Sports Analytics

 

By Joshua Thompson

Fresh from winning the Stanley Cup championship, the Chicago Blackhawks were among the top ten most valuable teams in the NHL in 2010. Forbes pinned their franchise value at $300 million, and to sports fans and industry experts the Blackhawks were undeniably the premier organization in professional ice hockey. Nevertheless, their worth still lagged behind that of the other “Original Six” NHL enterprises—the Boston Bruins, Montreal Canadiens, and others—by hundreds of millions of dollars.

Five years later, after two more Stanley Cup championships and consistently deep playoff runs, the star-studded Blackhawks are worth a whole $925 million, with their 12% annual growth rate ranking near the top of the league and well ahead of other less-winning members of the Original Six. Success on the ice, unsurprisingly, directly correlates to success in business. Winning makes money, and John Chayka thinks he has a new way of doing it.

Chayka, now 27 years old, became the youngest General Manager in the history of professional sports last year when the Arizona Coyotes promoted him from Assistant GM. Chayka began his career in hockey operations at just 19, when he realized that analyzing game film for nuanced patterns and complex events revealed a number of statistical insights that could improve in game performance. Soon thereafter, he created Stathletes with a friend with a math and science background. An ironic blend of science and sport, of hardened analysis and sporadic passion, of statistical order and of chaos, the company attempts to bring analytics to the world of hockey. Unfortunately, science and the sport have had a tenuous relationship. Many hockey higher-ups are still skeptical of the role of analytics in their organizational scheme. Nevertheless, in the years since Billy Beane’s success in the MLB, as chronicled in book and film Moneyball, a number of teams have begun to implement statistical modeling into their systems, and Chayka’s unprecedented rise can attest to that.

Just as in baseball, the hockey analytics movement started with the consumer. The internet age provided super fans with a wealth of data to muse on, and online forums such as, “The Irreverent Oilers Fans,” quickly began to develop analytical tools for interpreting the on-ice chaos of elite-level ice hockey. Soon thereafter, systems like CORSI and Zone Starts began to take hold in the NHL, and teams have taken their insights and adjusted their strategies accordingly. These adjustments appear to be directly correlated to an organization’s performance; the rise of, “puckalytics,” is often reflected in the success of teams such as the Blackhawks and the Los Angeles Kings, who are heralded as teams at the forefront of the analytics movement. In light of this information, the obvious question is why isn’t everyone doing it?

There are two possible answers—both rely on faith. The first is all business. Analytics is complicated, and to do it right requires a number of key moves on the part of team managers. Acquiring the right professionals, integrating proper technology, and delving into new research all implies the allocation of significant resources. Teams are careful not to commit to extensive spending so long as their faith in data is tepid, and the evidence is not overly convincing yet. For example, ESPN says that organizations like the Buffalo Sabres as well as the Boston Bruins have poured money into developing analytics systems, yet both organizations have seen relatively little real success in the past few years. Although, teams like the Kings and the Blackhawks play the motley.

The second answer rests on a belief in the essence of sport. Chaotic and harmonious, at times eloquent and at times awkward, passionate and stoic, sports are an age old expression of human imperfection and unpredictability, and the information age seeks to reduce the beauty of the game to something unnatural--to predictable outcomes, and to perfection. This romantic perception of analytics isn’t new to hockey. Rather, it has colored the relationship between science and human sports since the integration of computers into chess, since Bill James’ Baseball Abstract, and since the implementation of data science into the world of basketball just a few years ago.

Yet, this skepticism is not merely the chorus of those unwilling to adapt to the changing nature of sports. Ironically, backlash against science in sports may even have some science to back it up.

Exciting offense and scoring points, a lot of points, is perhaps the best indicator of the overall entertainment value of a game. While super-fans and connoisseurs of a sport may have a preference for defensive systems and tight, well executed strategies, the lay consumer’s tastes usually tend towards goals, points, touchdowns, or whatever other values associated with a dynamic, high paced offensive clinic may be. There is some evidence that suggests that analytics may be correlated to a decrease in offensive output in certain sports. For example, since the introduction of analytics into professional ice hockey in 2005, average goals per game in the NHL has been on a steady decline from 3.08 goals per game to just 2.7, despite rule changes that sought to encourage goal-scoring. This criticism isn’t only pertinent in hockey. Basketball, too, has shown certain trends towards a less-spontaneous, more run-of-the-mill game style. As Braedon Clark of Raptor’s HQ says, “the plays that make watching the NBA so fun—an impossible fade away jumper, a driving layup around three defenders—are in many cases the same plays analytics tell us should be avoided.” Clark isn’t alone in his criticism, in fact, many, “experts,” denounce the effects of analytics on the essence of basketball. Former professional Charles Barkley hasn’t held back on analytics, claiming that data analysis is making the game, “boring.” However, if there is a real connection between analytics and less offense, it doesn’t seem to be showing in either league’s revenue stream—the NHL has gone from pulling in $2.27 billion to just over $3.7 in the last decade, and the NBA has shown as much as 20% growth in certain years.

Analytics has also opened up an entire new world to potential sports consumers. Hockey, as well as other sports, is no longer confined to the realm of athletics alone. Mathematics and science give sport an intellectual application that can be supplemented throughout the game to give fans a deeper understanding of what’s going on at base level.

Additionally, there is evidence to prove that data analysis may actually make sports more dynamic and entertaining. Despite a steady decline in goal scoring, recent NHL seasons have shown that offense may actually be on the rise, and the growth of analytics expected to come with their integration into professional hockey is likely to coincide developments in offensive strategies and systems that are favorable to a fast, entertaining brand of hockey. Furthermore, experts in sports analytics like Dr. Philip Maymin, co-founder and co-editor of the Journal of Sports Analytics, believe that analytics may actually make sports more entertaining. In a recent interview, Maymin said, “If the results from analytics are more boring games, then rule changes will and should be forthcoming. But it seems as if the analytics approach tends to create better games: high-scoring, fast-paced games with lots of threes, lots of dunks, and lots of passing; in short, lots and lots of highlights.”

Agree with Maymin or not, it’s important to remember that the goal of competitive sports is not to simply make money, but to win, and people like John Chayka are explicitly hired for that purpose. Nevertheless, economic success is not mutually exclusive to organizational success, instead, the two are inextricably linked. Like them or not, analytics are helping organizations like the Chicago Blackhawks win games, and their fanbase, which had the highest average ticket sales and game attendance in the league last year, is perfectly entertained.