has insight into the power of sports and grasps the pulse of the industry! Big Sports Business Issue 712, welcome to pay attention to the best sports industry information platform.
Editor's note: On April 25 (next Monday), the medieval epic fantasy-themed American drama "Game of Thrones" will officially return for its sixth season, but what we are talking about today is "the game of numbers, the iceberg in the field of North American sports data analysis" A Song with Fire.”
Interested readers should still remember that in mid-March, we launched an issue of "What trends will the 2016 North American MIT Sports Analysis Summit bring to the sports industry" brought by Peng Minggao from the Ivy League Cornell University in the United States, which briefly introduced The annual MIT Sloan Sports Analytics Conference (MIT SSAC) and the 2016 Sports Analytics Summit focus on the topics and trends discussed.
The sports analysis monograph "The Numbers Game: Why Everything You Know About Football Is Wrong" translated by Peng Minggao will soon be officially launched in China. This book Combined with many specific cases in the Premier League, Champions League and even the NBA, it completely subverts some of our common sense about the data of professional sports events.
Let me give you a spoiler by the way. In view of the growing popularity of sports big data in the domestic sports industry, Sports Big Business will recently invite David Salley, the author of "The Numbers Game" and a professor at Ivy League Dartmouth University in the United States. David Sally came to China to hold a forum on sports big data to share and communicate with experts, scholars, and practitioners in the domestic sports data field. Interested friends please pay close attention to our relevant event push, or get in touch with us (customer service WeChat: tiyudashengyi).
In fact, another Chinese student appeared at the summit - Continuing from the Statistics Department of Boston University. He covered the entire summit on behalf of Hupu Sports. Continuous has a very professional understanding of statistics and sports analysis. We also asked him to share the development of sports data analysis in North American professional sports events, including bottlenecks and controversies in NBA, MLB, NFL and other events. We hope that It can provide some inspiration for the field of sports data analysis in the early stages of the development of my country's sports industry.
The author of this article: Continuous
Sports Big Business reporter in the United States and Boston University statistics student
In 2003, the best-selling American author Michael Lewis published "Moneyball: The Art of Winning an Unfair Game". The book tells the story of Billy Beane, the general manager of the MLB team Oakland Athletics, who used analytics to build a low-cost, small-market team into a league powerhouse.
The Oakland Athletics' total player spending in the 2002 season was $40 million, while the New York Yankees' total spending was $126 million. However, the Athletics had the same regular season record as the "Evil Empire" that season (103 wins and 59 losses). This story, which essentially means "spend one-third of the money to do the same thing, and use a hand full of phone numbers to achieve a 'king bomb' effect" was immediately popular in the United States after it was published. Precisely because of the popularity of "Moneyball", sports data analysis has received unprecedented attention from the public.
The development history and current situation of sports data analysis in North America
. In 1944, the MLB Brooklyn Dodgers hired data analysts
Although "Moneyball" has achieved great success, it has still been subject to a lot of controversy. The history of sports analytics is much richer than author Lewis lets on.
For example, the decisive statistic on-base percentage that created the Oakland Athletics' productive offensive performance was recognized as early as 1944. Branch Rich, the general manager of the MLB's Brooklyn Dodgers, hired Alan Ross as the team's data analyst that year, and later published "Goodbye to Old Thoughts" in the famous "LIFE Magazine" Some Old Baseball Ideas" shares the story of Rose's discovery of on-base percentage data.
▼ Blanche Ridge is an outstanding contributor to baseball.He once fought against all odds to sign Jackie Robinson, the first black player in MLB history, breaking racial boundaries. (Picture: LIFE Magazine)
. The Kraft family uses mathematical models to manage the salary cap
Some people also question that Billy Beane and his Oakland Athletics were only promoted to the altar because of the book "Moneyball" . There were pioneers before them, such as the New England Patriots: In 1994, the NFL officially introduced the "salary cap" system. The Kraft family, who took over the Patriots in the same year, realized the correlation between salary cap management and team performance. Jonathan Kraft, who was the team's vice president at the time and now the team's chairman, teamed up with colleagues in the financial world and began to use data models to manage the team's salary distribution and structure. Smart salary cap management also allows this Northeastern club to make free decisions on player personnel and maintain long-term competitiveness in the NFL.
. "Moneyball" opened the door for people with a statistical background to enter professional league management
However, the book "Moneyball" did successfully spread the concept of bringing data analysis into personnel decision-making and changed the entire The face of MLB and professional sports in North America as a whole. Prior to the publication of the book, a team's decision to hire someone with a purely statistical background into the front office would have been unable to escape criticism from the media and fan base if it had been publicly reported. But in the past decade or so, sports data analysis has grown into a "necessity" to improve team competitiveness.
