This is our old blog. It hasn't been active since 2011. Please see the link above for our current blog or click the logo above to see all of the great data and content on this site.

Fan Loyalty, 2000-2010

Posted by Neil Paine on January 19, 2010

Here's a question that's sure to spark debate: Which NBA teams have the most loyal fans?

Well, to answer that, first you have to define how you want to measure "loyalty" in a fan base. I think we'd all agree that it's easy to come out and support the team when they're winning, but what about making the trip to the arena and dropping x amount of cash to see a demonstrably lousy team lose again and again? Irrational or not, a lot of sports fans would call that kind of behavior the mark of a true die-hard, the kind of rooter against whom we all measure our own love for our favorite teams. Obviously, this is an oversimplification -- you can fervently support a team without necessarily going to games, especially if you're not living in the same area as that team -- but the assumption remains that a team's support "at home" will track with their fandom "abroad", that the two are strongly correlated. Maybe this isn't true, but for our purposes here it's a lot easier to track attendance than it is to go out and poll people in Nevada about their support for the Lakers. So we'll go with arena attendance as a proxy for team support.

With that assumption out of the way, we now have to think about how to set up our model. Not every arena can hold the same number of fans, so it would be unfair to compare teams on the basis of raw per-home-game attendance; instead, we should make our dependent variable the percentage of capacity filled in the team's home venue. To that end, I pulled each arena's capacity off of Wikipedia... Here are the teams that filled their arenas at the highest rate since 2000:

Team Att%
Los Angeles Lakers 0.997
Dallas Mavericks 0.990
San Antonio Spurs 0.982
New York Knickerbockers 0.969
Chicago Bulls 0.963
Utah Jazz 0.962
Phoenix Suns 0.955
Sacramento Kings 0.943
Toronto Raptors 0.935
Portland Trail Blazers 0.921
Boston Celtics 0.909
Detroit Pistons 0.909
Miami Heat 0.902
Orlando Magic 0.901
Indiana Pacers 0.871
Houston Rockets 0.868
Los Angeles Clippers 0.867
Milwaukee Bucks 0.866
Washington Wizards 0.860
Denver Nuggets 0.858
Cleveland Cavaliers 0.856
Golden State Warriors 0.840
New Orleans Hornets 0.836
Philadelphia 76ers 0.814
Minnesota Timberwolves 0.790
Atlanta Hawks 0.780
Charlotte Bobcats 0.766
Memphis Grizzlies 0.762
New Jersey Nets 0.761

(Note: All data through Sunday, 1/17. Seattle/Oklahoma City is excluded from this entire study, including the Hornets' tenure in OKC.)

Now we have to think about which factors contribute to these attendance figures. Winning percentage is obviously huge, since as we said, it's easier to support a team when they're winning. Here are the teams with the best winning percentages since 2000:

Team WPct
San Antonio Spurs 0.698
Dallas Mavericks 0.667
Los Angeles Lakers 0.654
Phoenix Suns 0.598
Detroit Pistons 0.595
Utah Jazz 0.562
Sacramento Kings 0.548
Houston Rockets 0.537
Boston Celtics 0.532
Indiana Pacers 0.528
Portland Trail Blazers 0.522
New Orleans Hornets 0.517
Philadelphia 76ers 0.511
Orlando Magic 0.511
Miami Heat 0.510
Denver Nuggets 0.500
Cleveland Cavaliers 0.493
Minnesota Timberwolves 0.487
New Jersey Nets 0.475
Milwaukee Bucks 0.457
Toronto Raptors 0.456
New York Knickerbockers 0.423
Memphis Grizzlies 0.408
Washington Wizards 0.408
Chicago Bulls 0.391
Golden State Warriors 0.385
Atlanta Hawks 0.382
Los Angeles Clippers 0.376
Charlotte Bobcats 0.364

Also, metro population plays a factor -- the more people there are in an area, the more basketball fans there will be, which will contribute to the team's ability to sell tickets (independent of how well the team performs or even how many people the arena can hold). For this figure, I again consulted Wikipedia, and linearly adjusted each population up or down for every year based on its growth rate between 2000 and 2008 (Toronto was found here). Within each season, I calculated the Z-score for the metropolitan area, and calculated the city's average for the 2000-2010 period:

Team Pop
New Jersey Nets 2.84
New York Knickerbockers 2.84
Los Angeles Clippers 1.57
Los Angeles Lakers 1.57
Chicago Bulls 0.88
Dallas Mavericks 0.14
Philadelphia 76ers 0.14
Houston Rockets 0.03
Miami Heat 0.03
Washington Wizards -0.01
Toronto Raptors -0.03
Atlanta Hawks -0.05
Boston Celtics -0.15
Detroit Pistons -0.15
Golden State Warriors -0.20
Phoenix Suns -0.27
Minnesota Timberwolves -0.43
Denver Nuggets -0.59
Cleveland Cavaliers -0.64
Portland Trail Blazers -0.64
Sacramento Kings -0.67
San Antonio Spurs -0.68
Orlando Magic -0.69
Charlotte Bobcats -0.74
Indiana Pacers -0.74
Milwaukee Bucks -0.76
Memphis Grizzlies -0.82
New Orleans Hornets -0.83
Utah Jazz -0.86

