July 26, 2014

The Diff: Market saturation and Indians attendance

In last week’s edition of The Diff, I shared a number of statistical notes about the streaking Cleveland Cavaliers. This week, I’m covering a topic I’ve been trying to get to for a while, but is perfect for right now: Cleveland Indians attendance.

The Diff

Cleveland Indians attendance is a topical news item this week. Firing right off the Nick Swisher arrival and the surprise signing of Michael Bourn, the Indians unveiled their new promotional schedule and then sold out of Opening Day tickets in six minutes on Monday. Excitement about the Tribe is as palpable this February as it’s been in the last decade — save maybe for right after ’07′s surprising playoff run. And that brings people to the age-old question: Will it show up in the ticket booths? Fortunately, I’m here to help — as I teased in Monday’s podcast.

Back in my college days (i.e. last year) at the University of Dayton, I majored in applied mathematical economics and wrote my undergraduate honors thesis with the economics/finance department on Minor League Baseball attendance. You can read my thesis at this link for yourself if you’d like. The title is “Minor League Baseball Attendance: Tradition, Success and the Recession.” I covered existing literature on the topic of sports attendance, interviewed 9 different minor league organizations for case study arguments and then featured a regression analysis of attendance data from 1998-2011. It’s a doozy. In fact, it’s about 70 pages long.

Cutting to the gist of my addition to the field of study: A new variable I introduced in this thesis was the concept of market saturation. Many of my interviews indirectly mentioned this topic. And, to a certain extent, it also relates to the concept of “fan identification” that other papers discussed. This ID variable was used to relate specific areas to their affinity to a certain team or sport — think of St. Louis and the Cardinals, or Pittsburgh and the Steelers. However, since I didn’t have a team of sociologists or psychologists working with me on my project, I just stuck to market saturation. And it relates to the Cleveland Indians in 2013 and beyond.

Here’s what I mean by market saturation exactly as it related to the regression analysis: Two separate variables, one for the number of major professional sports teams (NFL, MLB, NBA, NHL) in a MiLB team’s Combined Statistical Area1, and the other for the number of total MiLB teams in said CSA. In 2009, the last year of accurate economic data by the time of my analysis, the average MiLB team had 1.9 major professional sports teams in its CSA and 1.7 MiLB teams (including itself) in its CSA. Obviously, there were many outliers for areas such as New York City, Los Angeles and conversely for small towns such as Butte, Montana, or El Paso, Texas.

So my thesis had two different regression models: A full model (Model 1) that included all 12 variables I considered, and then an adjusted model (Model 2) that removed economic/demographic numbers, only focusing on dummy variables for the various MiLB leagues and market saturation. Here are the results for my market saturation variables in these two models:

The data below is presented as = coefficient (t-statistics), with * indicating significance at the 0.01 level.

Variable Model 1 Model 2
CSA Major 248.2 (6.47)* 407.4 (17.20)*
CSA MiLB -432.6 (-9.20)* -311.5 (-7.44)*

 

For the non-math majors, what does this mean? It shows that in both instances, both variables were significant at a very high statistical level. By variable:

CSA Major (again, the number of major pro sports teams in a given MiLB team’s CSA) was positive and significant. So for every CSA Major, it positively impacted the expectation of a team’s average attendance by 250-400.

Then, CSA MiLB (again, the number of total MiLB teams in the CSA), was negative and significant. So for every CSA MiLB, it negatively impacted the expectation of a team’s average attendance by 300-400.

So this is interesting to consider. There obviously are other factors at play — most notably, the fact that there are strict territorial restrictions of relocation and expansion of new professional baseball teams. Also, it is my belief that average annual attendance in MiLB has expanded most significantly because of location-optimization in the last 10-15 years. Teams are moving to larger, more desirable locations, and that obviously is related to these market saturation variables.

As a final summary of my thesis results, my two models only had overall R-squared values of 0.644 and 0.600. So even for all of these variables and samples spanning up to 13 years of attendance data, my models only theoretically explained up to 64.4% of the variation in MiLB attendance2. Which means I’m still missing a lot. But it’s still fun to consider for what it means in the practical

Now, for what this all means for the Cleveland Indians. But first, in order to explain this, I need to go back to the minors one more time and the Akron Aeros. As many of you may know from my previous writings, I worked for the Aeros in their media relations department from 2008-10. In my last year, I served as the interim media relations director between my sophomore and junior years at UD. This relationship was listed on page 2 of my thesis as a disclaimer, just to be safe, as the Aeros were one of my 9 case studies. I had the opportunity to interview several members of their front office in 2011 for the purpose of my research and the team is undoubtedly one of the more intriguing cases of any MiLB team in the last 15 years.

