A Quick Example of the Limitations of Analytics: Sports Analytics Series Part 3.1

In Part 3 we started to talk about the complementary role of human decision makers and models.  Before we get to the next topic – Decision Biases – I wanted to take a moment to present an example that helps illustrate the points being made in the last entry.

I’m going to make the point using an intentionally nontraditional example.  Part of the reason I’m using this example is that I think it’s worthwhile to think about what might be “questionable” in terms of the analysis.  So rather than look at some well-studied relationships in contexts like NFL quarterbacks or NBA players, I’m going to develop a model of Fullback performance in Major League Soccer.

To keep this simple, I’m going to try and figure out the relationship between a player’s Plus-Minus statistic and a few key performance variables.  I’m not going to provide a critique of Plus-Minus but I encourage everyone to think about the value of such a statistic in soccer in general and for the Fullback position in particular.  This is an important exercise for thinking about combining statistical analysis and human insight.  What is the right bottom line metric for a defensive player in a team sport?

The specific analysis is a simple regression model that quantifies the relationship between Plus-Minus and the following performance measures:

  • % of Defensive Ground Duels Won
  • % of Defensive Aerial Duels Won
  • Tackling Success Rate (%)
  • % of Successful Passes in the Opponents ½

This is obviously a very limited set of statistics.  One thing to think about is that if I am creating this statistical model with even multiple years of data, I probably don’t have very many observations.  This is a common problem.  In any league there are usually about 30 teams and maybe 5 players at any position.  We can potentially capture massive amounts of data but maybe we only have 150 observations a year.  Note that in the case of MLS fullbacks we have less than that.  This is important because it means that in sports contexts we need to have parsimonious models.  We can’t throw all of our data into the models because we don’t have enough observations.

The table below lists the regression output.  Basically, the output is saying that % Successful passes in the opponent’s half is the only statistic that is significantly and positively correlated with a Fullback’s Plus-Minus statistic.

Parameter Estimates
Variable DF Parameter
t Value Pr > |t|
Intercept 1 -1.66764 0.41380 -4.03 <.0001
% Defensive Ground Duels Won 1 -0.00433 0.00314 -1.38 0.1692
% Def Aerial Duels Won 1 -0.00088542 0.00182 -0.49 0.6263
 Tackling Success Percentage 1 0.39149 0.25846 1.51 0.1305
% Successful Passes in Opponents 1/2 1 0.02319 0.00480 4.83 <.0001

The more statistically oriented reader might be asking the question of how well does this model actually fit the data.  What is the R-Square?  It is small.  The preceding model explains about 5% of the variation in Fullback’s Plus-Minus statistics.

And that is the important point.  The model does its job in that it tells us there is a significant relationship between passing skill and goal differential.  But it is far from a complete picture.  The decision maker needs to understand what the model shows.  However, the decision maker also needs to understand what the model doesn’t reveal.   This model (and the vast majority of other models) is inherently limited.  Like I said last time – the model is a decision support tool / not something that makes the decision.

Admittedly I didn’t try to find a model that fits the data really well.  But I can tell you that in my experience in sports and really any context that involves predicting or explaining individual human behavior, the models usually only explain a small fraction of variance in performance data.

A Non-Judgmental Analysis of the NFL Rating Decline

Over the last week there has been a lot of discussion regarding the decline in NFL ratings this season.  The facts seem to be that the NFL is experiencing a weakness in prime time games that has resulted in an 11% drop in ratings.  The NFL has circulated a memo that cites a variety of factors such as the presidential campaign.  Notably the memo states that there is no evidence that “concern over player protests is having a material impact on our ratings.”

I have gone on record in multiple articles over the past few years talking about the likely impact of high profile or controversial events such as domestic violence incidents and concussions on NFL fandom.  My opinion has been that the NFL would continue to be strong and fans would continue to watch.  So what’s different now?  On this blog, the emphasis is almost always on data driven analyses.  In this case, it’s not possible to take that approach.  I would need much more detailed data on TV ratings and even then I likely wouldn’t have the ability to rule out different possible causes.

The NFL has suggested a confluence of events as the culprit.  I think this is true but perhaps not in the manner the NFL is implying.  I think the NFL is right that the presidential campaign is having an impact.  But I suspect it is having less of a direct impact due to people’s attention being shifted in a different direction.  College football does not seem to have experienced a decline in viewership.

