April 12, 2006
Lies, Damned Lies
Neighborly BaseballOne of the underlying themes behind the aggressive approach that Kenny Williams and the White Sox took this off-season is that the Sox are competing not only with their AL Central rivals, but also against the Cubs for the hearts, minds, and pocketbooks of Chicago's baseball-loving citizens. Indeed, there is a plausible case that the Cubs are on the brink of a perfect storm of circumstances that could shift the pendulum back to the White Sox--diminishing performance coupled with increasing ticket prices and rising expectations. The Sox foresaw a window of opportunity, and gave it their best shot to stick the Cubs in their jugular.
Is it working? Well, the White Sox sold 109,979 seats during their opening series, and the Cubs 120,917. (It should be noted that the Cubs were playing the Cardinals, and could probably sell a Cardinals game out if it was Abbott Laboratories Anthrax Giveaway Day and the game was played in an igloo overlooking Lake Michigan in mid-January). So the early returns seem to be that the White Sox are going to draw plenty of fans this season--2.85 million is my guess--but won't necessarily have "stolen" those fans from the Cubs. It's just that a lot of Chicagoans are going to be watching a lot of baseball.
I decided to examine this question more systematically, looking at attendance and related data going back to the mid-80s. The question I was trying to answer was this: suppose that the White Sox produce an extra $1.00 in attendance revenues. What effect, if any, does this have on the attendance revenues of the Cubs? Predictably, this involved a series of regression equations with only slightly fewer ingredients than the Hot and Sour Soup at the Chinese takeout place down the street. In particular, I looked at attendance revenues relative to both recent and long-term on-field performance, ticket prices, stadium changes and renovations, and the attendance revenues of the other team in a two-team market. Revenue figures were adjusted relative to major league average revenues in the given season, in order to account for macroscopic economic fluctuations from year to year.
The figures that I'm going to present look like this:
Jets -> Sharks $0.25 [$0.04, $0.45] Sharks -> Jets -$0.15 [-$0.36, $0.10]What this notation means is that for every additional dollar in attendance revenues that the Jets bring in, the Sharks can expect an increase in attendance revenues of 25 cents. Conversely, for every additional dollar in attendance revenues that the Sharks bring in, the Jets can expect a decrease of 15 cents. The 95 percent confidence intervals from the regression equations are presented in brackets.
Let's save the Chicago case for later and look at New York first.
Yankees -> Mets $0.37 [$0.10, $0.63] Mets -> Yankees $0.15 [$0.03, $0.27]You'll notice that both values are positive--in fact, both values stay positive within the 95 percent confidence interval. In other words, when the Yankees are taking in more at the box office, that is a good thing, all else being equal, for the Mets. And vice versa: the Yankees would prefer to see the Mets do well.
Perhaps this is an aberration--we all know that New Yorkers are a strange lot. Here is the same analysis for Southern California:
Dodgers -> Angels $0.42 [$0.23, $0.60] Angels -> Dodgers $0.29 [$0.08, $0.50]Same thing going on; in fact the effect is a bit stronger. Then again, the only people stranger than New Yorkers are Angelinos. Let's see if Chicagoans get it right and understand what a true rivalry is:
Cubs -> White Sox $0.17 [-$0.03, $0.37] White Sox -> Cubs -$0.11 [-$0.23, $0.01]This dynamic is a bit more convoluted. Although the White Sox benefit when the Cubs are drawing well, the Cubs do slightly worse when the White Sox are drawing well. Both figures are significant at the 90 percent confidence level (although not at the 95 percent confidence level). We'll have to think about this one.
Finally, let's check out the Bay Area teams.
Giants -> A's $0.03 [-$0.15, $0.21] A's -> Giants $0.06 [-$0.02, $0.13]Not much to look at here. Multiple choice--this is because (a) there's a cleaner geographical distinction between the West and East Bays; (b) whatever Mapquest says, you couldn't get from the Coliseum to SBC Park in 20 minutes if you were flying a Concorde; (c) Bay Area fans are characteristically non-committal and indifferent.
Ignoring the Bay Area case, it appears that there are two significant dynamics going on here. Firstly, rather than a competition effect, there is something of a substitution effect. That is, the commodity is not Dodger Baseball or Angel Baseball, but Baseball, Period. If it becomes more difficult to go to a Dodgers game--because the stadium is sold out, because you can't find a season ticket package suitably close to the field, because you're simply not willing to wait until the Dodgers get back from their East Coast swing--the logical thing to do is not to give up and go to the movies, but to go to the Angels game.
There may also be something that I'd call a Buzz Effect. If more people are talking baseball because the Angels are playing well, that increases the amount of attention you're paying to the game, and your desire to 'consume' baseball in general. Even things like advertising campaigns could have a spillover or overflow effect. This is not my area of expertise, but I would bet that, if GM ran a saturation campaign featuring its mid-priced family sedans, it would tend to increase the sales of GM sedans, but would also slightly increase the sales of similarly equipped Ford family sedans. If America West launched a discount on flights to Las Vegas from a particular market, it would tend to increase Southwest's load factors on flights to Las Vegas (especially after the America West flights had sold out).
There's a second, smaller effect at work too, which is that the secondary team in a two-team market (the Mets, Angels, and White Sox, respectively) tends to benefit more from the primary team doing well than the other way around. This would tend to support the overflow hypothesis. Yankees games sell out, so if you want to see some baseball, you start going to Mets games. But the Mets usually have good seats still available, so there is some excess capacity to fill at Shea Stadium before the Yankees get a spillover. (Also keep in mind that degrees of "selling out" are relative. The Yankees might not be completely sold out, but there are no affordable, non-nosebleed seats available, and you'd rather sit in field boxes at Shea than the right field bleachers at The Stadium.)
However, there is still the strange White Sox-Cubs dynamic to explain. It may simply be that Chicago fans really are more loyal to their clubs than fans in New York or Los Angeles. Among my friends, I can certainly identify a few Cubs fans who would tell me that it would be a cold day in hell before they'd spend three hours sitting at the Cell, and vice versa. The White Sox benefit a bit when the Cubs are doing well because of the spillover effect from crowded Wrigley Field, but the competition effect tends to prevail in the White Sox-on-Cubs case.
I suspect that this also has to do with the fact that the White Sox have always relied very heavily on walk-up sales. As anyone who has lived in Chicago for some time can tell you, The Cell is completely bizarre. You can go on an evening when you're expecting 15,000 fans and get 35,000 if a warm front passes through the city at about 4 PM, or just the opposite if there's a Bears game on the tube or thunderstorms are expected. Walk-up sales are a bit more spontaneous and discretionary than tickets purchased in advance, and may be less subject to the overflow effect.
In any event, general managers are wasting their time if they're worrying about competing with the second team in their market. Rather, they should be thinking about ways to cooperate, such as joint marketing initiatives or even cross-town partial season ticket plans. All of this, by the way, should be great news for baseball--the game itself is stronger than any one franchise. Baseball needn't be a zero-sum game; in fact, the opposite is true.