Here’s why: After moving from Canton’s Thurman Munson Stadium to beautiful and well-reputed Canal Park in downtown Akron in 1997, the Aeros were known as one of the premier entities in the minors. The stadium, right downtown, was designed by HOK, the same firm responsible for Camden Yards in Baltimore and Jacobs Field in Cleveland. It was a perfect combination of downtown re-development and a new minor league team3. And attendance skyrocketed quickly — especially with the 5-year ticket plans that the team promoted heavily.

Unfortunately, under the previous ownership of the Agganis family4, no significant upgrades were made over time to the product at Canal Park. So as the window of 5-year renewals came up in 2002 and 2003, season-ticket holders started falling by the boatloads. And the same thing happened in 2007 and 2008. The result? From 2004-2010, the team had six consecutive seasons of decreasing attendance, falling 46.1% in total. Here’s a year-by-year look at the team’s attendance figures:

Team Games Total Avg Year
Akron 68 473,232 6,959 1997
Akron 70 521,122 7,445 1998
Akron 69 522,459 7,572 1999
Akron 66 481,060 7,289 2000
Akron 69 485,582 7,037 2001
Akron 67 400,187 5,973 2002
Akron 67 445,603 6,651 2003
Akron 68 478,611 7,038 2004
Akron 66 455,056 6,895 2005
Akron 65 412,995 6,354 2006
Akron 64 355,376 5,553 2007
Akron 67 342,816 5,117 2008
Akron 68 316,836 4,659 2009
Akron 69 261,563 3,791 2010
Akron 68 266,265 3,916 2011
Akron 68 256,473 3,772 2012

 

From the absolute class of the Eastern League and one of the best in the minors, the Aeros have fallen to the ranks one of the worst attendance figures in Double-A baseball. They’ve practically reached a plateau-esque residence in their attendance tumble, settling around the 3,800 average attendance mark over the past three seasons.

For the Indians, however, everyone in Northeast Ohio practically knows their recent story of success and decline. Buoyed by the on-field success of the team, the Indians didn’t initially suffer through a phenomenon many call the “honeymoon effect.” This effect is that a new stadium only provides a temporary boost in attendance for any given team, and that said boost starts to die off after only a few years. So given that Jacobs Field opened in 1994, most models would have expected a decline in average attendance to begin within the next five years or so5.

Instead, the team kept selling out game after game for those first several years — as the team’s success potentially artificially delayed the honeymoon boost, the topic of one paper in particular that I read. The famous run of 455 sell-outs lasted from June 12, 1995, to April 4, 2001. But then, along with the end of the streak and the run of good Indians teams, things came crashing down in a hurry. Attendance dropped 18.6% from 2001 to 2002, and then another 33.9% from 2002 to 2003. Overall, the average attendance has only been 22,996 in the last 10 seasons, compared to 40,228 in the first 9 years at Jacobs Field. Here is a look at the Indians year-by-year numbers at Gateway:

Team Games Total Avg Year
Cleveland 51 1,995,174 39,121 1994
Cleveland 72 2,842,745 39,483 1995
Cleveland 80 3,318,174 41,477 1996
Cleveland 81 3,404,750 42,034 1997
Cleveland 81 3,467,299 42,806 1998
Cleveland 81 3,468,456 42,820 1999
Cleveland 81 3,456,278 42,670 2000
Cleveland 80 3,175,523 39,694 2001
Cleveland 81 2,616,940 32,308 2002
Cleveland 81 1,730,002 21,358 2003
Cleveland 81 1,814,401 22,400 2004
Cleveland 81 2,013,763 24,861 2005
Cleveland 81 1,997,995 24,667 2006
Cleveland 80 2,275,912 28,449 2007
Cleveland 81 2,169,760 26,787 2008
Cleveland 81 1,766,242 21,805 2009
Cleveland 81 1,391,644 17,181 2010
Cleveland 81 1,840,835 22,726 2011
Cleveland 81 1,603,596 19,797 2012

 

While the success of the Indians is certainly a factor to be discussed when comparing the 1994-2002 era to the 2003-2012 era, there also is something notable that you might have picked up on from everything else in this article thus far: market saturation. What else has changed in the Northeast Ohio framework since 1994? More baseball options. Before I get to any more stats, let me share an actual story.