I think it is the nature of the current political campaign and the emotions the campaign is generating.  This campaign has highlighted very distinct cultural differences.  The world views of Trump and Clinton supporters seem to be fundamentally different.  The potential problem is that a lot (majority?) of the NFL’s fan base may lean in the Trump direction while the protests lean in the Clinton direction.  In what follows I’m going to talk about this situation on a theoretical level.  I am making no value judgments about any protests or response to protests – I’m just looking at the marketing and branding issues.

Why are the protests potentially damaging to the NFL brand?  I think there are a couple of related issues.

First, the NFL has been known for shutting down individual expression by players.  Remember it is the No Fun League and it’s all about protecting the shield (brand).  Now, however, we seem to have a protest that is allowed.  And it is a protest with which many fans may disagree.  On some level the NFL seems to be changing its policies to accommodate the protests.  I think it is this “change” that may be the key issue.  Especially if the “change” is to accommodate something that is controversial to the core audience.  If a league is known for shutting down everything from TD celebrations to minor uniform violations then is not shutting something down an implicit endorsement?

The stridency of the current presidential campaign in terms of insiders versus outsider and political correctness makes this type of “authenticity” issue especially salient to certain segments of fans.  The impact may be  subtle.  It may manifest as a softening in enthusiasm or engagement with the NFL brand rather than a decrease in stated preference.  Fans still like the game and the players but maybe they are just not as compelled to watch.  (I don’t have access to the NFL’s data but this may be a tricky issue to assess using traditional marketing research techniques.)

The second, and related issue, is that there are other factors impacting the brand.  The current protests occurred in the wake of seasons that featured domestic abuse and the concussion issues.  The NFL brand may be resilient to any one event but over time problems can weaken the foundation.  This type of subtle brand weakness may be especially relevant given that the NFL is currently lacking some star power.  While the NFL is less of a star driven league than say the NBA, having Peyton Manning retired and Tom Brady suspended makes the league more vulnerable.


Questioning the Value of Analytics: Sports Analytics Series Part 3

Continuing the discussion about organizational issues and challenges, a fundamental issue is understanding and balancing the relative strengths and weaknesses of human decision makers and mathematical models.  This is an important discussion because before diving into specific questions related to predicting player performance it’s worthwhile to first think about how modeling and statistics should fit into an overall structure for decision making.  The short answer is that analytics should serve as a complement to human insight. 

The “value” of analytics in sports has been the topic of debate.  A high profile example of this occurred between Charles Barkley and Daryl Morey.  Barkley has gone on record questioning the value of analytics.

“Analytics don’t work at all. It’s just some crap that people who were really smart made up to try to get in the game because they had no talent. Because they had no talent to be able to play, so smart guys wanted to fit in, so they made up a term called analytics.  Analytics don’t work.” 

The quote reflects an extreme perspective and it is legitimate to question whether Charles Barkley has the background to assess the value of analytics (or maybe he does, who knows?).  But, I do think that Barkley’s opinion does have significant merit.

In much of the popular press surrounding books like Moneyball or The Extra 2% analytics often seem like a magic bullet.  The reality is that statistical models are better viewed as decision support aids.  Note that I am talking about the press rather than then books.

The fundamental issue is that models and statistics are incomplete.  They don’t tell the whole story.  A lot of analytics revolves around summarizing performance into statistics and then predicting how performance will evolve. Defining a player based on a single number is efficient but it can only capture a slice of the person’s strengths and weaknesses.  Predicting how human performance will evolve over time is a tenuous proposition.

What statistics and models are good at is quantifying objective relationships in the data.  For example, if we were interested in building a model of how quarterback performance translates from college to professional football we could estimate the mathematical relationship between touchdown passes at the college level and touchdown passes at the pro level.  A regression model would give us the numerical patterns in the data but such a model would likely have little predictive power since many other factors come in to play.

The question is whether the insights generated from analytics or the incremental forecasting power actually translate into something meaningful.  They can.  But the effects may be subtle and they may play out over years.  And remember we are not even considering the financial side of things.  If the best predictive models improve player evaluations by a couple of percent maybe it translates to your catcher having a 5% higher on base percentage or your quarterback having a passer rating that is 1 or 2 points higher.  These things matter.  But are they dwarfed by being able to throw 10 or 20 million more into signing a key player?

If the key to winning a championship is having a couple of superstars.  Then maybe analytics don’t matter much.  What matters is being able to manage the salary cap and attract the talent.  But maybe the goal is to make the playoffs in a resource or salary cap constrained environment.  Then maybe spending efficiently and generating a couple of extra is the objective.  In this case analytics can be a difference maker.

