email us

SMARTs Logo350

Introduction and Purpose

The NFL is the most popular and highest revenue generating sports league in the United States. It earned approximately $9.5 billion in 2012 compare d to baseball’s (MLB) $7.5 billion and basketball’s (NBA) $4 billion. The NFL is comprised of 32 teams across a variety of markets, and the bulk of its income comes from media, sponsorship, and NFL Ventures deals. As a private company, it reveals limited financial information to the general public. For insights into the revenue composition of a team, we turn to the Green Bay Packers, the only one required to release detailed financial information due to its status as a public company. As shown on Figure 1.1, Green Bay’s recent report suggests that over one-half of a team’s revenue is contributed by the NFL via a league-wide revenue sharing agreement.

Based on this report, we estimate that about 40% of a team’s revenue is likely to come from its own local initiatives. Mark Murphy, president of the Green Bay Packers, attributes significant revenue growth to his team’s strong 13-0 start in 2012 when it was the defending Superbowl champ. Accordingly, each local Green Bay fan is estimated to be worth $386 in revenue (Forbes, 2013). It appears, therefore, that a winning team is the obvious first step to growing local revenue. It improves home game attendance and boosts the opportunity to sell on premise and at local events.

Many NFL teams now actively pursue an analytics strategy to win on and off the field. At the MIT Sloan Sports Analytics Conference 2013, a panel discussion on football analytics focused on the use of algorithms to accurately predict player performance relative to team personnel and salary cap constraints. In Green Bay, for example, players’ expense equals nearly one-half of the team’s revenue (Forbes, 2013). Consequently, teams that successfully apply analytics can more cost-effectively navigate personnel costs—by far the largest expense and initial impetus for analytics among teams—in the college draft, free agency, and related salary caps while increasing the probability of winning. Professors Massey and Thaler studied player personnel management in the NFL and concluded that teams generally do not follow a rational approach in drafting and, thus, often overvalue and pay for a high draft pick to the detriment of improving their winning chances (Massey & Thaler, 2013). As Massey and Thaler proved, analytics is a key factor to a winning business model in the NFL.

To inform our analysis, we gathered information from a variety of sources including: Northwestern University library articles and peer-reviewed journals, strategy books, sports/business popular press online, and football panel discussions conducted at the recent MIT Sloan Sports Analytics Conference 2013. Worth noting is that analytics in the NFL has only gained wider coverage in the last five years and, therefore, most of the recent data on specific teams can only be found in the business and sports press. In these instances, we have taken additional steps to corroborate the information across multiple sources.

The purpose of this paper is to examine use of analytics among analytically advanced teams to achieve and sustain a competitive advantage in the NFL. In doing so, we initially reviewed the following teams on the extent of their analytical capabilities: San Francisco 49ers, New England Patriots, Dallas Cowboys, Jacksonville Jaguars, and Buffalo Bills. The area of analytics is a broad one and, in the specific business of the NFL, there are many factors that define success. We have determined that these teams have made a commitment to competing on analytics specifically in areas of player recruiting/team composition, in-game strategy, and fan experience. As a result, they are achieving varying degrees of success as they proceed through the adoption cycle.

Teams at Various Stages
Davenport and Harris, in their book, “Competing on Analytics,” describe the five stages of an organization’s analytical life cycle, from least enabled to most. These include firms that are analytically 1) impaired, 2) localized, 3) aspirational, 4) starting to compete, and 5) true competitors (Davenport & Harris, 2007). In the context of the NFL, the graphic below shows the teams we researched relative to their adoption timeline.

From our research, we have identified the New England Patriots and San Francisco 49ers as analytical competitors as defined by Davenport and Harris. From C-level management endorsement to hiring highly skilled analysts to cultivating a fact-based culture of testing and learning—both teams have integrated analytics as a core component of their business strategies. In contrast, other teams with analytical aspirations have only recently achieved executive support to embark on the journey undertaken by the Patriots and 49ers more than a decade ago. For the remainder of this paper, we will focus on the New England Patriots and San Francisco 49ers, teams that have overcome challenges and achieved success in their analytics programs, all the while becoming examples for other aspiring teams.

