FanPost

A Statistical Look at D.C. United Attendance

Ned Dishman

Unlike the stadiums of other MLS teams, our beloved RFK has for the past few decades boasted more raccoons than sellouts. Since I’m only a college student, I have only ever seen RFK as the home of D.C. United and briefly the Nationals, neither of whom routinely filled the stadium to the levels that I am told the Redskins did pre-FedEx Field. My memory of a filled RFK consists of only one game: the US’s epic defeat of Germany’s B-team last summer. Outside of that, I’ve only ever seen the half-filled lower bowl customary of D.C. United games.

Since I’m studying economics and statistics in college, I figured I would try my hand at some statistical analysis of D.C. United attendance over the past two seasons. Using Stephen Whiting’s excellent articles as a starting point, I collected data on a number of different factors that could affect attendance numbers and ran some basic regressions to see what I could find out. I’ll try not to bore anyone with the statistical details, but for anyone with no knowledge of statistics, essentially what I am doing is breaking down total attendance to determine the individual effects of different variables, such as weather or team form. If a variable is "statistically insignificant," its effect on attendance is no stronger than pure chance. "Significant" variables, on the other hand, have distinct, measurable effects on attendance. Feel free to ask me for more details or clarification in the comments. I should also point out that I can count the number of statistics classes I have taken on one hand, so I welcome any suggestions, questions, or constructive criticism from anyone who knows more about data analysis than I do.

My data consists of every home MLS game over the past two seasons (2012 and 2013), which gave me 36 observations (34 regular season games and 2 playoff games). Though it’s a relatively small sample, it does offer a wide range of days, times, weather, and on-field performance. The following are the variables I use in my analysis.

Game Time: indicators of when the game began (1, 3, 4, 5, 6, 7, or 8 o’clock)

Game Day: indicator of day of game (every day except Tuesday)

Form: number of points earned in previous five regular season or playoff games (i.e. not Open Cup, carries over between seasons)

Time Elapsed: number of days since last DCU home game

Temperature: temperature at game time according to Weather Underground

Rain: whether or not it rained during the game

Nationals Game: whether or not the Nationals played the same day

Opener/Playoffs: whether the game was the home opener or a playoff game

College Night: whether there was a College Night promotion

Holiday: whether or not the game took place on a holiday weekend

Opponent’s Attractiveness: I split this into three groups using a general rule of "How easy would it be for me to convince my non-MLS-fan friends to come to a game against this team?"

Attractive: NY, LA

Neutral: HOU, PHI, CHI, MON, SJ, POR, SKC, SEA, RSL

Unattractive: CLB, CHV, VAN, TOR, COL, FCD, NE

One piece of data I am missing but would have liked to have is ticket prices; I was unable to easily locate any useful data. As a result, I am assuming that ticket prices are fairly consistent over the course of a season, but this could be wrong, as I haven’t gone to many games recently since I go to school in Missouri.

I ran ordinary least squares regressions using EViews to analyze the data. After playing around with it and removing irrelevant variables, I found most of the variables to be statistically insignificant or irrelevant. The time of the game had no significant effect on attendance, but this may have been due to the fact that there were 26 games at 7:00 and no more than three at any other time. Similarly, only Wednesday games were statistically different from Saturday games, drawing about 3800 fewer fans. Surprisingly, opponent attractiveness was insignificant, as were holiday weekends, Nationals games, college night promotions, and precipitation. I left out the variable for season opener, as its effect was captured in the variable for days elapsed between home games.

So what does have a significant effect on attendance? One result was far from shocking: playoff games draw nearly 5,000 more fans than regular season games. Form was also very significant, with an additional 250 fans for each additional point gained in the last five games. When comparing between the last two seasons, you can see how large of an effect this has on attendance. The highest attendance in 2012, the LEWIS NEAL game against Columbus, was on 13 points of form. Based on my regression results, this game would see 3,250 more fans than any of the two home games this season where we lost the previous five. My regression also proves that absence makes the heart grow fonder: the number of days elapsed between home games was also significant, with each additional day adding 47 extra fans. This is clearly evident in the home openers (140 and 111 days elapsed in 2012 and ’13), but also in the LEWIS NEAL game (27 days elapsed) and our June 2012 game against Montreal (35 days elapsed, second-highest regular season attendance in 2012). Finally, temperature significantly affected attendance. Unlike the other variables, temperature acts "quadratically" on attendance; it acts negatively if it is too high or too low. According to my regression, there is an "optimal" temperature of about 75 degrees, which seems reasonable.

Variable

Significant?

Effect on Attendance

Game Time

No

-

Wednesday Game

Yes

3800 less than Saturday games

All Other Days (excluding Saturday)

No

-

Form

Yes

250 more per point

Days Elapsed

Yes

47 more per day

Temperature

Yes

Optimal temperature of 75 degrees

Playoffs

Yes

5000 more

Precipitation

No

-

Nats, College Night, Holiday

No

-

Opponent Attractiveness

No

-

My main takeaway from this analysis is that on-the-field performance is the key factor. Form and playoffs were the only significant variables about which D.C. United has any say. Every other factor either lies in the hands of the MLS (days between home games, Wednesday vs. Saturday games) or some higher power (temperature). Unless we build a dome on Buzzard Point or bribe MLS to give us more well-spaced Saturday games, the only way to boost attendance is to win more games. Differences in form can add or subtract up to 4,000 fans, which is substantial when considering that our average attendance is just under 14,000.

The main flaw in my analysis is the small sample size, especially for factors like College Night or Tuesday games. I’ll continue to collect data through future seasons to make my dataset more robust, but I think for the meantime the conclusions make a lot of sense. Though this may be a stretch, the data speaks both positively and negatively of us fans. On the bright side, we go to the games rain or shine, afternoon or night, and regardless of whom we’re playing. On the other hand, though, there’s a fair contingent of fair-weather fandom, and we don’t like to drag ourselves out to RFK on weekdays. Excluding Kansas City, Seattle, and Portland, though, I would think that these generalizations would be true for most MLS teams’ fans. Overall, a better on-the-field product seems to be the key to higher attendance. Though we can all debate whether our offseason moves will turn out to be worthwhile or not, I think the vast majority would agree that our team is moving in the right direction, which hopefully means higher attendance numbers next season.

Data: https://docs.google.com/spreadsheet/ccc?key=0AsBYmGWEnV8pdE5jajFxUDVENHl1YV9iVlh3SEY3dkE&usp=sharing

X
Log In Sign Up

forgot?
Log In Sign Up

Please choose a new SB Nation username and password

As part of the new SB Nation launch, prior users will need to choose a permanent username, along with a new password.

Your username will be used to login to SB Nation going forward.

I already have a Vox Media account!

Verify Vox Media account

Please login to your Vox Media account. This account will be linked to your previously existing Eater account.

Please choose a new SB Nation username and password

As part of the new SB Nation launch, prior MT authors will need to choose a new username and password.

Your username will be used to login to SB Nation going forward.

Forgot password?

We'll email you a reset link.

If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.

Forgot password?

Try another email?

Almost done,

By becoming a registered user, you are also agreeing to our Terms and confirming that you have read our Privacy Policy.

Join Black And Red United

You must be a member of Black And Red United to participate.

We have our own Community Guidelines at Black And Red United. You should read them.

Join Black And Red United

You must be a member of Black And Red United to participate.

We have our own Community Guidelines at Black And Red United. You should read them.

Spinner.vc97ec6e

Authenticating

Great!

Choose an available username to complete sign up.

In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.

tracking_pixel_9353_tracker