Understanding the Organization: Sports Analytics Series Part 2

The purpose of this series is to discuss the use of analytics in sports organizations (see part 1).  Rather than jump into a discussion of models, I want to start with something more fundamental.  I want to talk about how organizations work and how people make decisions.  Sophisticated statistics and detailed data are potentially of great value.  However, if the organization or the decision maker is not interested in or comfortable with advanced statistics then it really doesn’t matter if the analyses are of high quality.

Analytics efforts can fail to deliver optimal value for a variety of reasons in almost any industry.  The idea that we can use data to guide decisions is intuitively appealing.  It seems like more data can only create more understanding and therefore better decisions.  But going from this logic to improved decision making can be a difficult journey.

Difficulties can arise from a variety of sources.  The organization may lack commitment in terms of time and resources.  Individual decision makers may lack sufficient interest in, or understanding of analytics.  Sometimes the issue can be the lack of vision as to what analytics is supposed accomplish.  There can also be a disconnect between the problems to be solved and the skills of the analytics group.

These challenges can be particularly significant in the sports industry because there is often a lack of institutional history of using analytics.  Usually organizations have existing approaches and structures for decision making and the incorporation of new data structures or analytical techniques requires some sort of change.  In the earliest stages, the shift towards analytics involves moving into uncharted territory.  The decision maker is (implicitly) asked to alter how he operates and this change may be driven by information that is derived from unfamiliar techniques.

Several key concerns can be best illustrated by considering two categories of analyses.  The first category involves long-term projects for addressing repeated decisions.  For instance, a common repeated decision might be drafting players.  Since a team drafts every year it makes sense to assemble extensive data and to build high quality predictive models to support annual player evaluation.  This kind of organizational decision demands a consistent and committed approach.  But the important point is that this type of decision may require years of investments before a team can harvest significant value. 

It is also important to realize that with repeated tasks there will be an existing decision making structure in place.  The key is to think about how the “analytics” add to or compliment this structure rather than thinking that “analytics” is a new or replacement system (we will discuss why this is true in detail soon).  The existing approach to scouting and drafting likely involves many people and multiple systems.  The analytics elements need to be integrated rather than imposed.

A second category of analyses are short-term one-off types of projects.  These projects can be almost anything ranging from questions about in game strategies or very specific evaluations of player performance.  These projects primarily demand flexibility.  Someone in the organization may see or hear something that generates a question.  This question then gets tossed to the analytics group (or person) and a quick turn-around is required.

Since these questions can come from anywhere the analytics function may struggle with even having the right data or having the data in an accessible format.  Given the time sensitive nature of these requests there will likely be a need to use flawed data or imperfect methods.  The organization needs to be realistic about what is possible in the short-term and more critically the analysis needs to be understood at a level where the human decision maker can adjust for any shortcomings (and there are always shortcomings).  In other words, the decision maker needs to understand the limitations associated with a given analysis so that the analytics can inform rather than mislead.

The preceding two classes of problems highlight issues that arise when an organization starts on the path towards being more analytically driven.  In addition, there can also be problems caused by inexperienced analysts.  For example, many analysts (particularly those coming from academia) fail to grasp is that problems are seldom solved through the creation of an ideal statistic or equation.  Decision making in organizations is often driven by short-term challenges (putting out fires).  Decision support capabilities need to be designed to support fast moving, dynamic organizations rather than perfectly and permanently solving well defined problems.

In the next entry, we will start to take a more in depth look at how analytics and human decision making can work together.  We will talk about the relative merits of human decision making versus statistical models.  After that we will get into a more psychological topic –decision making biases.

Part 2 Key Takeaways…

  • The key decision makers need to be committed to and interested in analytics.
  • Sufficient investment in people and data is a necessary condition.
  • Many projects require a long-term commitment. It may be necessary to invest in multiyear database building efforts before value can be obtained.

A Short Course on Sports Analytics – Part 1

  1. Sports Analytics in Organizations

This fall the plan is to do something a little different with the blog.  Rather than data driven analyses of sports marketing topics, I want to spend some time talking about using analytics to support player and in-game decision making.  The “Moneyball” side of the sports analytics space.

The focus will mainly be at the level of the organization rather than at the level of specific research questions.  In other words, we will talk about providing effective analytics support within an organization, rather than presenting a series of analyses.  My hope is that this evolves to being something of a web based course on using analytics to drive decisions in sports.

