I'll probably need to take time to post this in more detail, but after reading the conference realignment discussions - I wanted to know - which teams attract the most viewers and which the least.
I couldn't find a resource that keeps track of the TV ratings for every team - and that would create a lot of variance issues anyway in trying to compare a team on one network at a featured time vs. a team on another network at say noon.
What I did find was a link that had ratings performance in bowl games going back several years - Bowl Performance Ratings
From that - I was able to first trasnscribe into Excel the TV ratings for each bowl going back 8 years - but as you'll see I ended up, for now, only settling on the last 4 years of data for this fanpost (too much work in one sitting).
Because I didn't want the relative popularity of college football to have an affect - I standardized the TV ratings for each bowl using each season's total viewership. In other words if the TV viewership has risen 10% from 2008 to 2009 - I didn't want that to affect my numbers (actually the viewership held pretty steady from '06-'07 to '08-'09 - but it did tick up 6% in '09-'10 as it turned out). If that doesn't make sense ask me about it and I'll try to explain it better.
From that I could then determine what the average viewership was for each bowl over those 4 years. So, for example the BCS National Championship game averaged a 16.98 rating; while the Gator Bowl averaged 3.62.
Armed with average data, I could now sum up the individual data for each team and compare how say Alabama did in their games with each bowl's 4 year average.
For example - Bama was in the BCS NC game last year, the Sugar Bowl the year before and the Independence bowl in 07-08 and 06-07. The standardized ratings in those games were 16.11, 7.85, 2.01 and 2.98. The average standardized ratings for those bowls were 16.98, 8.14, 2.05, and 2.05 respectively. So Bama drew a total of 28.95 standardized viewers vs. 29.22 average standardized viewers or 99.05% of average. So they underperformed (only slightly).
The way to read the number is that Bama drew 1% fewer viewers than an average teams that appeared in those same bowl games.
I repeated the exercise for every Div I team that appeared in bowls each of the last 4 years.
The results are after the jump.
I thought going in that FSU was a good draw - but it wasn't until I sorted the data that I saw that FSU came out on top.
I also ran data for Miami who only had 3 bowl appearances in the last 4 years. They came in at 108.99 which would have tied them for 4th with Ok State.
And also for ND who only made 2 bowl appearances (they declined a bowl that would have given them 3 appearances last year). ND came in at 118.84 which would have been good enough to just barely take 1st from FSU.
The numbers are certainly affected by BOTH teams in a bowl game. Pair an attractive team with an unattractive team and the results would surely be less than if two attractive teams were paired. Over time this will work itself out presumably. I may run the numbers back 8 years, but the individual team data entry is tedious.
Anyway, I think its interesting that FSU draws 18% more viewers than an average team. I was surprised to see GT so high and Oregon so low (I rechecked the numbers on Oregon - I'm not sure what would explain their ratings performance).
I was expecting UF to do better - but honestly - since they were in the BCS NC game twice in this study - the game is already pretty hyped - its hard to drive the viewership alot higher. Then last year they drew Cinn in the Sugar which was a downer after Bama beat them in the SECCG (ask Urban how much of a downer that was - ha).
I think this also points out how valuable FSU would be as a national draw for TV viewers to any conference thinking of expanding (SEC?). Only ND outdrew FSU relative to average performance and then only slightly. The SEC could do alot worse than to pull in FSU, GT, Miami and Clemson.