NFL Week 8 Elimination Pool Pick

Ok, the model says to go with the Chiefs at home against the Bills this week. However, I am personally not going to listen to my model this week and I'm going with the Jets instead. My model was going to save the Jets for week 17. But with the best teams resting players in week 17, I don't think it's worth saving one of the few best teams till then. Also, the Jets are coming off a bye week and the best teams tend to win coming off bye weeks. The winless Bills are really due for a win following their OT loss to the Ravens last week. So, in summary, my week 8 pick is the Jets.

NFL Elimination Pool - Week 7 Pick

6 Weeks down and 11 to go! As a reminder, I have already used (team's ESPN power ranking shown in parantheses):

Week 1 - Bears (12)
Week 2 - Packers (15)
Week 3 - Patriots (3)
Week 4 - Falcons (7)
Week 5 - Colts (5)
Week 6 - Steelers (1)

So I still have some good team's left (Notably Jets, Ravens, and Saints). This week there are two potential blowouts on the schedule for me to choose from: Bills at Ravens is the worst team playing the 4th best team and Browns and Saints is the 3rd worst at the 6th best team. The Saints I want to save for week 11 when they have the largest projected blowout on the calendar of the teams I still have available. So, I'm going with the Ravens this week as this is their easiest game left on the 2010 schedule and no other future weeks when they would present the best choice as well.

NFL Elimination Pool - Week 6 Pick

Week 6 is an easy decision for my model. We're going with the Steelers (ESPN's #1 power ranked team) is facing the lowly Browns at home. The only other option I'd even consider is the Giants to defeat the Lions at home. However, the spread between the Steelers and Browns is wider by 15 ranks. By the way, if I still had the Bears available, I would use them over the Giants (but not over the Steelers).

MIT Sloan Sports Analytics Conference

Well, fantasy basketball season is over, so I guess this isn't too relevant. But nonetheless, I have not posted in a while and that's because I've been doing fantasy research. A friend of mine, Paul Raff, and I entered this year's research paper contest at MIT Sloan's Sports Analytics Contest. We were extremely excited when our paper was selected as one of 4 finalists in the contest and we were invited to present the paper at the conference. Below is the abstract of the paper and you can find a link for the paper in its entirety here. Presenting at and attending the conference were great experiences and I am already looking forward to next year's conference and thinking about a new entry for the 2nd annual research paper contest.

In head-to-head fantasy sports leagues, it is common belief that managers try to do their best in all statistics categories. In this paper, we work to turn this notion on its head and investigate the strategy of effectively forfeiting certain categories while focusing on a certain subset of all categories. Through millions of draft and match-up simulations based on 2008-2009 NBA statistics, we found that approximately one-quarter of all possible subsets yielded strategies that defeated the "all statistics" strategy in a head-to-head match-up, and that the "all statistics" strategy is not the overall best one.