I could have adjusted for the skew in the distribution, but for our purposes the numbers above will work. I also added variables for per-capita income and team payroll, but they were found to be statistically insignificant. Anyway, the end result was that you could establish an "expected" attendance % by using the equation:

xAtt% = exp(4.82*wpct + 0.326*pop_z)/(exp(4.82*wpct + 0.326*pop_z) + 1)

This yields the following expected arena attendance %'s:

Team WPct Pop xAtt%
Los Angeles Lakers 0.654 1.57 0.975
Dallas Mavericks 0.667 0.14 0.963
New Jersey Nets 0.475 2.84 0.961
San Antonio Spurs 0.698 -0.68 0.958
New York Knickerbockers 0.423 2.84 0.951
Detroit Pistons 0.595 -0.15 0.944
Phoenix Suns 0.598 -0.27 0.942
Houston Rockets 0.537 0.03 0.931
Boston Celtics 0.532 -0.15 0.925
Philadelphia 76ers 0.511 0.14 0.925
Miami Heat 0.510 0.03 0.922
Utah Jazz 0.562 -0.86 0.919
Sacramento Kings 0.548 -0.67 0.918
Los Angeles Clippers 0.376 1.57 0.911
Portland Trail Blazers 0.522 -0.64 0.909
Indiana Pacers 0.528 -0.74 0.909
Orlando Magic 0.511 -0.69 0.903
Denver Nuggets 0.500 -0.59 0.902
New Orleans Hornets 0.517 -0.83 0.902
Minnesota Timberwolves 0.487 -0.43 0.901
Toronto Raptors 0.456 -0.03 0.899
Chicago Bulls 0.391 0.88 0.897
Cleveland Cavaliers 0.493 -0.64 0.897
Washington Wizards 0.408 -0.01 0.877
Milwaukee Bucks 0.457 -0.76 0.876
Atlanta Hawks 0.382 -0.05 0.861
Golden State Warriors 0.385 -0.20 0.857
Memphis Grizzlies 0.408 -0.82 0.845
Charlotte Bobcats 0.364 -0.74 0.820

Fan Loyalty, then, would be the difference between a team's actual attendance % and their expected rate -- the rate at which fans supported the team above and beyond what can be explained by team success (@#$% bandwagoners!) and the size of the metropolitan area:

Team WPct Pop xAtt% Att% Loyalty
Chicago Bulls 0.391 0.88 0.897 0.963 0.066
Utah Jazz 0.562 -0.86 0.919 0.962 0.043
Toronto Raptors 0.456 -0.03 0.899 0.935 0.036
Dallas Mavericks 0.667 0.14 0.963 0.990 0.027
Sacramento Kings 0.548 -0.67 0.918 0.943 0.025
San Antonio Spurs 0.698 -0.68 0.958 0.982 0.024
Los Angeles Lakers 0.654 1.57 0.975 0.997 0.022
New York Knickerbockers 0.423 2.84 0.951 0.969 0.018
Phoenix Suns 0.598 -0.27 0.942 0.955 0.013
Portland Trail Blazers 0.522 -0.64 0.909 0.921 0.012
Orlando Magic 0.511 -0.69 0.903 0.901 -0.003
Milwaukee Bucks 0.457 -0.76 0.876 0.866 -0.010
Boston Celtics 0.532 -0.15 0.925 0.909 -0.016
Washington Wizards 0.408 -0.01 0.877 0.860 -0.017
Golden State Warriors 0.385 -0.20 0.857 0.840 -0.017
Miami Heat 0.510 0.03 0.922 0.902 -0.020
Detroit Pistons 0.595 -0.15 0.944 0.909 -0.035
Indiana Pacers 0.528 -0.74 0.909 0.871 -0.038
Cleveland Cavaliers 0.493 -0.64 0.897 0.856 -0.041
Los Angeles Clippers 0.376 1.57 0.911 0.867 -0.043
Denver Nuggets 0.500 -0.59 0.902 0.858 -0.044
Charlotte Bobcats 0.364 -0.74 0.820 0.766 -0.053
Houston Rockets 0.537 0.03 0.931 0.868 -0.062
New Orleans Hornets 0.517 -0.83 0.902 0.836 -0.065
Atlanta Hawks 0.382 -0.05 0.861 0.780 -0.081
Memphis Grizzlies 0.408 -0.82 0.845 0.762 -0.083
Philadelphia 76ers 0.511 0.14 0.925 0.814 -0.111
Minnesota Timberwolves 0.487 -0.43 0.901 0.790 -0.111
New Jersey Nets 0.475 2.84 0.961 0.761 -0.201

Despite one of the lowest winning percentages of the decade, the Chicago Bulls have had one of the highest rates of attendance over the same span. The United Center held 21,711 + standing room for the majority of the 2000s, and Chicago averaged 20,858 fans a night, even selling out as they went 32-132 in 2000 and '01 (one of the worst 2-year stretches by any team in NBA history). MJ wasn't walking through that door, but Bulls fans still came out and cheered, proving they didn't just abandon the team when Jerry Krause dismantled it in 1998.

