Nate Silver takes a look at the geography of college football

[nmajdan]

Silver, a sports and political statistics blogger, wrote about the geography of college football and the roll it is playing in the conference realignment debate. Pretty interesting read.

http://thequad.blogs.nytimes.com/2011/09/19/the-geography-of-college-football-fans-and-realignment-chaos/

[sooner KB]

Great article.

[Soonerfan88]

It's a great theory but I think the writer didn't put much thought into his sources.

First, he uses Google searches to determine level of interest in college football. That completely leaves me out because A) I don't use Google and B) I don't search for college football, I already know where I'm going.

Second, he cites the CommonCensus Sports Map and it's 30,000 national responses. Even the makers of the map acknowledge it's not yet reliable. From their website: "Important: this map is highly inaccurate and should be understood only as a demonstration, and not as any kind of reliable data yet. It is based on only a few thousands of votes spread across all teams, across the country."

Third, he estimates that OU & OSU combined have less than 2 million fans. The state of Oklahoma has well over 3.5 million people alone. No, not everyone in Oklahoma is a fan of OU/OSU or even football. But if you combine their alums from just the surrounding states, it has to be over 2 million.

[Jacie]

As interesting as it was as well as the intriguing numbers, the comments section was enlightening as well. Several readers pointed out examples where it appeared the data analysis (Google searches) might have skewed the numbers that would affect the total # of fans. If the methodology is in fact a good way to estimate these numbers, however, and the respective school rankings are relatively accurate, it sheds light on conference realignment that disregards emotion, rivalries and geographic location (which it would seem is now given practically no consideration unlike the days when the conferences were first created).

[OUHOMER]

Third, he estimates that OU & OSU combined have less than 2 million fans. The state of Oklahoma has well over 3.5 million people alone. No, not everyone in Oklahoma is a fan of OU/OSU or even football. But if you combine their alums from just the surrounding states, it has to be over 2 million.

the same could be said for every football program

[BigTip]

http://motorbicycling.com/attachments/f37/32356d1294074621-better-change-new-eng-oil-soon-vedy-interesting.jpg

[OUHOMER]

If these numbers were even close , it make you wonder why anyone would go west

[silverwheels]

Seems pretty inaccurate to me. Anyone who lives in Chicagoland will tell you that Notre Dame is #1 there, not Illinois. And also that no one cares about Northwestern except for NW alums.

[Peach Fuzz]

Interesting that Nebraska nearly has as many fans as its state's population, and OU has less fans, despite OK having twice the population of neb.

[BigTip]

Interesting that Nebraska nearly has as many fans as its state's population, and OU has less fans, despite OK having twice the population of neb.

But the state of Oklahoma has three FBS programs to split the fans up. Nebraska only has Nebraska. No Nebraska State (thank heavens!) Or would that be redundant anyway? lol

[the-rover]

When was the last time you did a google search for "college football"? Not sure if I ever have.

I realize this might not be the only way a study like this might be done, but .....

[Ground_Attack]

He claims that Auburn has more fans than 'Bama? In what universe?

[Ground_Attack]

When was the last time you did a google search for "college football"? Not sure if I ever have.

I realize this might not be the only way a study like this might be done, but .....

he must only be counting t-shirt fans. I think I'll set up a script that googles "college football" over and over. In a few weeks we'll have 8 billion fans.

[LASooner]

Yeah.. I thought that was a flawed method as well.

I'm an Oklahoma fan, I want to find out about Oklahoma Sooners... I shall google 'College Football'

That totally happens daily.

[Jacie]

To be fair, the algorithm to estimate the # fans/team was more than just a count of Google searches but for brevity in my post, I mentioned that because it was a point of contention from some of the comments to the article. Google searches was only part of the algorithm and was to used to estimate the # college football fans/region. To break the data down into # fans/team the author used results of an online survey which had 28,000 responses and adjusted that data based on teams' annual revenue. The author stated that the survey results were skewed, even some specific examples such as Arkansas State rating high in the poll but low in revenue and average attendance. As I said before, some comments pointed out even more examples where the poster thought the numbers were off, such as Georgia Tech being more popular in Atlanta than U. of Georgia. So yes, nitpick specific cases but step back to look at the team rankings and it makes sense.

West Coast football teams really don't have as many fans as teams in the Midwest. So despite having large television markets, the Pac-12 looks to add teams from the Big XII because it will increase the # fans of teams in that conference by 70%. That sounds pretty believable to me.

The Big 10 and SEC have not had to go looking for teams to add but could wait for those that were highly rated to become available because the average # fans/team are at the top of the list.

Notre Dame has always insisted that it's fanbase is national, not centered in one state or even region and this data supports that (ND is the #2 team in the NYC market, for example). This is given as a reason the Irish have resisted joining a conference lest it alienate part of the fanbase.

This study is the first I have ever seen of any kind of estimate of # fans/team and it was based on methodology, admittedly flawed (a trait common to any survey), which is why an algorithm is used to adjust the data to get a more accurate result.

[OklahomaTuba]

So Miami has a bigger fan base than LSU, Nebraska and Oklahoma? Interesting....

[EatLeadCommie]

It's an interesting approach to analyzing team popularity, but so much of it doesn't pass the eyeball test. Clemson at #10? Really? Does anybody believe that? And Georgia Tech at #2 in the ACC? They only have about 14k undergrads, half of whom are nerds who don't even know what football is.

Also, nobody gives a good damn about Rutgers football. Nobody.

[TMcGee86]

Yeah, the problem with using Google searches at all is that it's automatically biased towards whatever team is currently newsworthy.

I think this explains why Auburn would be ahead of Bama, and teams like Clemson get a bump because no ones ever heard of them and they assume they are in the SEC and they have to google to find out they are actually in the ACC.

Doesn't prove much, but he's right, it's basically impossible to do a valid analysis without spending millions.

[nmajdan]

And Georgia Tech at #2 in the ACC? They only have about 14k undergrads, half of whom are nerds who don't even know what football is.

Your methodology is clearly superior.

[KantoSooner]

I like the approach, but the details need work. I know that when I have lived in 'non-football' areas of the country, it has been difficult to communicate exactly what something like the RRS was like.
I think it highly likely, for example, that not a single Rutgers fan approaches the intensity of the lowest decile of OU, or Bama, or UT fans. If you want dedication to your team, there is a band of the country from around Savannah to Lubbock that extends as far North as Lincoln....and that's it. There simply are no college football fans outside of this area.

It is because of the style of parentling and the genetics of the people in those areas. Raised to be whining, puling, theiving types and having had their lesser impulses encouraged, they can not relate to the soul-cleansing purity of that quasi-religious gladiatorial combat spectacle known as college football. But it's not really their fault. They were, mostly, born that way. I feel sad for them more than condemnatory.

[htownsooner7]

This is nonsense, there is probably a 30 to 40% error rate on this analysis. Some of the comparisons are just silly and google search of 'college football' is a very poor data point. I'm interested in politics but am not googling "president" or "republican party". There also is a crazy tilt towards certain big cities.

I would bet you could do a better job (although not as broad in scope) by measuring TV viewership in certain areas. This Saturday would be a perfect opportunity to measure/evaluate the ratings of the Aggie/Okie St gm on ABC v the Bama/Arky game on CBS vs the Fla St/Clem game on Espn. All at 2:30, all on main stream channels, all with major conference implications (clearly I don't believe the first matchup will determine the conference champ).