NHL Attendance and Performance Revisited

Yesterday afternoon, I happened across some new data which broke out teams’ post-lockout (2005-2006 season through 2009-2010) attendance and records, which I hadn’t seen before. It’s not perfect, but it does add a layer of detail which hadn’t been available before and allows at least a little bit of trend analysis.

I had a fairly lengthy internal debate about the complexity of the filters. The visualizations are far too crowded to allow for nice, neat display within the 625 pixels worth of width here, particularly with all thirty teams toggled on (and, honestly, even full-screen at 1600×1200). I’m not completely satisfied with the solution on which I settled, but it seemed that most of the target audience would have interest in a few teams, perhaps a division or geographical region, and possibly a whole conference at a time so there didn’t seem to be a happy way to make certain they all worked without making lots of worksheets and linking them all- which seemed complicated enough that no one who wasn’t a data geek would really do so (and a tip of the virtual chapeau to those of you out there who are data and visualization geeks, particularly those of you I had the honour of meeting last week at TCC 2010).

My objectives were two-fold: first, to try to put myself in the position of an interested potential ownership group possibly looking to relocate a team hoping to target one with a flagging support base; and secondly, to see if the article in which I found the data had reached good conclusions based on the data rather than anecdote- there didn’t seem to be much logic behind some of the rankings.

The second of the two was by far the easier. I think many of the conclusions Tom Schreier drew in the original article were at best flawed. I mean, c’mon now, ranking the Air Canada Centre in Toronto as having the best home-ice advantage in the National Hockey League when despite playing to sellout+ crowds for every home date over the last five years the Maple Leafs have averaged 5.4 points under the league median of points earned in home games. No, if there’s an arena which seems to give a boost to its team just from the data, it’d have to be Detroit’s Joe Louis Arena, where the host Red Wings have performed 10.7 points better than the median averaged over the last five seasons, and have done so in front of a shade over 98.1% filled seats.

Although the Bank Atlantic Center, home of the Florida Panthers, brought up the rear of Schreier’s list, that seemed tenuous as well. True, the Panthers’ drew a league low average percentage (at 80.61%), but the St. Louis Blues’ home at Scottrade Center seems like a far less daunting place to play wearing road white jerseys, as the Blues rank lowest in the league, posting only 41.6 points at home on average, while playing in front of average crowds about nine percent smaller than the league median, making them almost as much an outlier as their divisional opponent in Detroit is on the other end of the spectrum.

The other objective gets a little more difficult. With the ownership question mark in Phoenix, the Coyotes will come up in discussions, but excluding last year, the Coyotes haven’t drawn that poorly, although they still place solidly in the sub-median quadrant. It’s too early to tell if last season’s performance will make hockey fans in the Arizona desert flock to Jobing Arena, but if the team continues to perform well, there may be other more tempting targets.

The relative weakness of the Southeastern Conference may for move rumours surrounding any team there, and it’s possible to make a case for most of them. Realistically, though, the Washington Capitals aren’t going anywhere, and are a solid draw at home. The former Hartford Whalers seem to have settled into a new home in North Carolina’s Research Triangle, and have on average both outdrawn (in percentage) and outpaced their division-mates to the north. Tampa Bay may have underachieved in the standings, but draw near the median- and statistically better than any other team in their division, and with fresh faces in the lineup, a new general manager and ownership group, the Lightning aren’t likely to bolt soon either (sorry, couldn’t resist the pun!).

This leaves Atlanta and the Panthers as potential candidates from the Southeast. Florida has underperformed- but not as badly as Atlanta, trailing the median by 2.1 points as opposed to the Thrashers’ 5.3, but their attendance figures seem to indicate that they play in a building which is about 3-4k seats bigger than they can fill in the Miami-Ft. Lauderdale area as they average just over 80.6%- or one in five empty seats. Atlanta did well with the Flames before they left for Calgary thirty years ago, but with a team which has been unable to retain star players, attendance has fallen sharply, dropping about 2600 in the last four years. Either might be available at the right price, but the Panthers’ better play may make them more viable in a new market, albeit likely the more expensive of the two.

