| the ncaa needs some open source values |
[Dec. 1st, 2008|06:46 pm] |
College football in the United States is a perfect example of a small group of people who have huge power, and who practice monopolistic and proprietary software business practices in order to keep their power, because the power equals insane amounts of money that they would rather not share with others, even if others are deserving.
Every December, sports fans like me get all fired up over the fact that college football doesn't have a playoff system. Failing a playoff system, we are stuck with a bunch of human voters and computer algorithms that come together in order to determine the rankings that are used in setting up the "championship" game.
Every year, the system fails in some horrific way that leaves at least one (and sometimes more) schools justifiably angry and slighted.
There are three specific areas that need full transparency:
(1) How the pool of people (including the maintainers of the computer programs that get to vote) who are allowed to vote is selected, and what the criteria is for both getting a vote, and keeping your vote.
(2) The manner in which each person votes, every single week, as opposed to only the last week, which is the way it is set up now.
(3) The source code to each of the 6 computer algorithms that are also used to generate a ranking.
Jim Rome and ESPN can debate college football forever, but until these three requirements are met, the whole thing is a complete sham. Lucky for me, I went to a college that has never, and will never, be good at football. As such, my anger can be purely philosophical, and I'll never have to actually watch my team get cheated by the system. I know there are plenty of injustices in the world that are far worse than this one, but it just really annoys me.
I should start the "OpenBCS" project that tries to reverse engineer the current computer algorithms, and then uses the community to improve them from there.
But really, we just need a playoff like every other sport in the country has. |
|
|
| Comments: |
... you just have the same lack of transparency in your playoff-picking committee. For example, the basketball selection committee makes closed decisions on a variety of subjective criteria, and there is hand-wringing about what teams don't get in.
I suspect the main difference is that none of the teams that complain about not getting into the basketball tournament really have a shot at winning it.
AFAIK, only one of the BCS computer algorithms is documented (http://www.colleyrankings.com/method.html).
![[User Picture]](http://l-userpic.livejournal.com/82413996/10036337) | From: ![[info]](http://l-stat.livejournal.com/img/userinfo.gif) spevack
2008-12-01 07:33 pm (UTC)
Re: Ah, but then... | (Link)
|
Obviously you'd have to be transparent all the way to the bottom of the "who selects the selectors" tree, and if there was a playoff, equal transparency would be needed in how those teams are chosen.
The basketball system has the same flaws, but as you point out, they are less egregious because no one who really has a shot at winning gets left out. I think a 16 team playoff for football would probably at least bring you to the same level of grumbles as basketball gets.
Some credit to Colley for publishing the mathematics behind his method. But that still isn't source code and data in a format where it can be reproduced.
Edited at 2008-12-01 07:38 pm (UTC) | |
![[info]](http://l-stat.livejournal.com/img/openid-profile.gif){kind=small}