Last week, I examined a few different ways to anticipate the attendance for the Columbus Crew’s second home game. Competing models included past changes from games one to two, and a weather-centric model. The first indicated a crowd of around 13,000, while the weather model predicted a lower figure of 11,500.
The real figure turned out to be somewhere in the middle: 12,045.
This is something of a disappointment, as there was a bit of buzz around this game – at least among the soccer cognoscenti. The Crew were off to a 3-0 start. The Nordecke sold out midway through the week. The visitors were Toronto FC, with unprecedented star power (Michael Bradley! Jermain Defoe! Julio Cesar!) and by some measures one of the team’s most significant rivalry. Some predicted that the attendance could actually increase from the home opener.
(In the back of my head, I’m wondering whether the Nordecke’s sellout didn’t cannibalize other sections of the crowd, shifting ticket sales from one area of the stadium to another. Not that that would be a bad thing – the atmosphere from that corner was very strong, and drew glowing praise from the announcing team on Canada’s TSN. More on this if I can work out how to test this theory.)
Comparison With Other Teams
In order to better understand this decrease, I pulled the numbers from other MLS teams tonight. Of the 19 teams in the league, 16 have played at least two home games – the exceptions are Los Angeles, New England, and Toronto. Here are the attendance figures for those 16 teams, plotted in absolute terms.
A few key points as I look at this chart are:
- No team increased their attendance from the first to the second game.
- The only teams that stayed level were those who sold out their opener: San Jose (10,525), Portland (20,814), and Vancouver (21,000).
Decreases in Absolute Terms
To focus only on the various decreases, here is the same data calibrated from a uniform starting point:
When the numbers are plotted this way, the Crew’s decrease stands out more starkly. No other team in the league saw its crowd shrink as significantly – although New York came close (dropping from 20,542 to 15,225). It should be noted, however, that a number of other clubs saw fairly large decreases as well. Six clubs saw their attendance drop by more than 3,000. Columbus is not alone in that respect.
Decreases in Percentage Terms
Here is the same data, now expressed in terms of percentage decline, rather than absolute figures.
When looked at in percentage terms, DC sinks to the last position. Their drop from 13,840 to 9,445 is slightly more significant – 47% – than was the Crew’s 45% decline. New York fares better here due to their greater starting point, but may still be concerning with a 35% drop. There also appears to be somewhat more separation from this group to the rest of the league, although seven teams still saw a drop of more than 20%.
The Long View
While there are aspects of this data that may cause concern, a few points need to be kept in mind. A glance at the attendance figures for the Crew over the last few seasons shows very similar patterns. While this drop is significant among MLS, it is consistent with past trends in Columbus. In fact, both of the first two games outdrew the same games last year. Over a slightly longer view, these numbers are at the high side of the range over the last several years:
This figure for game 2 is still the highest its been since 2011, and the two-game average (14,781) is the second-highest since 2006 #crew96
— Matt Bernhardt (@bernhardtsoccer) April 6, 2014
Attendance will almost certainly increase as the season progresses. It will be interesting to see what the crowd is for the team’s next home game, in two weeks against traditional rival DC United. It is still too early for a reliable weather forecast, but as game day approaches I plan to re-visit a few models to generally anticipate what the crowd size might be.