Akron native Jim Pfander was hired to be the new CEO of the Aeros back in 2011. He then worked down for the Charlotte Stone Crabs in 2012, but returned back home again to be the Aeros GM after that season once Ken Babby became the new owner of the Akron team. Pfander and I chatted for about an hour on Friday, August 12, 2011, to discuss my thesis research. Obviously, I wanted to get his feedback on my concept of market saturation as it relates to minor league attendance. The following comment of his relates to market saturation and the Aeros’ hopeful goal to reach their earlier attendance figures:

“You know, I think back to when the team first moved here, it was the convergence of a lot of unique situations where, you know, the Browns moved, this is a football town first and foremost, but the Browns were out of the picture, they were in Baltimore, they weren’t even in existence. The Indians were selling out every game, so there was no other [option]. If you wanted to see Indians baseball, you came to Akron. There’s 3.5 million people that live between Akron, Canton, Cleveland right now, but even 10 years ago, there was more, and that’s just a ton of people to draw from. The Lake County Captains didn’t exist, the Mahoning Valley Scrappers didn’t exist, the Avon Crushers didn’t exist, even though they’re not affiliated with the Indians, and the Columbus team was affiliated with the Yankees. So you know, you had no other alternatives to see Indians baseball and the prospects, and you couldn’t get a ticket to Jacobs Field so you came here.”

Pfander went on to say that day that he still thought that Akron could carve out a niche in the area as a family-friendly destination for affordable entertainment. That even though the market was saturated with sports options (especially baseball ones), it probably wasn’t over-saturated as it relates to the Aeros still having some success long-term — maybe not as much as those first few years, but at least a better trend than the most recent few.

So this is what market saturation could mean for the Akron Aeros, simply a Double-A team that in an ideal world, was selling $5 tickets and competing with move theaters and putt-putt for the entertainment option of Summit County families. It’s not exactly the same market as Cleveland Indians attendance. But there is something there potentially in Pfander’s comments as it relates to the overall ceiling/potential for attendance success for the Tribe.

Thus, maybe, because of these four new baseball teams that have joined the fold since 1994 — the Aeros in ’97, the Mahoning Valley Scrappers in ’99, the Lake County Captains in ’03 and the independent Lake Erie Crushers in ’09– the Indians won’t ever shine like they once did. Obviously, there are a number of other factors at play, such as the Browns departure, as Pfander mentioned. And, of course, these things are incredibly cyclical: It’s very difficult to get out of a constant downswing in bad PR and bad attendance, just as there were a whole slew of factors that led to the improbably 455-game streak.

Now I’m going to share the breakdown of professional baseball attendance among these four teams:6

Year AKR + LC + MV CLE Total
1997 473,232 3,404,750 3,877,982
1998 521,122 3,467,299 3,988,421
1999 725,532 3,468,456 4,193,988
2000 687,347 3,456,278 4,143,625
2001 667,199 3,175,523 3,842,722
2002 560,294 2,616,940 3,177,234
2003 1,025,007 1,730,002 2,755,009
2004 1,045,539 1,814,401 2,859,940
2005 1,003,143 2,013,763 3,016,906
2006 889,233 1,997,995 2,887,228
2007 815,329 2,275,912 3,091,241
2008 782,752 2,169,760 2,952,512
2009 705,486 1,766,242 2,471,728
2010 664,054 1,391,644 2,055,698
2011 613,210 1,840,835 2,454,045
2012 614,543 1,603,596 2,218,139

 

So the theory isn’t quite complete nor accurate. Total NE Ohio attendance in 1999 and 2000, the first and second years of the Scrappers, exceeded 4 million. It has then been under 2.5 million for each of the last 4 seasons. So while market saturation could be a factor, it obviously isn’t the complete story.

But to a certain extent, I’ll maintain that there is some impact. And it all relates to Pfander’s comments I shared above. In a confluence of factors, the Indians were a hot ticket in the mid-’90s. There was no other way to watch Indians baseball. Then the Aeros came about, the Scrappers and Captains joined the fold, and the Crushers also are another avenue for baseball in NE Ohio.

Are they all competing for the same fans? Not necessarily. And the overall Greater Cleveland-Akron-Youngstown area is likely large enough to support a few teams (maybe not necessarily all of them, or all of them that well), but it probably means the Indians won’t ever total 3 million again — like they did from 1996-2001.

As I shared on Monday’s podcast, to a certain extent, the Indians are like the Indiana Pacers in basketball. Although the Pacers’ attendance struggles are also because of the small arena in Indianapolis, the Indians share the same mid-size city mentality that they won’t ever be able to rank in the top 10 in their league’s attendance ever again.

