Tuesday, July 14, 2009

Conference Attendance: a Double Shot of Reality

Attention all college presidents, AD's, and wussy coaches - I've got a deal for you.

Us fans will acknowledge the fact that in order to be prudent stewards of your college or university, sometimes you have to do what's fiscally sound (aka, schedule cupcakes in order to make money, get an easy win, and go to a bowl game). Sometimes.

But if we accept that, you have to acknowledge the fact that a lot of us fans think that despite the money, this is a cowardly and bullshit path that leads to uncompetitive games that none of us want to watch. (You can't ignore the fact that when playing non-BCS opponents your attendance drops an average of -14%, and when playing I-AA teams it drops an average of -22%.)


Non-conf games (2000-2008) Home Games Total Attendance Average Attendance Drop from Average % Drop
ACC vs BCS 65 03,623,533 55,747
ACC vs non-BCS 88 04,512,727 51,281 -04,465 -08.0%
ACC vs I-AA 53 02,673,821 50,449 -05,297 -09.5%
Big10 vs BCS 62 04,562,836 73,594
Big10 vs non-BCS 130 09,299,744 71,536 -02,057 -02.8%
Big10 vs I-AA 36 02,034,731 56,520 -17,073 -23.2%
Big12 vs BCS 41 02,580,134 62,930
Big12 vs non-BCS 142 08,197,073 57,726 -05,204 -08.3%
Big12 vs I-AA 66 03,184,454 48,249 -14,680 -23.3%
BigEast vs BCS 53 02,614,131 49,323
BigEast vs non-BCS 90 03,580,581 39,784 -09,538 -19.3%
BigEast vs I-AA 41 01,573,839 38,386 -10,936 -22.2%
CUSA vs BCS 61 02,254,231 36,955
CUSA vs non-BCS 53 01,477,181 27,871 -09,083 -24.6%
CUSA vs I-AA 49 01,339,694 27,341 -09,613 -26.0%
MAC vs BCS 38 00930,994 24,500
MAC vs non-BCS 32 00594,874 18,590 -05,910 -24.1%
MAC vs I-AA 82 01,511,589 18,434 -06,065 -24.8%
MtnWest vs BCS 47 01,929,690 41,057
MtnWest vs non-BCS 60 02,183,753 36,396 -04,661 -11.4%
MtnWest vs I-AA 37 01,163,088 31,435 -09,622 -23.4%
Pac10 vs BCS 49 03,056,303 62,374
Pac10 vs non-BCS 98 04,843,257 49,421 -12,952 -20.8%
Pac10 vs I-AA 26 01,169,157 44,968 -17,405 -27.9%
SEC vs BCS 52 04,124,898 79,325
SEC vs non-BCS 169 12,529,826 74,141 -05,183 -06.5%
SEC vs I-AA 57 03,730,076 65,440 -13,885 -17.5%
SunBelt vs BCS 12 00247,774 20,648
SunBelt vs non-BCS 40 00738,810 18,470 -02,177 -10.5%
SunBelt vs I-AA 42 00709,942 16,903 -03,744 -18.1%
WAC vs BCS 36 01,234,032 34,279
WAC vs non-BCS 59 01,670,569 28,315 -05,963 -17.4%
WAC vs I-AA 48 01,222,606 25,471 -08,807 -25.7%

Speaking of attendance, we seem to have found actual proof of a diving line between the haves and the have-nots. It's at about 36,000. (Note to the Mountain West fans - I know you want to be taken more seriously as a conference and be a part of the BCS, but in addition to winning more games against BCS competition, it wouldn't hurt if you'd start buying more tickets. Your 2008 average attendance was your lowest this decade...)

I'm just sayin'.


bevo said...

Cool analysis and graphs. Good work. Take your data to the conclusion though.

Many ADs at BCS conference member schools state that football has to pay the athletics department bills. Your analysis shows that those ADs are leaving money on the table. Or, to use economists' speak, they are not maximizing revenue.

If they believed their argument that football pays the bills, then those ADs would schedule more games against other BCS conference member schools.

Ed Gunther said...

Thanks, bevo. Scheduling cupcakes does make financial sense - but you've gotta add scheduling differences to attendance to see it.

Most BCS conference opponents insist on a home-and-home series, while a lot of non-BCS teams will settle for a one time game, or a one-and-two series (one game at the opponent, two at home). Games against I-AA cupcakes are always a one time game (the only teams that have traveled TO a I-AA opponent in the BCS era were from the SunBelt, and it's only happened 7 times).

So for example, Tennessee can schedule a home-and-home with UCLA, play two games, and have a total attendance (at their one home game) of around 100,000. Or they can schedule a two cupcakes, play two games, and have a total attendance of 160,000. So when it comes to raw attendance per game, scheduling cupcakes is going to get them more tickets sold overall, even though their per game average is going to be 20% lower.

Of course you've gotta add in how much they're paying the cupcake for that one game, usually in the neighborhood of hundreds of thousands of dollars (an amount which is increasing almost exponentially, incidentally). But even with such a payout, they're still gonna make more money.

These numbers are just a very rough guesstimate/example, but...

100,000 x $50 ticket = $5,000,000
160,000 x $50 ticket = $8,000,000 - two $500,000 payouts = $7,000,000

When universities and colleges are in penny-pinching mode, as they are now, the cupcakes' financial incentive can be too big to pass up.

Unknown said...

Excellent analysis as usual.

A couple questions and a possible, though not certain, critique.

First, what was your source on attendance? Is there a database somewhere that tracked all this or did you just take attendance from individual boxscores and compile them?

Second, not only does attendance vary against lower-tier teams, but in many cases, the ticket prices do as well. This is not a hard and fast rule, and I'm sure the biggest programs don't adhere to it, but some of the marginal BCS conference teams will often charge a significantly reduced price on single game tickets when it is not a marquee opponent. I'm not positive how you would factor that in, but it seems like something worth considering if an AD were going to attempt a real NPV analysis.

Ed Gunther said...

Hey there Nathaniel, the attendance figures are from the NCAA's official records of the games, so (in theory) they should be more accurate than totals coming from the schools.

You bring up a good point about ticket prices being radically different too - in reality, schools probably lose more money (per game) playing cupcakes than my above example estimates. I would love to try to incorporate ticket prices, but just finding an accurate measure of them (especially the retroactive ones going back nearly ten years) is both nearly impossible and a bigger undertaking than I want to attempt.

If a school was really serious about focusing solely on revenue and maximizing profit, they could work up such a study, but it's tricky, mainly because even though higher-caliber make the number of fans/amount of $ go up, the chance of winning the game goes down. It seems that in the SEC it's not really that big of a deal for most teams - their fans are gonna show up whether they're 11-2 or 7-6. But in the Pac10, more losses would probably mean fewer fans, which means less money too. So it's a tricky balance that schools have to work with.

xlpharmacy coupons said...

I really enjoy this article and of course learn more about the topic

price per head services said...

It makes a lot of sense that they can schedule that kind of games to make money in an easy manner.