Sergey Nivens - Fotolia
The 2013 Major League Soccer season ended for the Seattle Sounders in the Western Conference semifinals round of the playoffs. Not a bad showing, but David Tenney, sports science and performance manager, knew the team could go further.
He set about analyzing data collected throughout the season. He looked at how hard players worked going back to the preseason, how much distance they covered during practices and games, and how hard they exerted their bodies. He found a correlation between missing portions of the preseason and injuries. This made it possible to develop a training program that better prepared players for the rigors of the 34-game season.
So far, the effort is paying off. With 14 games remaining in the 2014 season, the Sounders are at the top of the Western Conference standings and among the league leaders in goals scored. Tenney attributes part of this improvement to better conditioning and fewer injuries, benefits that were made possible by in-depth data analysis.
"That's one of the reasons I feel like we're a fitter team and one of the reasons why we feel like we're able to sustain higher work rates toward the end of games," he said.
The Sounders collect mountains of data on their players. During practices, players wear a body monitor made by Catapult Sports. This device collects information on accelerations, decelerations and distance run. Twice a week the team puts players through electrocardiogram scans performed by a system from Omega Wave to measure their level of bodily strain. All this data goes into a SQL Server database, where Tenney and his team use a visualization tool from Tableau to analyze it.
But the data collection and analysis operation at the Sounders wasn't always this mature. When Tenney joined the team in 2009 there was little infrastructure in place for making productive use of data.
"At the time there was a little bit of data, and we put stuff in Excel spreadsheets and stored things on external hard drives," he said. But after his first year, the team began generating more biometric data, and that was when Tenney got serious about managing and analyzing it.
In addition to bringing in more technology tools to handle all the data, he started building a team that could help the organization become more data-driven. He hired Ravi Ramineni, a long-time manager at Microsoft, to be the lead data analyst. He also brought in Chad Kolarick, who holds a degree in computer science, mathematics and economics, as his head strength and conditioning coach.
David Tenneysports science and performance manager, Seattle Sounders
Tenney may not have the resume of someone you would think would be leading this kind of analytically driven team. His background is entirely in coaching soccer and fitness, not technology or data analysis. But as his career has progressed, he's seen teams collecting more data on players and trying to base more decisions on data analyses. In addition to learning the Tableau visualization system, he is learning to manipulate data using the R programming language through online courses posted to the site Coursera. To stay relevant, he had to learn the technology.
"I realized we were getting a lot of data and we've got to figure out what we're going to do with this data, and I've got to be better at handling it and making sense out of it," he said.
Not every professional sports team is heading in this direction. At the MIT Sports Analytics Conference held in Boston in March, former NBA head coach Stan Van Gundy said he never felt confident his staff were recording data properly. This data quality issue made him skeptical about incorporating data analysis into his decision-making process.
But Tenney said it is possible to at least measure player fitness and make objective recommendations. For example, the Sounders have been collecting data on all-star midfielder Osvaldo Alonso for all five years he's been with the team. This depth of data makes it easy to see what his level of conditioning was just before past injuries. If he gets to that same point, Tenney feels comfortable recommending changes to the coaches.
"With the data, we can say, 'Listen, this isn't my opinion. We know when we see this type of data from this athlete that this is a likely outcome.' The goal from our analytics is we can create enough data so that it's not my opinion against the coach's," he said.
And as teams like the Sounders that have embraced analytics continue seeing results, more are looking to jump on the bandwagon. Tenney said he has received inquiries from teams across different sports and around the world to learn more about how he has managed to use analytics successfully. It's a sign that sports, one of the most human, and therefore most chance-prone of endeavors is ready to be swept by the analytics revolution that has changed other industries.
"If you would have asked me in 2007 if I would be doing what I'm doing today, I would have thought you were crazy," Tenney said. "But I do think this is part of where pro sports is going."
Hear why analytics in sports is a hot topic
See what CIOs can learn from sports teams
Learn why Germany tied its World Cup hopes to analytics
- Data Visualization Techniques: From Basics to Big Data with SAS Visual Analytics –SAS
- Data Visualization in F1 Racing –TIBCO
- Visual Data Mining and Machine Learning –SAS
- Good Enough to Great: A Quick Guide for Better Data Visualizations –Tableau Software