FanPost

Attendance Statistics Round 2: Now with More Data!

Last week I wrote up a statistical analysis of D.C. United attendance over the past two seasons. My findings were interesting, but were ultimately inconclusive due to a small sample size. Now, however, I have tripled the number of observations, extending my dataset out to the 2008 season for a total of six seasons, or 98 total games. I will continue to collect both future and past data, but unfortunately as I go farther back in time some of the data becomes more difficult to find.

I added a few variables to my analysis, most based on some excellent suggestions in the comments. In addition to Nationals games, I now also include Wizards and Capitals games. I changed my measure of days between home games; before, I included the between-season gap, but now I ignore those values, to look only at the time elapsed between games in an individual season. I added a variable for nationally televised games, which I defined to be on any NBC, Fox, or ESPN network (I think there was one game a few years ago on ABC, but I excluded Galavision). I consolidated my time and date variables to afternoon and night (before and after 6:00) and weekday and weekend (with the weekend being Friday and Saturday, since at least in my opinion Sunday is considered a work/school day). Finally, I added indicator variables for each season, to see if there are any yearly trends in attendance numbers (this could account for economic factors or demographic changes over time).

The results were somewhat different from what I found out in my first attempt. Playoff games were also insignificant, likely due to the fact that only two of our last 98 games were playoff games. Temperature, days elapsed between games, openers, games for other D.C. teams, college nights, national television, unattractive opponents, and afternoon games were also now insignificant to attendance numbers. Interestingly, the variables for every year but 2008 were statistically different from 2013.

2008 saw nearly 4,800 more fans than games in any other year, independent of all other factors. Form remained significant, with 182 more fans for each additional point of form (as a reminder, I calculated form as the total points earned in our last five games). This is about 70 fewer fans per point than my first attempt. Though temperature was not significant, rain did have an effect on attendance this time, with a drop of about 2,000 fans if there was rain before or during the game. Games against attractive opponents (New York and LA) increased attendance by 2,900 fans. While this isn’t unreasonable, it may be skewed by a game in June 2008 that drew 36,000 fans. Games on federal holiday weekends drew nearly 2,700 fewer fans. There was a fairly even distribution of Memorial Day, July 4th, and Labor Day games so this does seem like more than a seasonal trend. Finally, to no one’s surprise, playing a game on a weekday (Sunday through Thursday) brings 2,200 fewer fans.

Variable

1st Analysis Effect

2nd Analysis Effect

Game Time

- (Insignificant)

-

Weekday Games

3800 fewer on Wednesdays

2200 fewer on weekdays

Form

250 more per point

180 more per point

Days Elapsed

47 more per day (including between season gaps)

-

Temperature

Optimal at 75 degrees

-

Precipitation

-

2000 fewer

Playoffs

5000 more

-

Nats, Caps, Wizards Games

-

-

College Night

-

-

Holiday Weekends

-

2700 less

Playing NY and LA

-

2900 more

"Unattractive" opponents

-

-

Games in 2008

N/A

4800 more

Games in other years

N/A

-

Though the results changed somewhat, none of my new results were incredibly surprising. Form is still important, though slightly less so. Weekday games are still toxic to attendance numbers. Interestingly, playing New York or LA or playing on a holiday weekend have relatively large effects on attendance. Also, I found it slightly odd that 2008 had such larger crowds but none of the other years were significant. Perhaps this shows the end of D.C. United’s "golden age," when we were one of the top clubs in MLS every year.

Though my dataset is still not perfect, the results for the most part make sense. A better team means more fans. Rain means less fans. Playing Beckham or Henry draws larger crowds than playing the likes of Wondolowski or Beckerman or whomever Toronto happens to have at the time (have to get all of my TFC jokes out now before they inevitably run rampant this season). Like before, feel free to look at my data and offer me suggestions, I will keep up with this and post updates when time allows.

Data

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