Is a possible 2013 improvement to nearly 2 million enough for fans to think it’s not depressing at Progressive Field anymore? I’m not certain. That’s likely the goal for the Tribe this season with all of the new excitement and energy. Do I think it will happen? Probably not, as I’d probably peg my odds for now closer to 1.85 million, which would still be the higest total since 2008. I’ll certainly be watching and keeping an eye on how this all pertains to market saturation and my college thesis research.

Note #1: Another topic I could explore further in a future article is month-by-month attendance data. For tables on what I mean, take a look at this about the Aeros, and I’ve got a similar table as well for the Indians. Would you readers be interested in further analysis in this topic? Just let me know in the comments.

Note #2: I’m hoping to write a follow-up article on this topic as a guest post for Ben’s Biz Blog in the coming week. I’ve been in contact with Ben for the past few weeks. Keep an eye out for that, and thanks to Ben for his feedback thus far on my writing.

___________________________________________________

Footnotes:

  1. If you’re looking for more info on what designates a CSA, check out Wikipedia. I settled on this area definition because it was the largest census block. For the several Micro- or Metropolitan statistical areas that didn’t reside in a larger CSA, I simply used whatever their base statistical area was. []
  2. Note again: This thesis covered Minor League Baseball attendance, not the majors. In the majors, undoubtedly, attendance figures are much more related to success of the team. There are varying opinions I read on the impact of winning in MiLB attendance, but likely, these figures can be much more impacted by external factors to the team’s success. []
  3. Construction of new stadiums is a sticky topic that was quite polarizing in many of the articles I read. Financially speaking, does it likely ever make sense for a city to invest $31 million — as the city of Akron did — on a sports facility? Probably not, in the actual accounting sense. But there certainly can be some points whereas the city and a new team can help to spark some economic development of a slighted area. That was exactly what Mayor Don Plusquellic envisioned for the revival of downtown Akron and he nailed it down perfectly. []
  4. Another notable factor of minor league attendance that I hope to explore sometime in the future: The effect of local ownership. The Agganis family was not based in Akron and only visited occasionally in the summer time. New Aeros owner Ken Babby has officially moved to the city. Think his 24/7/365 engagement with the area can be a boost for awareness of the team? Undoubtedly. []
  5. Or, in the case of the New York Mets and Citi Field, in year two. In fact, the Mets had one of the more notable honeymoon effects in history with their new stadium as attendance dropped from 39,118 in year one (2009) to 31,602 in year two (2010) — that’s a drop of 19.2%. []
  6. Passing my love out to the Crushers, and my former colleague Mike Link, but I’m not using their data here. My research was in affiliated pro baseball and I’m adding in the Indians’ context here. If you’re interested in the Crushers’ attendance, you can visit this link. []
  • c3j1v62

    Impressive work.

  • mgbode

    Fantastic article here and yes, I’d like to see the Indians month-to-month attendance figures over the years.

    One other important effect on the attendance drop was after the sellout streak ended, the Indians not only had the Browns back to contend with for attendance dollars but the Cavaliers entered the fray as a legit #1-sport-in-town contender with LeBron James.

  • JacobWFNY

    Yet another factor at play here. So many perfect things went into play for that 455-game streak. It’s why 3 million likely never will happen again. But there’s still moderate success and improvement to be had, undoubtedly. Maybe this is the season they’ll creep back up to 2 million.

  • JacobWFNY

    Thank you much. I appreciate the nice comment.

  • cynosurescence

    It would be nice to see how the attendance data track with large-scale economic statistics like GDP growth or purchasing power. By eyeball, it looks like the dips in attendance track pretty closely with the recessions of 2001 and 2008. Correlation is not causation, obviously, but that was the first thing that jumped out at me.

    Maybe you controlled for those factors already and I simply missed that mention, which wouldn’t be surprising as economics is not my primary field.

  • JacobWFNY

    cynosurescence — I did not perform any regression analysis on the Indians figures. Certainly, your explanation also makes sense as well.

    My thesis covered minor league baseball only. In my full model, here are some of the external variables I used:

    CSA Major — explained above
    CSA MiLB — explained above
    CSA Population — obvious
    CSA TPIPC — total personal income, per capita
    Age — of stadium
    Micro — dummy variable for whether the area’s base was a micro- or metro-politan statistical area.

    So yes, obviously, CSA TPIPC was positive and significant. Not as significant as some other variables, but ya never know.

    TPIPC is especially a factor in a rich suburb such as Frisco, Texas, home of the Frisco RoughRiders. They’ve averaged spectacular attendance numbers throughout their history, as the DFW suburb is one of the wealthiest in the country. I also wrote a separate paper for my econometrics class on the Texas League alone — specifically covering some of the economic differences in this Double-A league.