NFL Analytics Adoption Cycle

Recruit and Manage Player Personnel

In 2000, Robert Kraft, owner of the New England Patriots, hired Bill Belichick as head coach and general manager, giving Belichick nearly complete control of his football operations. Belichick, an economics major in college, was known for his analytical approach to coaching, salary cap negotiations, and team building. He is a consummate teacher and student of the game. To date, much of the academic research on personnel management in the NFL have concluded that franchises taking a superstar approach to recruiting contribute to market inefficiencies and ultimately hurt their teams (Borghesi, 2007; Massey & Thaler, 2013). In this regard, Belichick, in his first year used a combination of traditional and nontraditional metrics to uncover the overlooked and undervalued Tom Brady in the 2000 NFL draft. According to the Patriots’ president at the recent MIT Sports Analytics Conference, Belichick was surprised that a player of such caliber by his metrics was still available after several rounds. Belichick pushed hard, ultimately drafting Tom Brady as the 199th pick in the 6th round. Today, with three Super Bowl wins together (Brady was MVP in two) and more than a decade later, the Belichick and Brady coach-quarterback tandem is the winningest ever.

The Patriots’ extensive culture of testing and learning gives the team the empirical data to understand which key performance indicators (KPIs) predict success within its system. Hence, it is not surprising that the Patriots often de-emphasize more traditional KPIs in favor of those designed internally. One example of a less favored conventional metric in New England is the 40-yard dash—the default speed test that most teams employ to find the fastest player for a given position. In contrast, the Patriots’ evaluation places greater weight on reactive speed—the speed in which a player processes what is happening and responds quickly. Wes Welker, former wide receiver for the team, is an example of a non-conventional evaluation and good fit within the Patriots’ system. At 5’ 9” in height and running the 40-yard dash in 4.65 seconds, Welker is neither tall nor fast by traditional wide receiver measures. From previous success data, the Patriots knew that more important in its system was a wide receiver’s reaction speed, ability to handle press coverage (getting jammed) at the line, and achieve separation in small closed spaces. Welker is compact, strong, and possesses the closed space quickness needed to succeed in the Patriot’s offense. As important from a personnel management and salary cap strategy, the Patriots did not pay the premium price for the rights to sign Welker, a free agent with the Dolphins in 2007, opting instead to trade for him using often overvalued draft picks (Massey & Thaler, 2013). Welker went on to lead the league in receptions in 2007, 2009, and 2011. He currently holds the four highest single season reception totals in Patriots history.

The Patriots learn from their mistakes in their unwavering commitment to identifying the most effective KPIs. According to Scott Pioli, former VP of Player Development, certain physical measurements such as the size of a quarterback’s hand matters. Pioli “obtained this data” in 2003 when he evaluated Kliff Kingsbury, a productive college quarterback, who had one of the smallest hands among quarterbacks trying out. When asked to throw in New England’s inclement weather, Kingsbury had difficulty controlling the ball. It was then that Pioli realized Tom Brady’s large hands helped him grip and throw the ball accurately. As a result, Pioli added hand size as another variable predictive of a quarterback’s success (MIT Sports Analytics Conference, 2013).

Like the New England Patriots before them, the San Francisco 49ers have also achieved the status of analytical competitor. Since 2001, when the 49ers brought in Paraag Marathe from Bain to head its analytics strategy, the team continues to hire data scientists and increase its capabilities. Marathe is now the team president and, according to him, football analytics presents a challenge due to so many covariant factors going on at the same time (MIT Sports Analytics Conference, 2013). Ultimately, success lies in the combination of talented players, coaches, and systematic ability to build a cost-effective roster. Marathe goes on to say that analytics plays a very important role in player evaluation, and the teams doing it well are the ones that understand how use it in the context of their systems. Whatever is quantifiable can be analyzed but more crucial is finding the metrics that strongly predict peak performance (MIT Sports Analytics Conference, 2013).

The 49ers’ culture and analytics experience is unlike other aspiring teams, including the Baltimore Ravens and Jacksonville Jaguars, who have only recently obtained management support to develop analytics capabilities. Consequently, whereas less analytically enabled teams may only be using analytics to identify performance outliers, the 49ers evaluate a variety of performance indicators specific to its system. Referring to Jerry Rice, former 49er and all-time football great as a model, Marathe describes the team’s focus on speed in the “Flying 20”—the distance going from 20 to 40 yards—as a key predictor of success at its wide receiver position. This is the speed in which a wide receiver achieves separation from the defender and, thus, has the best opportunity to break away for a big play. Marathe goes on, saying that Rice was slow by conventional metrics but, more importantly for his team, is among the fastest ever in the “Flying 20.”

The 49ers are at the forefront of exploring innovative ways to apply analytics. In the first quarter of 2013, the tech-aggressive team collaborated with SAP to develop a draft “app” that allows scouts to access and process large amounts of data in real time. The information can be retrieved on any secure device at any time, thus fully leveraging technology and analytics to reduce the burden of subjective manpower scouting. In this regard, its team president stated, “As the fight for each player becomes more intense, we are looking for advantages.” The app provides just that. Now a scout visiting a school can enter the 40-yard dash time of a draft candidate, and the new information can be processed quickly to show how the time compares with other players under consideration (Henschen, 2013). Hence, the data and subsequent analysis puts the 49ers in a better position to identify market inefficiencies and then trade over or undervalued draft picks to productively accommodate salary cap constraints and optimize team composition.