In mid-December 2012, after the new owner completed the acquisition of the team, the NBA Memphis Grizzlies hired ESPN basketball mathematician John Hollinger as the team's vice president of basketball affairs. In ESPN's farewell article, Hollinger wrote: "It's hard to believe that this has become true. But eight years ago, few teams were interested in data analysis, and those who were interested were unwilling to disclose it. Admit it...but in less than 10 years, many teams have changed their thinking: Now, many teams are even showing off their investment in data analysis for at least two-thirds of the league. Teams have invested in this, and although some teams are only showing off in form, most are still very involved. "
It is not an exaggeration to say that two-thirds of NBA teams have begun to invest in data analysis. According to a recent statistical report by ESPN, as many as 22 teams in the entire MLB have invested in data analysis, while only 2 teams do not believe in data analysis at all. In the NBA, only three teams, the Brooklyn Nets, New York Knicks and Los Angeles Lakers, do not believe in data analysis at all.
. The MIT Sports Analysis Summit came into being
It is precisely because "Moneyball" has changed the public's thinking about sports data analysis that many "outsiders" with a background in data statistics can enter the circle with dignity and shine. This also gave birth to the formation of an ecosystem. Creating a communication platform for sharing and discussion has also become inevitable. After
jumped to the NBA's Houston Rockets on April 3, 2006, Daryl Morey, who originally taught sports business courses at MIT, had no time to travel back and forth between Boston and Houston, so he chose to transfer his courses to transformation. The MIT Sloan Sports Analytics Conference (hereinafter referred to as the MIT Sports Analytics Summit) came into being and has gradually grown into a standard setter and leader in the field of sports data analysis over the past ten years. , further catalyzing the development of sports data analysis.
▼ Daryl Morey, general manager of the Houston Rockets and chairman of the MIT Sports Analytics Summit (Picture: Thomas Shea)
5. The Celtics are the first NBA team to delve into big data about fans
Nowadays, sports data analysis The scope is no longer limited to the team's performance on the field. It has permeated every aspect of North American professional sports. More and more teams and organizations realize that in-depth data analysis in the commercial field can enhance their profitability.
In January 2003, a consortium headed by Vic Grosbeck and Steve Pagliuca acquired the NBA Boston Celtics.Grosbeck, who has a background in venture capital, believes that in addition to using data analysis in basketball affairs, the team must also be practical in business affairs. It was under Grosbeck's leadership that the Celtics became the first team in the NBA to conduct in-depth information mining in terms of fan experience. This thinking model of the Celtics will also be learned and used by other teams and the NFL in the future.
At this year's MIT Sports Analytics Summit, Grosbeck said: "On the business side, we began to better understand our fans through data analysis. We modified the entire membership system. You are no longer a Celtic. Season ticket holders. You are a part of our team. We want to get to know you as much as possible and then give back to you as quickly as possible. Technology allows us to appreciate the fans. It’s time to thank the fans.”
6. The blogosphere and the flexible application of high-level data
As the Internet flourishes in the 21st century, the emergence of the blogosphere has greatly lowered the threshold for outsiders in the sports circle to enter the media industry. This has also resulted in the emergence of a large number of excellent sports data analysis authors in the media industry in the past decade or so. Articles full of informative articles and viewpoints different from those of traditional reporters force mainstream sports media to rethink.
Former ESPN chief writer and Grantland founder Bill Simmons, who emerged from the blogosphere, once said frankly that nutritious data analysis can enhance the quality of his works and satisfy his audience. Zach Lowe, Bill Barnwell, Jonah Kerry, Kirk Galsberry and others he recruited for Grantland were all excellent writers who were good at using numbers to illustrate their opinions. Today, high-level data vocabulary appears frequently in daily programming on various sports television networks. Media like ESPN have also begun to invest more in high-end data research and development to better serve fans.
The bottleneck and breakthrough of sports data analysis in the NBA
. From the TENDEX equation to player efficiency value
In recent years, North American professional sports have made frequent moves to develop the Chinese market. MLB has reached a full ecological strategic cooperation with LeTV Sports and plans to broadcast more than 120 events every year on LeTV's various platforms. The NFL is also actively exploring the possibility of holding regular season games in China.
But there is no doubt that the NBA is the most popular North American professional sports event among Chinese people. As the league with the earliest launch and the widest audience, the NBA’s business expansion in Greater China has been very mature. Houston Rockets CEO Ted Brown lamented at this year's MIT Sports Analytics Summit: "We broadcast a game in China, for example, the Rockets play the Celtics. The number of viewers watching this game live in China is greater than that in our local area. Taiwan has more in one season.” Therefore, basketball sports data analysis is also the first thing that Chinese sports fans are exposed to.