    In a nutshell: You are probably right. MLB attendance likely trended down as a whole during those years (independent of new stadiums), which could certainly have a been a multiplier effect on top of all of the other factors going against the Indians.

  • mgbode

    if not, then I fear for the Tribe. they again have alot of good factors in play. the Browns are still losing, the Cavs will not be in the playoffs, and there is a ton of positive buzz about the Francona hire and the Swisher/Bourn signings. if they can add a third consecutive fast start, then I think the attendance will follow.

    they better get it going now though because the Irving Cavs will not cede the month of May (and hopefully June!) to the Indians for much longer.

  • mgbode

    well, as you even mentioned, not all areas are created equally. so, despite the recessions of those years, many areas were not nearly as affected as the Great Lakes region.

    so many variables at play here. glad it’s you analyzing and using the econometrics to figure out which ones are most important and not me :)

  • JacobWFNY

    All very true. But it’d be incredibly surprising if the Indians do jump well over 2 million this year. Such a jump — without any stadium-related boost — would be largely unprecedented in recent history (or, at least in terms of my knowledge of the minors, as the majors might be slightly different with the bigger effect of winning ball clubs and all that).

    We’ll know some more in April. I’ll share the month-by-month chart you were interested in and show what the April pace would need to be for them to reach that level.

  • mgbode

    Not unprecedented at all at the MLB level (at least taken at the completely unscientific study below):

    2011 Baltimore Orioles – 1,755,461 ( 69 wins)
    2012 Baltimore Orioles – 2,102,240 (93 wins)
    Difference: +346,779 (+19.8%)

    2012 Cleveland Indians – 1,603,596 (68 wins)
    *2013 Cleveland Indians – 2,000,000 (need ~94 wins?)
    Needed Difference: 396,404 (24.7%)

    Sure, it’s a bigger jump, but it doesn’t seem out of the realm.

    Oakland only received a 14% jump from 2011 to 2012, but I would also argue that is largely because they didn’t surge until late in the year. Attendance isn’t just about how you finish, but the expectations and how you get to that final win count (so the general consensus that the Angels/Rangers had the West locked up hurt them until late – I think at least).

  • JacobWFNY

    Hmm. Would very much be interested in exploring this more. Some of the papers I read did focus on major professional sports teams — not the minors. But I certainly had selective attention when it came to just the topic I was looking for.

    I’m fascinated by this topic — obviously. I wonder if I have any more data somewhere in my files on MLB attendance, before I decided to just do the minors. Stay tuned, sir. I think I might.

  • mgbode

    fantastic. I always love to see which of my thoughts have actually have true statistical bearing versus myth.

  • a_priori

    Nitpick: El Paso isn’t a small town, it’s 19th in the nation and part of the larger metropolitan area that includes Ciudad Juarez on the other side of border ;P
    - trivia nerd

  • Mookie

    I’m curious as to what the actual attendance was. What is the announced attendance? Is it the number of tickets sold or what passed through the gates? Also, if the Indians can win early with this schedule they could surpass 2 million. April is huge. The biggest draws (NYY and BOS) come in the first homestand.

  • mgbode

    i think it is usually tickets sold. however, as long as the accounting differences do not change from year to year, then the differences and % changes should be relative

    good point on the big draws in April. that may help.

  • Ezzie Goldish

    Would love to see monthly and also relate to Cavs and Browns attendance, including ticket prices. (If you like doing more work… :) )

    I’d really be curious if the aggregate spending on sports tickets has remained a bit tighter than the attendance of any individual team might imply.

  • http://twitter.com/tribestats Bryan Gattozzi

    Excellent work! Can the decrease be explained away by total entertainment competition? I’m just thinking about the amount of money I spend on cell / cable and other internet sports business today compared to 5-10 years ago. Please keep writing stats heavy long form stuff like this!

  • 02cents

    Jim Pfander was actually fired as GM of the Charlotte Stone Crabs and did not make it through his one and only season there. Within the industry he is largely viewed as an individual who has questionable business acumen.

  • Bobert

    Baseball is not as popular as it used to be, especially with America’s youth. ESPN’s age 12-24 poll http://betmlssoccer.com/wp-content/uploads/2012/03/Soccer.jpg

  • mgbode

    obviously. but, it’s still plenty popular. we have 12 different LL organizations in and surrounding Austin, Texas. And, the YMCA has started adding teams the past few years to it too.

    plenty of kids still like playing baseball and that will lead to future fans (just ask soccer, whoops. don’t ask them :) ).