In the business of football, accurately assessing a player’s skills to determine fit within a team’s environment represents a compelling advantage. Analytically advanced teams like the Patriots and 49ers can accurately assess what players are worth, both from a performance and monetary value standpoint, which enables them to optimally balance personnel costs between draft players and free agents. Consequently, these teams can rationally approach the college draft and free agent negotiations in their efforts to build a cost-effective winning team. As Davenport and Harris suggest, analytical competitors must continue to evolve in order to sustain their competitive advantages (Davenport & Harris, 2007). While salary cap and personnel management served as the primary impetus for developing analytics capabilities, early adopters like the Patriots and 49ers see the value of analytics across their entire operations. In the following section, we describe other business applications that present the natural evolution of analytics in these teams’ strategies.

Inform In-Game Strategy
Football is like a game of chess wherein it is a battle of the minds between two teams. Since the league’s inception, teams have employed “quality control” coaches to analyze and scout opposing teams. Therefore, it is only natural that an analytical competitor like the New England Patriots would take this function to a higher level by using empirical statistics to inform in-game decisions. The NFL, founded in 1920, has just wrapped up its 47th Super Bowl. There are plenty of historical data statisticians can model to see what plays in the past yielded what results. Bill Belichick and the New England Patriots are among the most ardent users of analytics to inform in-game decisions. When asked for his opinion regarding a controversial game theory study that concluded teams are “too conservative” on fourth down, Belichick undoubtedly agreed with the many points made by its author (Lavin, 2005).

In areas like game strategy, teams that utilize analytics can leverage statistics to eliminate guesswork when it comes down to fourth down and inches (Bell, 2013). Although not widely known, Thomas Dimitroff, former head of scouting for the Patriots and now general manager of the Falcons notes, “We use analytics to eliminate as much guesswork as we possibly can.” In addition to leading in the scouting and personnel development area, the New England Patriots also lead the league in leveraging analytics for in-game strategy. According to Davenport, the Patriots do not shy away from using analytical data to make play-calling decisions—whether it is deciding to punt on fourth down, or going for one point or two after a touchdown. In the last couple of years, Belichick has been one of the most aggressive coaches when it comes to going for it on fourth down (Sauser, 2008). To illustrate, one of the most controversial and notable football plays to date is Belichick’s decision to attempt a fourth down conversion with two minutes left in the game against the Colts. While results are not always favorable, what most people don’t factor in is that Belichick owns the highest success percentage of fourth down attempts. And when asked if he would go for it again in that situation, Belichick’s answered he would because it’s the situation which dictates that he takes a chance (Card, 2009). Along these lines, most people have forgotten that in the 2003-2004 AFC Championship game, the Patriots converted on a couple of critical fourth down situations to beat the Colts, ultimately going to the Super Bowl and winning (ESPN, 2004).

According to Burke, operator of the blog Advanced NFL Stats, the probability of a team winning is 79% when going for it on fourth down with less than two minutes left in the game as compared to a 70% winning probability when the team punts (Burke, 2009). The Patriots have obviously benefited from Belichick’s reliance on statistical analyses. Unfortunately, there is no available data that tracks a team’s decisions based on analytics versus intuition. Perhaps because of this, Nate Silver suggests, “The reluctance of NFL teams to adopt analytical approaches is sometimes attributed to the limitations of statistics in the sport (Silver, 2014)." There is no hard evidence that using in-game analytics will increase the odds of winning, but the Patriots definitely make a good argument for it.

In the article “On the Effect of Coaching,” Burke argues that the impact coaches have on the game is minimal. He found very little variance left to explain a game’s results outside of randomness and player impact. There are also instances where the same coach has wildly different results from year to year or when on different teams. One explanation for this is that the coaching staff in the NFL has gotten very efficient. As a result, it is hard to differentiate one team from another if each team is using the same data to make decisions. An argument against the efficiency of play calling is “that the typical NFL team sacrificed about half a win over the course of the 16-game regular season due to its inferior fourth down strategy (Silver, 2014).” Coaches may have the data to support a play decision but could override the call based on their gut feeling. Perhaps the edge NFL coaches have is mostly from how the team is built either through trades or the NFL draft, but the popular view remains that coaches play an important role in a game’s outcome. While difficult to quantify, people talk about the genius of coach Belichick and his influence on the Patriots’ success across multiple factors.