The Columbus Blue Jackets and Nashville Predators may come to mind for some as potential candidates. The Preds might actually be a steal were they to become available- they are a solid enough squad to perform better than the median, averaging 5.1 points better at home, while drawing roughly the same five year average as the Capitals or Blackhawks (although, to be fair, Chicago’s attendance was comparatively abysmal in the first two years of the sample, sufficient to skew the whole to about the third-year’s value; as was the Caps’ draw in the pre-Ovechkin era). Columbus has been a reasonably mediocre team drawing reasonably mediocre crowds which have slumped almost ten percent over our sample. As was the case with the Florida Panthers, this may be sufficient to make them sufficiently more attractive for relocation.

Two other teams stand out as potential candidates- although potentially shocking, and probably off of almost everyone’s radar. First, the New York Islanders. Well removed from the glory days of the early ’80s and their consecutive Cup wins, the Isles are the weakest drawing team in a market currently supporting three franchises. Their ownership group has made some dubious contract moves, and they’ve played to a building fifteen to twenty-five percent empty over the last three years. Owner Charles Wang has made noises about a new arena, but nothing has come of it, and with a flock of up-and-coming young talent, the Islanders might be attractive to someone with deep enough pockets to eat the bad contracts and uproot the team.

Finally, and perhaps most surprisingly, the team which has had the steepest drop in attendance in our sample, dropping over four thousand fans per home date- dropping from sellouts every night to under 77.5% capacity, the Colorado Avalanche. The Avs are another team somewhat removed from their glory days, but with a crop of emergent players, there is potential to turn the corner.

I’m not even going to broach the question of where these teams should go if new ownership should arise, or whether the NHL would approve moves aside from a few blanket statements: “Out of the Sun Belt” – if the climate is not capable of forming ice outside in the winter at that latitude, hockey may be a tough sell; “Canada”- the game was born in the Home and Native Land, and will find a way to be supported there; and “Seattle” because, honestly, I’d rather not have to travel in order to see pro games. Besides, the first Cup won outside of Canada was lifted here, so the NHL owes the Emerald City something. 😉

Agree? Disagree? Think you can make a better case for another team after playing with the data? I’d love to see your spin on it.

Share and Enjoy: These icons link to social bookmarking sites where readers can share and discover new web pages.
  • TwitThis
  • Facebook
  • del.icio.us
  • Digg
  • StumbleUpon
  • Reddit
  • BlogMemes
  • email

One Response to “NHL Attendance and Performance Revisited”

  • Vanya:

    Yes, the first graph would be a little easier to deal with if it were a line graph rather than a scatterplot. Anyone want to volunteer to figure out a way I can coax team colours for thirty different teams into a 20-colour palette, particularly with so many teams using red, navy, white and/or black in their schema?

    If there had been more than five years worth, I probably would have needed to come up with another way or simply abandon the idea of matching a team to their colours.

    Heck, I couldn’t manage to do it for even a single division. You want to try:
    Atlantic Division:
    New Jersey: Red, White, Black
    New York Islanders: Blue, White, Orange
    New York Rangers: Blue, Red, White
    Philadelphia: Orange, White, Black
    Pittsburgh: Black, White, Gold

    Yeah, only five teams worth of difficulty, and exactly one unique colour.

    Want to try another one?
    Northeast Division:
    Boston: Black, Gold, White
    Buffalo: Blue, Gold, White
    Montréal: Red, White, Blue
    Ottawa: Red, White, Black
    Toronto: Blue, White

    No unique colours there. Start to see the problem? 😉

Leave a Reply

Inside Crashing the Net
Support the Site
July 2015
« Dec    
Get Adobe Flash playerPlugin by wpburn.com wordpress themes
Easy AdSense by Unreal