The simple lessons you can still learn from sports analytics
MIT Sloan’s Ben Shields explains how franchises use data to make personnel, business decisions.
By Brian Eastwood |
February 9, 2017
“You drive change by integrating analytics on a daily basis,” said MIT Sloan's Ben Shields.
Very few industries are blind to the need for business analytics and data scientists. Job growth in the field shows no sign of slowing.
But hiring is only step one. For success, it’s helpful for managers to look back at founding lessons from the field of sports analytics, where organizations use data to make critical personnel and business decisions.
“We can view sports as a petri dish for how organizations can maximize the value of analytics,” said Ben Shields, a lecturer in managerial communication at MIT Sloan. The ways that sports franchises use analytics, he added, “reinforces the idea that analytics is a tool to make better decisions in a variety of disciplines within an organization.”
Business leaders may be skeptical of drawing lessons from a unique industry with multimillion-dollar salaries and months-long offseasons. However, sports franchises offer many case studies for taking a strategic approach to building an analytics program, said Shields, who has written three books, most recently “Social Media Management: Persuasion in Networked Culture,” on social media and marketing in sports, media, entertainment, and more broadly.
“Any strategy should begin with the business goals a firm or leader is trying to achieve,” he said. In sports — and in business — these goals are straightforward: Put together a winning team on the personnel side, and operate more efficiently and increase revenue on the business side.
Don’t just collect data. Use it. Every day.
The key to success, Shields said, is achieving the culture change that embraces analytics-based decision-making. The Oakland Athletics of the early 2000s faced criticism for using mathematical formulas instead of “old-school” scouting reports to rate players, but other franchises caught on, and now professional, college, and even high-school teams use “new-school” analytics to evaluate performance and make personnel decisions.
“Interestingly, one of the barriers to unlocking the full potential of analytics is organizational. A company may make an investment in analytics technology and data scientists, but if its analytics team operates in in its own silo, it can be difficult for the company to truly become data-driven in its decision making,” Shields said. “You drive change by integrating and using analytics on a daily basis.”
Today’s NBA teams illustrate this point. In addition to using statistics to decide which players to add or drop from a roster, professional basketball teams apply analytics to player health, using wearable technology and eye-in-the-sky cameras to identify signs of fatigue, and give players time to rest. Many teams have also implemented dynamic ticket pricing, with game-day prices based on an analysis of supply and demand as opposed to a standard seat price set at the beginning of the season.
Communicate effectively and know the limitations of analytics.
It’s also important to communicate the meaning of data analysis throughout an organization for maximum impact, Shields said. Flashy data visualizations are ineffective if they lack the proper context for the stakeholders using data to make decisions, be they players, coaches, personnel managers, or owners.
For all the benefits that analytics can bring to an organization, it’s important to remember that analytics is not a panacea, Shields said. Humans still have a role to play in decision-making — and the qualitative data and gut instinct from those old-school scouting reports can be valuable. If Player A scores more points, but Player B is a better cultural fit for a team, then a team shouldn’t just automatically choose Player A.
“Even the most ardent analytics supports should be up front about the limitations of analytics. That’s a key part of successful implementation,” Shields said. “You must maintain a measured mindset of what analytics can and can’t provide.”