I’ve spent a lot of time over the past few decades working on analytics projects (across multiple industries) and I’ve developed opinions about what firms do right and where mistakes are made.  Over the last few years, I’ve thought a lot about how analytics can be used by sports organizations.  Specifically, about how lessons from other industries can be applied, and instances where sports are just different.

The history of statistical analysis in sports goes way back, but obviously exploded with the publication of Moneyball.  A huge number of sports fans would love to be a General Manager but few people have the athletic ability to gain entry as a former player.  Using statistics to find ways to win is (maybe?) a more accessible route.

But this route is not without its complications.  Using stats and data to win games is an intriguing and challenging intellectual task.  What data should be collected?  How should the data be analyzed?  How should the analysis be included in the decision making structure?  These are all challenging questions that go beyond what a fan with some data can accomplish.

What I’m going to do in this series is talk about how to approach analytics from both a conceptual level and an operational level.  Conceptually, I will cover how humans make decisions in organizations.  At the operational level, we will discuss what types of analyses should be pursued.

What I won’t do in this series is talk about specific models.  At least not very much.  I may drop in a couple of analyses.  This limitation is done with purpose.  It’s my feeling that the sports analytics space is overly littered with too many isolated projects and analyses.  The goal here is to provide a structure for building an analytics function and some general guidance on how to approach several broad classes of analyses.

What will this series include?  Some of the content will be based on whatever becomes top of mind or based on the response I get from readers.  But some things will definitely appear.  There will be material about how analytics can best compliment a human decision maker.  I will also talk about how lessons from other industries can be helpful in the sports context.  There are more similarities than differences between sports and “standard” businesses.  But there are some important differences.

We will also talk about models and statistical analysis.  But this will be done in broad terms.  What I mean is that we will discuss classes of analysis rather than specific studies.  For example, we will discuss player selection analyses but the emphasis will be on how to approach the problem rather than the creation of a particular forecasting model.  There are a variety of ways to analyze players.  We can use simple models like linear regression or more complex models that yield probabilities.  We can also forgo the stats and use raw data to look for player comparisons.  We will discuss the implementation challenges and benefits of each approach.

This series is a work in progress.  I have a number of entries planned but I’m very open to questions.  Shoot me an email and I’ll be happy to respond in future entries or privately (time permitting).

Next: Understanding the Organization

Medaling at the Olympics: Is Corruption the Golden Ticket?

A Guest post from my friend and colleague at Emory – Tom Smith!

by Thomas More Smith

Even before the Olympic flame in Rio was lit, there were significant concerns regarding doping and competitive balance. In June, 2016, the IAFF banned the Russian athletic team (those competing in track-and-field events) from the Rio Olympics after Russia failed to show it had made progress in light of the World Anti-Doping Agency’s report on state sponsored doping by Russia. After a considerable amount of concern and angst by Russian Olympians, the IOC decided not to ban the entire Olympic squad.

The issue of fair-play at the Rio Olympics has been front and center since the opening ceremonies. There is clearly some bad blood between competitors in the Olympic swimming events. At a press conference on Monday, August 9, Lilly King, the U.S. swimmer and Gold medalist of the 100-meter breast stroke, made pointed remarks about the Russian Silver medalist, Yuli Efimova, who was, until several weeks ago, banned from Olympic competition because of positive drug tests.  The Gold medalist of the men’s 200-meter freestyle even, Sun Yang, was the subject of testy comments from Camille Lacourt, who took fifth in the event. Lacourt suggested his Chinese competitor “pisses purple” in reference Sun’s failed drug test several years ago.

In both of these situations, athletes who had at one time been found to have taken PEDs were standing on the medal podium. Are these athletes clean now and will their medals stand? In 2012, Nadzeya Ostapchuk from Belarus won the Gold medal in women’s shot put. The IOC subsequently withdrew her medal and her standing after she tested positive for anabolic steroids.  Other athletes at the 2012 games and 2008 and 2004 games were stripped of their medals after they tested positive for various PEDs.

This leads to an interesting question – do dirty athletes win more medals? Or, perhaps, do athletes from “dirty” programs or countries win more medals?

How Much Advantage do PEDs Provide?

There is no data on athletes currently taking PEDs – we only know about the athletes that have taken PEDs and eventually tested positive for them. Also, we can suspect that some athletes did or didn’t take PEDs during the Olympics but we don’t really know unless they were tested and the results were positive. Still, some athletes have been able to avoid positive tests for years because of the drugs, testing facilities or advanced systems in place to mask the drugs (see, for example, Lance Armstrong.) As such, it is a little tricky to test the relationship between PED use and performance in sporting events. However, we can examine the relationship between Olympic performance and the perceived level of corruption of the country of the athlete – what I will call the “dirty” country hypothesis.

