy's Five Essential Websites (Non-Major Media) for 2013
[+] Team Summaries

Monday, September 10, 2012

Week 1 EPAs, RB and QB

The computer is crunching week 2 numbers as we speak, but in the meantime we take a quick look back at week 1. Brent has finished grinding through 10,000 rows of play by play data to bring us running back and quarterback EPAs. The EPA and NEPA methodologies are still works in progress, but when perfected they will arguably represent the best possible reflection of a player's performance.

In short, Brent attaches a point value to each play based on the change in the situation - down, distance, field position, possession - using the expected points for the team before and after the play.

At running back, the week 1 champion was Air Force's Cody Getz, who rushed for 218 yards and 3 TDs. Taylor Martinez led the pack at quarterback. 

7 comments:

  1. please get back to me. I wanted to know why your predicted box score changes after the game has been played? and whats the percentage mean?

    ReplyDelete
    Replies
    1. 1) When you are commenting a) do not demand a response and b) if you are going to represent something someone else said do so accurately. Personal pet peeves.

      2) That being said, it is the Expected Box Score, not the predicted box score. That's an important distinction. A prediction is a probabilistic statement about a future outcome. A expectation is a hypothetical average. If they played the game a billion times, the model "expects" that to be the average outcome. When new data becomes available, the model adjusts its expectations.

      3) A percentage is the expected positive outcomes from 100 tries. There are a lot of percentages on this site. In order to answer your question I need to know what percentage, specifically, you want interpreted.

      Delete
  2. Scott, Let me say first of all I did not mean to demand anything from you guys, nor was I trying to be rude or pushy. I do apologize if I came off sadistic or said anything derogatory about your site, no harm intended. I use your site a lot and have a great appreciation for it and the information provided herein. Sorry about the wording and the question I had about the percentage was after the game has been played and the result of the game has been posted by the expected box score it has a percentage. Sometimes it is a negative, other times positive. I do not understand what the percentage represents. Example: actual score 31-12, expected box score 34-13 +59%. That was all. Sorry that I came off as a rude #$%. I meant no harm or ill-will. Your site has great information through out and has helped me tremendously. This is just an attempt to understand it better for the guys like me who have a small cranium, lol. Thanks a million and I can't wait to here back. Have a good one my friend.

    ReplyDelete
    Replies
    1. No worries. I'm just a bit anal about a few things.

      That percentage is the wins added above expected. For example, if a team has a 50% chance of winning each of its games we would expect them to win 6 games over a 12 game season. If that team wins its first game we would now expect them to win 6.5 games (11*.5+1). That win then add 50% of a win above expected. Upsets add more (higher percentage) easy wins add less. Its just a quick way of comparing a team's season rating - how well the season has gone in terms of wins and losses - and its power rating - how good a team actually is.

      Delete
  3. ok, I get it now. Your just a little too smart for me. Thanks very much for answering that. It was really confusing. I appreciate your patience and keep up the good work my man. Much love for this site and the work you do.

    ReplyDelete
    Replies
    1. Glad I could help. And I do appreciate the questions - it means people are checking out the site - so I'm usually not a prick in my responses.

      By the way, I'm not smarter than anyone; I know the answers because I made the stuff up. Glad you appreciate the site and I hope you keep coming back.

      Delete
    2. Certianly will sir. You have bben very helpful. I'ts much appreciated.

      Delete