Unlike in skills assessment, with its longer history and proven tangible benefits in personnel management, analytics to inform in-game decisions is a work-in-progress and, thus, its value proposition has yet to be proven. In this regard, it is likely that an analytical competitor, like the New England Patriots, will continue to test in-game analytics by collecting data and refining models within its system. And the Patriots will not stop there. Not only will the team apply analytics internally to optimize personnel and in-game decisions but, like its analytical counterpart the 49ers, will also take it to the next stage by optimizing the complete game experience and making it more fun for its external stakeholders—the fans.

Enhance Fan Experience

In recent years, analytics has been revolutionizing the way in which NFL fans can interact with their favorite players and teams, thus creating more personalized relationships with an already committed fan base. This trend continues to evolve, and in January 2014, the NFL signed a contract with Extreme Networks Inc. to be the official provider of Wi-Fi analytics for the league. “There's nothing that can quite replace that immersive experience of being surrounded by fellow fans at an NFL game. But we do know that the home viewing experience—or the ‘couch-gating’ phenomenon—has improved enormously," said NFL chief information officer Michelle McKenna-Doyle during a press conference at NFL headquarters in New York (Poeter, 2014).

“Couch-gating” is immensely popular in the NFL, as fans have demonstrated their willingness to tune in to games en masse every week of the season. Accordingly, television networks have paid out record contracts for the right to televise NFL games. At $5 billion in 2014, NFL television revenues exceed that of any of the other major U.S. sport, despite the fact that the 16-game NFL season is far shorter than that of its MLB or NBA counterparts. This represents more than 52% of the $9.5 billion in total revenue generated by the NFL in 2014 (Isidore, 2013). There is clearly a major incentive to keep fans happy and engaged, whether they are in-stadium or at home.

Not only have the New England Patriots established themselves as a dominant player in leveraging analytics for on-field success but, as our research suggests, they should also be the model against which all other NFL teams are judged for their utilization of analytics off the field. The Patriots have created an environment at Gillette Stadium that is enjoyable for all fans, regardless of age or gender. One strategy the team has employed involves capturing at-game, quantitative measurements of stadium amenities such as food queue waiting time, availability of parking, and bathroom cleanliness (Davenport, 2008). Such initiatives demonstrate a sustained commitment to maintaining the type of organizational excellence that has translated into rabid fans and a routinely sold out stadium.

The San Francisco 49ers have undergone a monumental turnaround over the past few seasons, transforming from “cellar dweller” to annual contender. As the 49ers break ground on a new stadium, the team’s executives have placed a critical importance on the ability to understand and interact with fans in real time. The 49ers’ previously referenced partnership with SAP placed them on the cutting edge of NFL fan experience offerings. Professional sports franchises and leagues are extremely concerned about slumping attendance at stadiums due to competition from television, computer gaming and lofty ticket costs (Veverka, 2013).In this regard, the new technology will enable the 49ers to create customized offers for fans, text them coupons as they enter the stadium, or recognize their birthdays with free concessions.

As teams such as the Patriots and 49ers dominate not only on the field, but in their stadiums and the homes of their fans as well, the rest of the league will surely take notice. NFL fans will soon broadly expect their favorite teams to offer a similar experience to that which the “true competitors” do today. Those teams that refuse to make similar investments run the risk of alienating their fans and will potentially leave significant streams of revenue under-utilized. The leading organizations are changing from top down and instilling analytics at the core of their operations. Creating an outstanding fan experience is simply one of the many desired outcomes.

In an ever-increasing competitive world, businesses are tasked with discovering new and innovative ways to differentiate themselves. This is especially true in the sports industry and, specifically, in the NFL where over 40% of a team’s revenue is likely to come from its own local initiatives. In order to make smarter, more efficient and effective business decisions, teams in the NFL must instill an analytics practice in some form or fashion, internally or externally, to compete and differentiate themselves. Analytical leaders such as the New England Patriots and 49ers have truly paved the path for the skeptics in this industry. In the business of football, these two teams have demonstrated a remarkable combination of on and off-field analytical practices that have given them a competitive advantage to revolutionize this game and industry. The Patriots’ and 49ers’ game records and loyal fan base are further proof of that. Aspiring teams, such as the Jaguars and Bills, who are just starting to adopt an analytical framework, have much to learn from these key players. To truly sustain this practice for the long haul though, we believe that “Analytics themselves don’t constitute a strategy, but using them to optimize a distinctive business capability certainly constitutes a strategy” (Davenport & Harris, 2007). Evidently, this combination of management heuristics and fact-based decision making—supported organization-wide—will continue to lead teams like the Patriots and 49ers on the long-term path of winning on and off the field.


You are viewing the text version of this site.

To view the full version please install the Adobe Flash Player and ensure your web browser has JavaScript enabled.

Need help? check the requirements page.

Get Flash Player