Jazz fans have been blessed with a much better team than the Bulls, but with a fraction of the Windy City's population, they still deserve credit for making EnergySolutions Arena one of the loudest venues in the league. They easily boasted the NBA's best ratio of total attendance to metro population over the past 10+ years, which is impressive even if you take into account how smoothly Jerry Sloan and Utah's brain trust were able to transition from the Stockton/Malone era to the Williams/Boozer/Kirilenko one.

At the bottom of the list, New Jersey might get a raw deal for being assigned the full NYC metro population, but it doesn't excuse the fact that the Nets sold at 72% capacity the two seasons they went to the Finals in the early-to-mid 2000s. Meanwhile, Minnesota's fans showed up at only an average clip during the KG era and haven't been anywhere near as supportive since -- as has been the case in Philly, where Allen Iverson's departure signaled a significant drop-off in fan support. While understandable, that's not going to earn you any credit in a ranking of loyal fan bases (by comparison, the Raptors won less than New Jersey, Minnesota, and Philadelphia over the same span, and still managed to sell Air Canada Centre seats at the 9th-best rate in the league). Because when it comes to fan loyalty, I think we can all agree that what you do when the team isn't winning defines you far more than what you do during good times.

17 Responses to “Fan Loyalty, 2000-2010”

  1. Nate D. Says:

    I think another factor that wasn't mentioned was how many (or few) other pro/major college sports teams were within any given market. Granted teams like the Spurs, Kings, and Jazz who had winning teams through much if not all of the 2000's would be expected to draw well in any market, however I think that NBA teams in Portland, Orlando and Milwaukee (each in the top half of the league in "fan loyalty") might have benefitted attendance-wise by being "the only show in town". Anybody agree or disagree? Just my two cents. Interesting topic and I enjoyed the article.

  2. Jason J Says:

    I have no idea if this is done by organizations (at all), and it may just be that Chicago has a lot of serious basketball fans, but did people get roped into long term season ticket deals while the BULLS were playing and then get forced to keep them while the bulls were playing?

  3. Neil Paine Says:

    Good point, Nate. I considered throwing in variables for attendance of NFL/NHL teams in the same metro area, but it would have involved acquiring some data I didn't have on hand at the time. Colleges would also be interesting to include -- perhaps I could just add a dummy variable for each sport (1 if they have a team, 0 if not).

    Jason, interesting idea, although Chicago also led in attendance % in 2007 and was 2nd in 2008, only dropping out of the top 10 twice all decade (15th in 2002, 12th in 2003). I don't know if people bought season tickets 3-4 years in advance in 1998, but even after that they've continued to sell at a rate that isn't commensurate with their W-L record. Granted, it's a huge market, but still.

  4. Gabe Says:

    Two things:

    1) I believe some arena sizes have changed over time (adding new seats, restructuring, etc). The Wikipedia page seems to only list the current attendance.

    2) For teams that share a city (eg, NY, LA), does it make sense to divide the total city population by two, in order to take that into account? Maybe not necessarily divide in half, but make some sort of adjustment? Perhaps add a binary variable for it?

  5. Quenton Says:

    Brilliant article.

    I'd love if you'd run the numbers for Seattle (pipe dream) up until The Thing That Shall Not Be Named, as my gut tells me the city's loyalty was much higher than certain entities would have people believe.

  6. mackie chang Says:

    how about loyalty be defined by more intangible things. Sure statistics make the argument black and white, but can it really tell the whole story? I'm pretty sure you have to be very die-hard and loyal if you are a clippers fan, in LA or abroad, especially with them constantly being hated on (the onion news story on all the clippers dieing in a staples center collapse comes to mind), as well as the horrible luck that's come upon them time and time again (everyone knows about the bad picks, and brand leaving).

    Just putting that out there.

  7. Jason J Says:

    Neil - by 2007 & 2008 there had been some recovery in Chicago. 1999-2001 was a basketball abortion (no offense to Jud, Toni, and Randy Brown) where the three most popular players and the coaching staff from the team's glory days all hit the pavement. I just wondered if perhaps the apparent popularity of the team (and a lot of Chicagoans were pretty pissed at management for the dynasty failing to defend the title) had something to do with ticket holders being locked into old deals at that stage - again if that's even something that happens, which I don't know.