H0: Athletes from countries with more corruption are more likely to win Olympic medals.

Perceived Level of Corruption

The organization Transparency International compiles a corruption perception index tracking the level of perceived corruption by country and by year. The Corruption Perceptions Index scores countries on a scale from 0 (highly corrupt) to 100 (very clean). No county has a perfect score (100); the top four countries of Denmark, Finland, New Zealand and Sweden regularly score between 82 and 92. Nearly two-thirds of the 170 countries identified by Transparency International score below 50.


Data: Transparency International and ESPN

Using data from the 2012 Olympics, I ran a correlation plot of the total Olympic medal count and the Corruption Perception Index (CPI) for each country with 10 or more total medals.  The plot of the total medal count for each country relative to the Country’s CPI is shown in the figure above. We can see that New Zealand, for example, is perceived as very un-corrupt (Index = 90) but also has a low medal count (13), while Russia has a much higher perceived level of corruption (index = 27) and a high medal count (79). The plot of the best-fit line shows a positive correlation. That is, although Russia and China have high medal counts and high levels of perceived corruption, the overall trend suggests that countries with less perceived corruption tend to also perform better at the Olympics.

Although it looks like some countries do poorly because of corruption, this may not be the case. Of course, correlation does not mean causation. In addition, this plot does not take into consideration the size of the Olympic team. Azerbaijan, for example, had 10 medals in the 2012 Olympics and had a PCI of 27. But, Azerbaijan only sent 53 athletes to the Olympics — a considerably smaller team than Ukraine, which had 19 medals, a PCI of 26 and 237 athletes. So, perhaps the countries with higher perceived corruption might have performed better at the Olympics if they had sent more athletes. When the medal count is adjusted for team size (Total Medals / Total Athletes) and plotted against the PCI, we get the figure below.


Data: Transparency International and ESPN

In this figure, the correlation has reversed– countries with higher perceived corruption also have higher level of medals per athlete in general. When accounting for the size of the team, countries such as Kenya and Azerbaijan tend to do pretty well (as does China and Russia). The United States still performs well, but does not have as high a medal per athlete count as China or Kenya.

What does this mean?

It is unwise to use figures like this to suggest that the Kenya Olympic team are full of drug cheats or that the Chinese team is engaged in dubious behavior. It’s also unwise to suggest the United States has completely clean athletes (we know, for a fact, that this is not the case!) But, given that there are seemingly strong correlations between perceived corruption and Olympic performance, it is understandable that some athletes would be vocal about the behavior of the person in the next lane based on the country the athlete is playing for.

Amateur Sports and Brands

HBO Sports recently created a detailed report on the IOC.  The RIO Olympics do not come off well.  Pollution, doping, corruption and athlete exploitation are at the top of the list.  It is a fascinating story that seems to play out with each Olympic Games.

This issue of fair compensation for the athletes is high on the list. The number discussed in the report was $4 billion.  The question is whether and how this money from rights fees and sponsors should be allocated to the athletes.  Is (and should) there be an Olympic Ed O’Bannon?

In many respects this starts to sound like the debates about college sports in the US.  These debates are usually cast in terms of fairness.   to the athletes versus arguments about the purity of the sport or appropriateness of academic institutions running pro teams.

These debates are at best incomplete without considering the role of marketing and brands.  While college football players supply the product, the brands owned by the colleges or the Olympics is what drives fan interest.  Leonard Fournette is a Heisman favorite and a huge star.  But does he draw fans to LSU.  the truth is he probably doesn’t (in the short-term).  In the long-term its stars like Fournette that create the brand equity. 


Likewise, in the case of the Olympics – we could ask how much interest in driven by the current athletes?  and how much is driven by the attachment people have to the Olympics (the brand).


I think (in the US) the Olympic brand is about Carl Lewis, Bruce Jenner, Mary Lou Retton, Jesse Owens, Cassius Clay or many others.  It remains to be seen who from the current crop breaks out.

The real problem, I believe is one of equity.  This is true in both college sports and the Olympics.  The fundamental issue is who gets to harvest the value of the brands.  The problem – to many folks – is that this seems to just end up being the people that control the institutions at any one moment.  The athletes that have built the brands (the stars of the past) and the athletes that create the product (this years athletes) tend to get left out in the cold.


