Reference no: EM133032969

Business Intelligence and Analytics in Major League Baseball 

Early in this century, the Oakland Athletics used readily available traditional player performance statistics in new ways to decide which players to put on the field, and this change led to better play and to several division-winning seasons. Their efforts were memorialized in Michael Lewis's book Moneyball, and in the 2011 movie of the same name.

Major league teams are now all using data analysis to improve player selection, player performance, in-game decision making, and player development. The techniques and tools now in use have moved way beyond what was described in Moneyball. Now, data on every pitch is captured by a doppler radar system that samples the ball position 2,000 times a second. At the same time, the batter's swing is recorded, capturing data about the ball's speed as it comes off the bat and the ball's launch angle. Cameras behind third base record the position of players on the field 30 times a second. A terabyte of data is captured each game. This is now done at all major and minor league parks, in most Division 1 college parks, and even at some high schools.

This wealth of performance data is used as input to analytical software for a variety of purposes. Here are some examples:

• In-game decision making: Teams can see where in the field each batter tends to hit the ball, and they now position fielders accordingly. Therefore, you now often see three infielders to the right (or left, as the case maybe) of second base, or four fielders in the outfield. These untraditional defensive configurations - rarely seen in baseball's 150-year history-look strange to the average fan, but they are very effective in cutting down on base hits.
• Player selection: Teams can acquire players from other teams, or sign players whose contracts with teams have run out. Teams have a rough idea of what pitchers they will face in a year and in what ball parks, which have different dimensions. From the data that is collected each game, a team can simulate how a batter would do against these pitchers in those parks during a full season. In this way, a team can project which players would succeed with them and which might not.
• Improved performance: Doppler radar-generated data shows in detail how each pitch was delivered - the ball's spin, the way the ball was released by the pitcher, the ball's direction and path taken, and other measures. Analysts are now able to show a pitcher how to change their delivery or motion for certain kinds of pitches. By analyzing data about his pitching, Justin Verlander revived his career after being traded to the Houston Astros.

In 2011 the Houston Astros were one of baseball's worst teams. They hired Jeff Luhnow away from the St. Louis Cardinals, one of the early leaders in the use of data analysis, to establish a program for the Astros. In a two-part McKinsey Quarterly interview, Luhnow described this work. Initially, many players were resistant to change, for example to new defensive configurations. But, upper management made it clear to all that the program would continue. A breakthrough occurred when (1) the club showed players how the data was gathered and used, and (2) assigned ex-players with software skills as coaches for the minor league teams to explain the program to players coming up. These moves generated trust and buy-in at all levels. Today, the Astros' program is recognized as one of baseball's best, and the Astros have been one of the most successful teams on the field. Many of Luhnow's staffers have been hired away by other teams.

Luhnow says data analysis in baseball will continue to evolve. In the future, he says, big data and artificial intelligence will be increasingly important. One area of interest is using biometric data to predict, and thus prevent, injuries, particularly to pitchers.

Critical Thinking Questions

1. Baseball executives typically call their analysis programs "analytics." Based on this chapter's BI and Analytics definitions, would you say that their programs are more Business Intelligence or more Analytics? Or, some of both?
2. Excel is a popular and powerful program with a good statistical package. Why do you think baseball teams use tailored software applications for their data analysis, instead of Excel?
3. Baseball teams have used "scouts" to watch young men play at the high school and college levels. Scouts would report the evaluations to the front office, and players were hired based on these reports. Teams still do employ scouts to do this, but increasingly player potential is based on an analysis of doppler and video data. Do you think there will come a day when scouts are no longer needed by major league teams?
4. Most teams have at least a dozen data scientists and other analysts in their programs. Analysts earn high salaries and benefits. Office space, equipment, hardware and software are costly as well. What would you roughly think the data analysis program would cost a major league team each year?