UC-Irvine statistician advocates boycotting the BCS

Slate’s Bill James has a great article on the failures of the BCS system. Most of James’s points are based on a 2006 paper (requires login) in the Journal of Quantitative Analysis in Sports by UC-Irvine’s Hal S. Stern.

I couldn’t access Stern’s original article, but James summarizes his main points in the Slate piece:

The problems with the BCS are:

1. That there is a profound lack of conceptual clarity about the goals of the method;
   
2. That there is no genuine interest here in using statistical analysis to figure out how the teams compare with one another. The real purpose is to create some gobbledygook math to endorse the coaches' and sportswriters' vote;
   
3. That the ground rules of the calculations are irrational and prevent the statisticians from making any meaningful contribution; and
   
4. That the existence of this system has the purpose of justifying a few rich conferences in hijacking the search for a national title, avoiding a postseason tournament that would be preferred by the overwhelming majority of fans.

The excellent play of the teams at the top of the polls this year has illustrated the need for a college football playoff. Despite the legitimate cases for USC, Texas, and Utah to the national championship, it will be decided by two other teams (who also have legitimate claims on the title) in a one-off game that will settle nothing.

James does a nice job elaborating on Stern’s objections, as in this comment relating to 1.:

There are several things that a ranking system could do. It could rank teams based on their accomplishments over the course of the season—whom they played and whom they beat—or it could rank them based on the probability that they would win against a given opponent. It could rank teams based on how they have played over the course of the season, including perhaps in some early-season games against teams that were not quite sure who their quarterback was, or it could rank them based on how strong they are at the end of the season. It could rank the teams based on consistency, or it could rank them based on dominance.

Which of these is the goal of the BCS system?

Nobody has any idea. It's never been debated. There is a perception among the people who are in charge of this monkey that if you just turn the rankings over to a computer, the computer will figure those things out. The reality is that it can't. It is very difficult to objectively measure anything if you don't know what it is you are measuring.

Update: it appears that the list of objections to the BCS come from James, not from Stern, as I originally stated. -John

Wordle as research tool

Lee Sherlock has posted some ideas for analyzing Wordle text clouds.

Wordle opens up some "new," and potentially surprising, combinations and juxtapositions of terms that can lead to new avenues for analysis. Out of this cloud, I might pull out "cyborg spaces" or "computer symbolic construction" or "political-economic transparency" or "Visible surveillance architectures" as potential (re)combinations. Some will be more useful than others, but this experiment might point to some of the gaps in how we define and employ these terms.

Plus, it looks like a submarine.

Free? Language and hidden costs

While traveling to and from New Orleans for the 4Cs conference last week, I got to catch up on my magazines and ran across this article by Wired senior editor Chris Anderson.

In the article, titled “Free! Why $0.00 is the Future of Business,” Anderson argues that improvements in technology have driven down the cost of serving individual web users, meaning that many products and services, both on and offline, can be provided to consumers virtually for free.

In the traditional media model, a publisher provides a product free (or nearly free) to consumers, and advertisers pay to ride along. Radio is "free to air," and so is much of television. Likewise, newspaper and magazine publishers don't charge readers anything close to the actual cost of creating, printing, and distributing their products. They're not selling papers and magazines to readers, they're selling readers to advertisers. It's a three-way market.

In a sense, what the Web represents is the extension of the media business model to industries of all sorts.

The way this works out in most instances is that companies provide services to users for little or no cost, and then make their money using one of the following models:

• “Freemium” – What’s free: Web software and services, some content. Free to whom: users of the basic version.
• Advertising – What’s free: content, services, software, and more. Free to whom: everyone.
• Cross-subsidies – What’s free: any product that entices you to pay for something else. Free to whom: everyone willing to pay eventually, one way or another.
• Zero marginal cost – What’s free: things that can be distributed without an appreciable cost to anyone. Free to whom: everyone.
• Labor exchange – What’s free: Web sites and services. Free to whom: all users, since the act of using these sites and services actually creates something of value.
• Gift economy – What’s free: the whole enchilada, be it open source software or user-generated content. Free to whom: everyone.

While I think this list is helpful, and I agree with Anderson that the internet has created interesting new business models, I was a little put off by some of his examples.

Consider this sidebar on Ryanair’s low-cost flights around Europe. Anderson explains that the airline can sell tickets for $20 even when the flight costs them $70 per passenger because the difference can be made up by selling premiums such as food or the ability to check extra baggage. However, one of the “optional” expenses listed in the article is a $6 credit processing fee. While Anderson presents this fee as an extra, as if consumers could choose to pay it, it seems to me to be a hidden cost—an example of “Gotcha!” capitalism—rather than a optional service.

Similarly, Anderson’s description of Comcast’s “free” DVRs is a little off. Anderson claims that “Comcast has given about 9 million subscribers free set-top digital video recorders,” and that the company makes it money back by charging users a monthly fee for the box. But Comcast customers can’t keep their equipment, so it’s not quite accurate to say that the DVRs are free. Rather, customers are hit with hidden “installation fees” and have to pay $13 a month for the equipment, which will then be reclaimed by Comcast when the customer moves or cancels their service.

In both these cases, what Anderson is describing as “free” is free in name only. Instead of describing a new business model, in the case of the Comcast DVRs the word ‘free’ seems to be used to mask the true cost of the product. While Anderson’s article is interesting, and he provides a helpful description of the uses of free products in the new economy, I think his failure to be more descriptive in this case somewhat tarnishes the overall argument of the piece.