NFL Bandwagon Fans and the Business of Fan Rankings

The Business behind Fan Base Analysis: Sponsorship Insights

Today’s post is a follow up to the NFL fan base rankings post.  The annual NFL Fan base ranking involves a combination of data analysis and marketing ideas (brand equity).  I do them as a single ranking to make it easily digestible and to encourage conversation.  Or in the case of Raider Fans – to generate threats.  Today, I go beyond a single ranking and present multiple fan base metrics.  The goal is to provide a richer description of how teams’ fans compare.  Specifically, we present rankings focused on brand equity, social media, road attendance and “bandwagon” behavior.

The fan analysis material is meant to be both instructive and to provide material for debate.  Sports brands are unique in the degree of loyalty that exists between fans and teams.  The reaction to the fan base rankings highlights the intensity of the relationships as people take it very personally when their fandom is questioned.  It’s interesting that it matters to fans not only that their team is competitive but that their passion for their team also exceeds the opposition’s.  As such it’s crucial for teams to thoroughly understand the strengths and weaknesses of their fan bases.

Something that tends to get lost in the discussion of fan base rankings is that the results have very significant business implications.  The fan equity and other measures that we discuss today tell an essential story about fans in each city.  If I am a brand looking to sponsor a stadium or a fast food company looking to do a deal with a team, then I very much want to know about the underlying long-term passion and behaviors of the fan base.

A common approach for valuing sports properties is the use of comparables.  The basic idea is that some entity, like a team or player, can be valued by looking at similar teams or players.  For example, a way to value a team is to look at previous sales and then make some adjustments for differences in population or income across markets.  Stadium naming deals are often similarly driven by past deals.

The Fan Equity work and rankings below provide extra factors that can be added to analyses based on comparables.  The rankings can be used to go beyond demographics driven comparisons to include a measure of engagement or loyalty.

In what follows, I provide a few insights about each of the metrics and then a Table that provides a complete breakdown.  I also discuss the business relevance of each of metric.  There are a number of caveats that should be offered such as the importance of looking at multiple metrics or noting that the results rely on public data.  But these explanations are a bit tedious and the key point is that the metrics should be carefully interpreted.

One important factor that should be stressed is that all of the measures are based on market place behaviors of fans like attending games and following on social media rather than consumer opinions collected via surveys.

The rankings should be interpreted with care.  A high ranking on the brand equity measures is something to strive for while a high ranking in the bandwagon category is something to avoid.


Fan Equity

The Winners: Cowboys, Patriots and Ravens

The Losers: Jaguars, Raiders and Dolphins

Fan Equity is the core of the Dynamic Fan Equity (DFE) metric used to summarize fan bases.  It looks at home revenues relative to expected revenue based on team performance and market characteristics.  The goal of the metric is to measure over (or under) performance relative to other teams in the league.  In other words, statistical models are used to create an apples to apples type comparison to avoid distortions due to long-term differences in market size or short-term differences in winning rates.

In terms of business concepts, this measure is similar to a “revenue premium” measure of brand equity.  It captures the differentials in fans willingness to financially support teams of similar quality.  From a business or marketing perspective this is a gold standard of metrics as it directly relates to how a strong brand translates to revenues and profits.

However, the Fan Equity context is sports, and that does make things different.  At a basic level sports organizations have dual objectives.  They care about winning and profit.  That is important because sometimes teams aren’t trying to maximize revenues (Packers, Steelers, etc…).   When this is the case the Fan Equity metric understates the engagement of fans.

What is the importance of Fan Equity for sponsorship?  Fan Equity shows the relative commitment to spend to support the team.  If we make the assumption that paying a premium (remember the model controls for the income differences across markets) is correlated with passion then teams with higher fan equity have fans that are more deeply bonded to the team.  These teams should receive a bump in terms of sponsorship deals.


Social Media Equity

Winners: Patriots, Cowboys and Broncos

Losers: Rams, Chiefs and Cardinals

An issue with the Fan Equity measure is that it can be constrained by capacity or by team pricing decisions.  If teams have a small stadium or are NOT pricing to maximize revenues then the Fan Equity measure can understate the team’s following.  In contrast to buying a ticket, following on social media is free and not impacted by geography.  It’s just as easy to follow the Seahawks as it is to follow the Falcons while sitting in Atlanta.

Social Media Equity is also an example of a “premium” based measure of brand equity.  It differs from the Fan Equity in that it focuses on how many fans a team has online rather than fans’ willingness to pay higher prices.  Social Media Equity is also constructed using statistical models that control for performance and market differences.