's early basketball data analysis was in the same direction as baseball, both creating high-level data based on primary data transformation. David Silen is a pioneer in this field. The TENDEX equation he created in 1958 (until it was published in 1989) was the first linear equation in basketball history used to evaluate a player's individual level (Silen's TENDEX efficiency value = [points + rebounds + assists + steals + blocks – Shots missed – 0.5 * Free throws missed – Turnovers – Fouls] / minutes).
Since then, many people have created various versions of linear models based on the TENDEX equation, and used them to evaluate players' personal efficiency. The most famous of these is the Player Efficiency Rating (PER) created by Hollinger, which is now widely used.
▼Houston Rockets general manager Daryl Morey and Memphis Grizzlies vice president of basketball affairs John Hollinger are two major representatives of sports data analysis in the basketball world. (Picture: NBC Sports)
. Why is it said that basketball data analysis is still in its "infancy"
However, the current mainstream view in North America is still that basketball and other sports are still in their "baby steps" in terms of data analysis compared to baseball. "status.
The reason is actually not difficult to understand. Baseball data consists of three parts: pitching data, batting data and fielding data.The preliminary data of the first two items have the characteristics of "black and white" and are relatively easy to collect (for example, "the batter can only be successful or unsuccessful in getting on base", or "the pitcher can only solve or not solve the batter"). Therefore, these two data are easier to develop in terms of high-level data than the fielding data involving personnel movement.
And basketball is a very "dynamic" sport. Its preliminary data does not reflect the player's ability well. Hollinger himself admits that rudimentary statistics like blocks and steals can by themselves mislead people about the defense itself. This also leads to the fact that his PER rating is not a trustworthy assessment tool when measuring a player's personal defensive performance. Players like Bruce Bowen, who is considered by the world to be the top perimeter defender, has always had a PER in the single digits (the league average PER is 15).
At the same time, high-level data similar to PER cannot bring much nutrition to the coaching staff. Although more "overall" data such as "plus-minus" have been officially introduced by the NBA in recent years, it cannot well explain the role of individual players. Therefore, the limited value of data for on-the-spot coaching has also become a major bottleneck in the research and development of basketball data analysis.
. The SportVU player tracking system allows analysts to have a language to communicate with coaches.
However, the emergence of the player tracking system solves this problem.
In 2005, a group of Israeli scientists with a background in missile tracking invented the SportVU player tracking system. Its essence is a camera system that is placed above the stadium and captures all activities on the basketball court 25 times per second. The system can provide real-time two-dimensional movement data of players and three-dimensional movement data of basketballs.
▼ The SportVU system installed above the United Center (Photo: Digital Trends)
Since 2013, SportVU has been installed in all 29 NBA arenas and has recorded more than 3,000 NBA games. The massive data provided by this system provides a strong foundation for each team's data analysis.
Before the birth of SportVU, if a team wanted to know how its players performed in pick-and-roll defense, the only way they could collect information was to manually watch the video. Such work is time-consuming, laborious, and inefficient. Now, they can download tens of thousands of pick-and-roll defense data from SportVU, use machine learning to identify and determine the type of each pick-and-roll defense, and analyze and summarize each pick-and-roll defense category one by one after the identification work is completed. . The
SportVU system allows analysts of each team to interpret all aspects of the field through coordinate data, such as players' shooting hotspots or their running distance on the field. Information such as this is undoubtedly valuable, because it is simple and easy-to-understand basketball language and can bring more intuitive understanding to coaches, trainers and players. Therefore, SportVU has also been listed as the most important innovative technology in the sports industry in the past ten years by the MIT Sports Summit.
Controversies about sports analytics
There have always been various controversies over the role of sports data analytics in the sports industry. Many people worry that professional sports will have negative effects after embracing data analysis and innovation. They are resistant to analysts taking over the team's front office and question whether data can maintain the team's long-term success. They even worry that data analysis will eventually take away from the charm of the sports themselves and affect their popularity.
Of course, it is irresponsible to respond to these controversies solely with the argument that "innovation is positive". These questioning voices are worth exploring. In the book "Diffusion of Innovations", sociologist Everett Rogers summarizes the negative consequences of innovation as "indirect", "unwelcome" and "unexpected". "External" three categories. We can also analyze these doubts one by one from these three perspectives.
1, Are traditional scouts and analysts incompatible?