  8. Neil Paine Says:

    Key word: "some" recovery -- you still don't expect a team that goes 41-41 to sell 99% of its seats. I agree, though, that a lot of their performance here was driven by sales at the beginning of the decade when they were beyond awful, so you may be on to something when it comes to being locked into longer season-ticket deals. I'm going to take another look at this next week, so I'll see what I can dig up.

  9. Gawad kalinga Philippines Says:

    Loyalty, being the difference of actual attendance and expected attendance, is quite flawed for me. Let's take LAL for example. The lakers, for the last decade has been winning. And so, as expected, their actual and expected attendance are high, thus making the loyalty % low. You said that the most loyal fans go to arenas even though their team is losing. You can't say that lakers fans aren't that loyal if they have a little opportunity to prove it because the lakers were and are still winning. Can you see my point? This case may be applied to the spurs as well.

    Maybe, your formula is more applicable to the teams that have low winning percentage for the last decade.

    And I pity the nets.

  10. Walter Says:

    I think the study has a lot of good information but I think the output of which teams were the most "loyal" is somewhat skewed.

    For instance, the Lakers expected attendance percentage was 97.5% meaning that even if fans were literally fighting to the death every night outside the arena to get in and had a perfect 100% attendance record they would have finished no higher than 5th on the list as they couldn't possibly exceed the expected attendance by more than 2.5% (and to their credit they did exceed their expected by 2.2%). When a team has no mathematical chance at being higher than 5th then the results aren't necessarily accurate.

    I think it would be interesting to include ticket price into the study as well (or possibly a ratio of ticket price to per capita income). Any team would probably be close to 100% attendance if the tickets prices were $1.00 and they all would suffer drastically if the average ticket price was $1,000 regardless of the product on the floor.

  11. Walter Says:

    Neil, this is a little bit of a tangent but I know that their are many regression (or MLE) models that incorporate home court advantage as an independent variable. However, whenever I have seen the results the home court advantage is assumed to be constant across all teams. Would it be possible for you to update your BBR Rankings system to include an separate homecourt advantage variable for each team (the model would have 60 variables instead of 31). I think it would be interesting to see which teams have the best and worst homecourt advantage.

    I would assume teams like Utah would have the best.

  12. marc Says:

    I think this tends to discount Portland's loyalty. Smack in the middle of the period in question there was a fan revolt due to the Jail Blazers backlash. Fans were still fans, but they were pissed off fans that were literally boycotting the team. This sort of analysis glosses over those sorts of things that are unique to certain franchises.

    In terms of Seattle's loyalty, as someone who lived up there I think loyalist views of the overall fanbase's loyalty are generally exaggerated or greatly exaggerated. The fans that were there were quite loyal, but huge swaths of the Seattle metro area simply didn't care. Seattle would also be more affected than most towns on this list from competition--no NHL, but the Seahawks, Mariners and Huskies were all significant factors in diluting Sonics interest which led to a very wishy-washy level of interest depending upon on-court success. Even when success was found, though, support still wasn't as widespread as for the other sports--when you can literally buy walk-up tickets for a Western Conference Finals game in 1996, the height of their popularity, and not have to go to scalpers as I did, it says that the fan base, while passionate, didn't cover large chunks of the population there.

  13. Stay Chisel Says:

    I've been a Bulls season ticket holder for a while and have not seen any long-term ticket plan options. As far as I know, tickets are only purchased one season at a time.

  14. Neil Paine Says:

    Well Marc, isn't boycotting the team (because of the Jail Blazers or whatever) still being disloyal? Isn't the whole point of fan loyalty to support a team unconditionally, theoretically even if the roster was stocked with felons and lost every game?

  15. Jason J Says:

    thanks, chisel

  16. Johnny Twisto Says:

    There's no perfect way to account for this, but just for anyone who doesn't know, it's hard to equate the Knicks and Nets, despite being in the same MSA. The Knicks play right in the middle of the most densely populated county in the country, and right on top of 6 subway lines, 2 regional rail lines, and the national rail line. The Nets play in the middle of a parking lot in the middle of a swamp.

    And as others have alluded to, it's possible the Nets fans are very loyal, but there just aren't many of them. Still, our definitions of "loyalty" could be semantics, and this is still an interesting study.

  17. Kevin Pelton Says:

    Marc, my vague recollection is you must be referring to the Sonics' decision to reserve something on the order of 500-1,000 tickets for each game in 1995-96 only to be sold on gameday starting two hours before tipoff. The team sold out every game that postseason, so to suggest that tickets were easily available due to a lack of interest is erroneous.

    The Sonics also sold out every home game during the 1996-97, 1997-98 and 1998-99 seasons. By my math, that streak included 114 regular-season games and 23 playoff games.