In terms of business application, the social media metric has several implications both on its own merits and in conjunction with the Fan Equity measure.  For example, the lack of local constraints, means that the Social Equity measure is more of a national level measure.  The Fan Equity metric focuses on local box office revenues while the social metric provides insight into how a team’s fandom extends beyond a metro area.

Social Media Equity may also serve as a leading indicator of a team’s future fortunes.  For a team to grow revenues it is often necessary to implement controversial price increases.  Convincing fans to sign expensive contracts to buy season tickets can also be a challenge.  Increasing prices and acquiring season ticket holders can therefore take time while social media communities can grow quickly.  Some preliminary analysis suggests that vibrant social communities are positively correlated with future revenue growth.

A comparison of Fan Equity and Social Media can also be useful.  If Social Media equity exceeds Fan Equity it is evidence that the team has some marketing potential that is not being exploited.  For example, one issue that is common in sports is that it is difficult to estimate the price elasticity of demand because demand is often highest for the best teams and best seats.  The unconstrained nature of social media can provide an important data point for assessing whether teams have additional pricing flexibility.


Road / Diaspora Equity

Winners: Eagles, Cowboys, Giants and the Bills in TOP TEN!

Losers: Chiefs, Cardinals and Texans

This is a new metric for the blog and a vocabulary lesson all in one.  One way to look at fan quality is to look at how a team draws on the Road.  For example, in the NBA these effects are pronounced.  Lebron or a retiring Kobe coming to town can often lead to sell outs.  College football is especially noted for traveling fans (SEC!).  A fan base that travels is almost by definition incredibly passionate.

This one has a bit of a muddled interpretation.  If a team has great road attendance is it because the fans are following the team or because they have a national following?  In other words, are fans traveling to the game or just showing up because it’s the Cowboys or Steelers?  Furthermore, if it is a national following is it because the team is popular across the country or because a lot of folks have moved from Pittsburgh or Buffalo to the Sun Belt?

Road Equity tells a story and suggests a need for additional research.  A national following is a great characteristic that might suggest that a team’s brand is on an upswing.  Or it might be that the city itself is on a downward trajectory.  Road equity might also be a matter of temporary factors (beyond winning) if fans are drawn to star or controversial players.


Band Wagon Fans

Biggest Bandwagon Fans: Cardinals and Cowboys

Loyal to a Fault: Bills, Lions and Redskins

This ranking looks at how responsive attendance is to winning.  This is a fun one because there are two really different interpretation of the results.  The more negative one is that a team whose fans show up less when the team is losing has a “fair weather” or “band wagon” fan base.  The other interpretation is that fans that are sensitive to winning are more demanding of quality.  The former seems most likely.

The rankings come directly from a statistical model of attendance.  The top ranked bandwagon fans are the ones whose attendance is most sensitive to winning.  Based on the data and models the Arizona Cardinal fans are the most “Bandwagon” of all the fan bases.  On the other extreme we have the Bills, Lions and Redskins fans as the most loyal.

From a sponsorship perspective, a high bandwagon ranking might make a sponsoring brand leery.  If fans only show up when a team is winning then the team might not have the relationship intensity with fans that a sponsor is trying to leverage.  An important reason for sports sponsorships is that brands want to be associated with teams that fans live and die with.  If a team is just entertainment then maybe a sponsorship is not going to generate the associations and connections desired.

There is complexity in the real world and all of these measures have limits.  The Cowboy fans are an interesting case study.  The Cowboys rank #2 in bandwagon fandom but they also rank very highly in the other brand equity measures.  Cowboy fans buy tickets and follow their team on social media.  The national stature of the Cowboys also brings in fans on the road.  But in terms of actually showing up at games it seems like the fans need a winner.  Loyalty in terms of spending but fair weather in terms of showing up.

The Best NFL Fans 2016: The Dynamic Fan Equity Methodology

The Winners (and Losers) of this years rankings!  First a quick graphic and then the details.


It’s become a tradition for me to rank NFL teams’ fan bases each summer.  The basic approach (more details here) is to use data to develop statistical models of fan interest.  These models are used to determine which cities fans are more willing to spend or follow their teams after controlling for factors like market size and short-term variations in performance.  In past years, two measures of engagement have been featured: Fan Equity and Social Media Equity.  Fan Equity focuses on home box office revenues (support via opening the wallet) and Social Media Equity focuses on fan willingness to engage as part of a team’s community (support exhibited by joining social media communities).