The reason why many insiders are frightened by the data storm is because they worry that the introduction of data analysis will "indirectly" reduce the value of traditional insiders (such as scouts) in the team.TNT celebrity and NBA Hall of Fame player Charles Barkley once criticized Morey on the show: "The NBA is a league where talent speaks for itself. Now the people who run the team talk about data analysis all day long, but They are essentially a group of people who have never played football. "
But the reality is that, at least in North America, the conflict between traditional scouts and analysts is not so fierce, especially among those good teams. The incompatibility between the two sides is mostly just the obscenity of some media. A bitter disagreement between analysts and traditionalists over the script for the "Moneyball" movie led Oakland Athletics assistant general manager Paul DePosta to ask the film team not to use his name in the film. De Posta believes that the opinions of both parties are basically the same. In the end, the film team transformed DeBosta's character into a fictional character named "Peter Brand" and gave it to Jonah Hill, who has a very different body shape from DeBosta.
In the past ten years, the mutual assimilation between the two has become a mainstream trend. As Boston Celtics assistant general manager Mike Zaren said, the word "analytics" no longer appears in management's daily communication. Nowadays, those who study statistical models will be required by teams to inspect players on site, and traditional scouts and coaches are equipped with tablet devices loaded with data information. Traditionalists still play an important role in team decision-making.
▼Hollywood star Jonah Hill played the role of Peter Brand in the movie version of "Moneyball" and was nominated for the 2012 Oscar for Best Supporting Actor. The prototype is former Oakland Athletics general manager Paul DePosta. (Picture: "Moneyball" movie stills)
. Information privatization has become a mainstream trend
However, the popularity of sports data analysis will indeed bring "unwelcome" consequences to some teams. The Oakland Athletics once became the "first to eat crab" because they spent a small amount of money to buy players with high on-base percentages. However, after the publication of "Moneyball", teams began to follow the practices of athletes, which also caused the value of players with high on-base percentages to rise steadily, causing athletes to lose their own advantages. This also forced Billy Beane's team to find the next "on-base percentage" that would allow them to succeed, but it is understandably difficult.
Therefore, "information privatization" has become a mainstream trend in North American professional sports. Most MLB teams have built their own unique databases or outsourced this task to third parties. Now, if you want to know the "secret" in his team's hands, the only way is to hack.
On June 16, 2015, the New York Times revealed that the FBI had launched an investigation into the MLB team's St. Louis Cardinals' hacking of the Houston Astros' information database. The Astros' current general manager, Jeff Luhnow, once served in the Cardinals' front office.
. Is the loss of young baseball fans related to data analysis?
Of course, whether technological innovation will make the sport itself "boring" is the issue that people are most concerned about. The issue is particularly heated within baseball. In the past five years, many MLB teams have reached huge broadcast contracts with local TV stations, demonstrating their strong ability to attract money. However, according to an ESPN survey, baseball's popularity among American teenagers has been declining year by year, and has even reached the same level as MLS. This also makes the future of baseball worrying.
Many, including John Thorne, MLB's official historian, blame analytics for baseball's loss of young fans. The main argument is that baseball is the sport most exploited by data. But this statement is open to question. Although data analysis can bring a lot of information that cannot be detected by the naked eye, it cannot exhaust the sport after all. In addition, the slow pace of baseball itself (MLB has actively begun to improve this situation), and the rise of football itself are obviously more important factors.
At least, MLB’s official attitude towards data analysis is still positive. Since the 2015 season, MLB has included tracking data elements in broadcasts on various platforms.The speed, angle and direction of each ball after it is hit will be presented to the audience in real time, providing a better viewing experience.
▼ In 2014, MLBAM announced at the 8th MIT Sports Analytics Summit that it would enable tracking data elements in broadcasts. (Picture: MLB official website)
Conclusion:
After evaluating the above objections one by one, it is not difficult to conclude that the negative impact of data analysis on professional sports is limited: the relationship between traditionalists and analysts is not contradictory. . As long as the team privatizes information, it is possible to remain competitive in the long term. Nor is the introduction of data analytics the determining factor in the decline in popularity of individual sports.
For now, data analysis is not destroying the competitive environment of North American professional sports, but further enhancing their competitiveness. As NBA Dallas Mavericks owner Mark Cuban believes, the motivation for professional teams to take data analysis into consideration is actually clear and straightforward: “I just want to do everything I can to give the Mavericks as many advantages as possible. ”
WeChat public account: sportsmoney
Industry information | Sports marketing | Sports venture capital
The best sports industry information platform
Contact us Welcome to submit news or all kinds of cooperation
Customer service WeChat: tiyudashengyi
Submission email: [email protected]
www. sportsmoney. cn
