OT Thesis Help

Submitted by mstorm88 on

So I'm a senior econ major at GW and am writing my capstone paper this semester. I'm trying to write it on how recruiting ranking fare in terms of predicting team win percentage. Its supposed to be a heavily quantitative paper, and as of now I plan on including the number of 2, 3, 4, 5, and NR star players(using Scout and Rivals) by class for each of the BCS conference schools as my regressors, while also including both time and entity fixed effects.

I've been thinking of how to refine/improve this topic, and figured that there's no better place than here to look for help. Anything you guys post will be much appreciated.

TomVH

January 22nd, 2010 at 7:41 PM ^

Thesis Statement:

If you have 3 star recruits, you win 60% of the time, every time.

You're all set, sit back and watch the "A" roll in.

Don

January 22nd, 2010 at 8:20 PM ^

What's your winning percentage now?

And how do you factor in coaching ability? And how does this fit into an econ degree (just curious)?

Is this George Washington U in St. Louis?

StMirhza

January 22nd, 2010 at 9:39 PM ^

...We're just Washington University in St. Louis; no "George". GW is in Washington DC.

I would like to go on record saying that Wash U (STL) should change their name to avoid just such confusion. My life would be much easier if I didn't have to explain to everyone that I don't go to law school in either DC or the pacific northwest.

maizenblue92

January 22nd, 2010 at 9:54 PM ^

better players? You win more, it is simple. Look at USC in 2003-2005 where they signed a shit ton of 5* players and won a shit ton of games. Or look at the best SEC schools. Now if you need to argue a counterpoint use ND, and say that that was bad coaching.

pullin4blue

January 25th, 2010 at 2:03 PM ^

I would be interested in seeing the recruiting rankings as a predictor in future player success. For instance, are the 5* recruits always the best? Looking at the info available on past Heismann winners and past number 1 NFL draft picks, the list is not populated by all 5*'s. Why is that? Coaching? Teammates? DNA that makes someone mature later rather than earlier?

mhwaldm

January 25th, 2010 at 2:30 PM ^

One interesting topic might be evaluating the quantity vs quality strategies. For instance, in the 2010 class thus far, USC has pulled 14 total commits to Oklahoma's 29. But the avg star rating for USC is 4.0 while the avg star rating to Oklahoma is 3.52 (as per rivals). It might be interesting to determine which of these strategies is historically more successful. Of course this might be harder than it sounds, as teams may not utilize this strategy on a consistent basis. I suppose you could base ur research on team performance 2/3 yrs after each particular class.

dtod

January 25th, 2010 at 2:35 PM ^

In my opinion players should be weighted on how much they contribute and how many years they have been in school. A senior 5-star recruit should be expected to produce much more than a freshmen 5-star.

I also would not use class rankings for your analysis. They do not take injuries, transfers or non-qualifiers in to account. They also count players twice who went to prep school (Slocum in counted in '05 and '06).

I'm sure there are many more issues such as these that will arise, but these are two things I have always noticed.