Ben_Slingerland

Sports statistics and box scores have existed for many, many years, but it wasn’t until recently that the perceived importance of “numbers” really exploded when it comes to sports. In 2003, Michael Lew published a book called Moneyball (which would turn into a film in 2011 starring Brad Pitt and Jonah Hill), which highlighted Billy Beane’s “sabermetric” approach with the Oakland Athletics. Beane’s rigorous statistical analysis demonstrated that on-base percentage and slugging percentage were better indicators of offensive success compared to the more popular statistics of stolen bases, runs batted in and batting average. This approach enabled the Oakland A’s to find undervalued players in the market while remaining competitive with teams like the New York Yankees whose salary cap was three times their size.

It is certainly a little surprising that this level of quantitative analysis has not existed to a greater level in the hyper competitive sports business industry. This level of quantitative analysis certainly exists in finance, real estate, medicine … the list goes on and on. Of all those industries, sports is one founded and based on getting that slight competitive edge to be better than your next competitor. Needless to say this heightened level of statistical analysis in baseball has seemingly caused a trickle down approach to other major sports, including soccer.

There is no better place to look at this emerging trend than Major League Soccer. Due to the salary-capped nature of the MLS, a disproportionate amount of competitive advantage can be derived from effective usage of the cap space. Unlike baseball, there are much less statistical categories in the game of soccer (goals, assists, shots, fouls, offsides, corner kicks) and so there is sometimes a perceived notion of less statistical analytical value. However, in a league like the MLS, where teams operate on extremely tight budgets in a vast, global, and competitive talent exchange; knowing what it is you are and what you are not, what you need and what you don’t, and how your limited dollars are best spent, effective analytics can offer concrete advantages. On top of all that, outside of the Designated Player exception, everyone is working under the same salary cap constraints, creating a very even playing field … increasing the incentive to recruit and evaluate current talent.

Because of this recent heightened importance of high-level analytics within the MLS, those few clubs that are using the analytics are really not willing to speak publicly about the specifics of what they’re doing. The industry’s reliance on proprietary data sets restricts worthwhile discussion to a select few, and that’s ultimately unhealthy and stunts overall growth. A lot of teams are re-inventing the wheel behind closed doors and there is no strategy in place for collective growth. This has ultimately resulted in a “cold-start” problem that has hindered club’s incentive to make an investment into analytics. There is no club incumbent to decipher the difference between good and bad analytics and moreover there is no tangible way to measure the return on investment; thus creating a bottleneck of sorts for the industry.

One club that has made a splash in these waters is Toronto FC. Toronto FC has a very young, forward thinking GM, Tim Bezbatchenko who has gone on record as saying “There’s more information available to coaches and General Managers. You need to collect it, organize it, and then look at it and try to figure out patterns and new ways of looking at the game… You don’t know what you don’t know.” Toronto has hired a Director of Analytics, Devin Pleuler, who previously worked for the Opta statistical service and has a strong soccer background writing as a columnist for MLSsoccer.com. Toronto will break down game film looking at trends in pattern of play while scouting their upcoming opponents. By processing games algorithmically, season trends are uncovered and it allows for game film to be watched with a specific focus. In addition to scouting opponents, this level of analytics also helps Toronto better understand their own strengths and weaknesses, which leads to useful information regarding player acquisition.

The San Jose Earthquakes, which coincidently share an owner with the Oakland Athletics, said they use a variety of services to provide data. The club has contracted with Wyscout for international scouting services, Match Analysis for in-game analytics, and Catapult to monitor players for overuse at practices. Match Analysis has a league-wide deal with the MLS and is an “xy coordinate system” that has become popular with many college programs within the United States. Catapult is a heart-monitoring device that is easily worn by players in practice, and is something that many teams throughout the league implement. However, they have not gone in greater detail about their specific use of these programs.

Two services that some teams use that San Jose didn’t mention are Pro Zone and Opta. Pro Zone, like Match Analysis, also has a MLS league wide deal. It’s a system that is used by over 70 college teams and is based in Leeds, England. They will turn a match back within 36 hours that has over 4,000 data points (touches, tackles, headers, etc). From these data points, coaching staffs are able to break down passing completion and accuracy as well as shots and shots on goal. Even more specifically, they can see where on the field certain players have success and where on the field they might have a high turnover rate. Opta, is a manually tracked system that is done live, giving it more practical use in the course of a match whether it be on a live broadcast or if it’s produced for a coaching staff going into the halftime locker room.

Peter Vermes, the Head Coach of Sporting KC, is a big believer in analytical data. However, Vermes admits that soccer is unlike baseball in that there are so many variables that go on within a game (from pitch size and condition, to formation, to a team’s style of play). Thus, Vermes argues, you have to use both the statistical analytical data along with the actual game film to pick up a greater understanding of player trends. Sporting will rank each of its eleven players comparatively to the rest of the 19 teams in the league through a set of statistical categories. In general Vermes points out, if Sporting has at least 6 of their players in the top 10 (half) of the league they are successful.

A former Los Angeles math teacher, Tim Crawford, runs the New England Revolution’s analytics and they take more of a “best practices” approach. What I mean by that is the Revolution will try to understand what is successful in the MLS, what works. From there, they will take those data points to evaluate their own style of play and then take it one step further to use those higher valued statistical data points to scout players. The Revolution take a more academic sabermetric approach where they come up with a thesis and try to find a proof for it.

The last club in the MLS that is known to use a fairly large amount of data specifically around injury prediction and prevention are the Seattle Sounders. While the club has become the model organization around the league from a business perspective, it should come to no surprise that they are out in front on the analytics side as well. Dave Tenney, the Sounders sports science and performance manager, uses the Catapult gps tracking heart-rate monitors in training to see how hard each player is working during certain sessions of training.

Meanwhile, the Sounders use Ravi Ramineni, the team’s performance analyst and former Microsoft employee, to automate and streamline the data. Tracking software allows the technical staff to measure how often a player reaches top speed during a given practice, to design drills that best simulate game action. Players that are pushing themselves too close to the limit get dialed back, while teammates that are loafing are put through extra paces the following day. Sounders Head Coach Sigi Schmid sums his viewpoint on this data:

Analytics is just part of the identification package. I don’t think anybody is at the stage where you’re going to make that your sole identifying source or the sole determinant of your decision, but it certainly factors in. Does the objective data support what you’re seeing?’ I’ve never been a person that just looks at the data and goes, ‘OK, that makes them a good or a bad player… You’re looking at whether the objective data supports what you’re seeing subjectively.

Schmid’s words capture what seems to be a shared sentiment amongst the league. That is, “we see the value in collecting and reviewing data, but we are not completely confident in relying on the data standing alone.”

While teams like Manchester City in the English Premier League have upwards of a dozen analytics staff, MLS teams are lucky to have one. Those few MLS teams that are investing into the analytics field have remained very secretive about their processes. The verdict is still out as to who is really capitalizing on their investment into analytics. And so it seems that the growing trend towards capturing data and finding meaning to it all is very quickly coming to a tipping point in Major League Soccer.