I added the following features with regards to NCAA tournament bracket projections:
1. ON the main bracket page (www.rpiforecast.com/bracket.html), I added more stats: Current and Expected RPI, Expected OOC RPI, Current and Expected Records, and Current and Expected Conference Records
2. The daily projected tournament seed can now be found on the individual team pages (for example, see Duke: www.rpiforecast.com/teams/Duke.html)
3. I've added a graph on the individual team pages below the other two graphs which tracks the history of projected seeds for each team. You will be hard pressed to find others willing to archive the history of their projections.
Next on the to-do list is BracketBusters and Conference Tournaments.
Subscribe to:
Post Comments (Atom)
6 comments:
I think there's a problem with the seeds on your page. There should be either 4 17-seeds and 2 16-seeds or 6 16-seeds. However there would only be two 16-seeds, because two 17-seeds would play the 1-seed and the other two 16-seeds would play the other two one-seeds...
Same with the 12 and 13 seeds. I admittedly do not know how this works, but instead of 4 12 seeds there are either 6 12 seeds or 5 12 seeds and 5 13 seeds.
Hope that makes sense. Basically, there are not 8 teams who are 16 and/or 17 seeds - there are 6...
Good work on the site, I visit it after every Arizona game!
Thanks. You're right. The problem is that it isn't clear which seeds will be considered the last at-large seeds at this point. Definitely should only be 6 16 seeds though. I'll change it.
Thanks for visiting.
how is BYU # 1 on your team forecast page, yet only a 7 seed in your bracket projection page? Doesn't seem to follow.
Yosh, good question. The RPI is only one small factor that goes into determining seeds. The selection committee looks at a number of things. The at-large bids and seeds are based on Jay Coleman's dance card formula. The short answer is that controlling for all other factors, the selection committee has given MWC teams lower seeds than teams from all other major conferences. You'll notice that all MWC teams have lower seeds than you might expect based on RPI. You can go to his website or read the paper if you want to see the formulas. If you look at BYU's "at-large" ranking you'll see that they are much higher than many of the teams seeded higher.
I recently read Coleman, DuMond, and Lynch's article in Managerial and Decision Economics. I also have their older Dance Card article. My question is which Dance Card model do you use for your site? There are five "specifications" that they listed, so I would presume it is one of those five. Obviously, it is their work so no need to post the equation, just if I already have purchased their article, which one are you using. Thanks for your help!
AMKroger,
Sorry, I just saw your question. I have been using equation number 3 (which includes conference affiliation and representation on the selection committee). However, just today, I switched to equation 1 which does just about as well in terms of picking teams, but not quite as well (in-sample) at seeding. This was mainly because of the problem with the MWC teams seeds (BYU, SDSU, etc). Because the MWC has never been this good, it may not make sense to assume that the same "bias" would exist against them this season.
Post a Comment