This year I have come up with a new method that combines these two measures: Dynamic Fan Equity (DFE).  The DFE measure leverages the best features of the two measures.  Fan Equity is based on the most important consumer trait – willingness to spend.  Social Equity captures fan support that occurs beyond the walls of the stadium and skews towards a younger demographic.  The key insight that allows for the two measures to be combined is that there is a significant relationship between the Social Media Equity trend and the Fan Equity measure.  Social media performance turns out to be a strong leading indicator for financial performance.

Dynamic Fan Equity is calculated using current fan equity and the trend in fan equity from the team’s social media performance.  I will spare the technical details on the blog but I’m happy to go into depth if there is interest.  On the data side we are working with 15 years of attendance data and 4 years of social data.

The Winners

We have a new number one on the list – the New England Patriots. Followed by the Cowboys, Broncos, 49ers and Eagles.  The Patriots victory is driven by fans willingness to pay premium prices, strong attendance and phenomenal social media following.  The final competition between the Cowboys and the Patriots was actually determined by the long-term value of the Patriots greater social following.  The Patriots have about 2.4 million Twitter followers compared to 1.7 for the Cowboys.  Of course this is all relative a team like the Jaguars has just 340 thousand followers.

The Eagles are the big surprise on the list.  The Eagles are also a good example of how the analysis works.  Most fan rankings are based on subjective judgments and lack controls for short-term winning rates.  This latter point is a critical shortcoming.  It’s easy to be supportive of a winning team. While Eagles fans might not be happy they are supportive in the face of mediocrity.  Last year the Eagles struggled on the field but fans still paid premium prices and filled the stadium.  We’ll come back to the Eagles in more detail in a moment.

The Strugglers

At the bottom we have the Bills, Rams, Chiefs, Raiders and Jaguars.  This is a similar list to last year.  The Jags, for example, only filled 91% of capacity (ranked 27th) despite an average ticket price of just $57.  The Chiefs struggle because the fan support doesn’t match the team’s performance.  The Chiefs capacity utilization rate ranks 17th in the league despite a winning record and low ticket prices.  The Raiders fans again finish low in our rankings.  And every year the response is a great deal of anger and often threats.

The Steelers

The one result that gives me the most doubt is for the Pittsburgh Steelers.  The Steelers have long been considered one of the league premier teams and brands.  The Steelers have a history of championships and have been known to turn opposing stadiums into seas of yellow and black.  So why are the Steelers ranked 18th?


A comparison between the Steelers and the Eagles highlights the underlying issues.  Last year the Steelers had an average attendance of 64,356 and had an average ticket price of $84 (from ESPN and Team Market Report).  In comparison the Eagles averaged 69,483 fans with an average price of $98.69.  In terms of filling capacity the Steelers were at 98.3% compared to the Eagles at 102.8%.  The key is that the greater support enjoyed by the Eagles was despite a much worse record.

One issue to consider is that of pricing.  It may well be that the Steelers ownership makes a conscious effort to underprice relative to what the market would allow.  The high attendance rates across the NFL do suggest that many teams could profitably raise prices.  It’s entirely reasonable to argue that the Steelers relationship to the Pittsburgh community results in a policy of pricing below market.

In past years the Steelers have been our social media champions.  This past year did see a bit of a dip.  In terms of the Social Media Equity rankings the Steelers dropped to 5th.    As a point of comparison, the Steelers have about 1.3 million Twitter followers compared to 2.4 million for the Patriots and 1.7 million for the Cowboys.


The Complete List

And finally, the complete rankings.  Enjoy!


End of an Era – Goodbye Manish

A fond farewell and a new era –

Things change.  Sometimes for the good and sometimes not.  We (Manish and myself) started this blog a few years ago as a means for turning our love into sports into an academic pursuit.  Its been a lot of fun and and a lot of work.  Its taken us into different ways of thinking and exposed us to a lot of interesting media.



But its come to an inflection point.  Manish has decided to leave academia.  Nothing wrong with that, but it does mean he needs to step off the platform.  Its one thing for an academic to publish findings that insult Raiders or Duke Blue Devil fans.  Its another for someone in the corporate world.

He is already missed.  The best thing about this line of work was that it was fun and we had a shared purpose.   We also did a lot of other stuff related like teach several sports courses here at Emory.  We will have to see how all this evolves.  at a minimum there will likely be far more spelling errors and typos.  But fewer !!!!!

I won’t get too sentimental but its a huge loss.  And I’m genuinely sad.




The latest work from Professor Mike Lewis