Men occasionally stumble across the truth, but most of them pick themselves up and hurry off as if nothing had happened. – Winston Churchill
Remember when your high school coaches relied only on their experience and instincts to evaluate teams and players? Well, times are a-changing! Business intelligence has come to high school sports!The Wall Street Journal
ran an article on March 21, 2012 titled “Game Film Goes to School: How One Stathead Is Carving Out a New Market – High School Coaches.”
The article reported that a high school coach had a series of charts and diagrams that plotted almost everything that was plottable about his basketball team and their opponents on his iPad. This included shot locations, scoring pace, the offensive and defensive potency of every five-man unit they'd put on the floor during the season and how effective their star players had been when they'd received the ball at any spot on the court.
Amazingly, this high school coach was but one of 475 high school coaches across the country that have paid a New York-based company named Krossover
as much as $2,000 a year to take their digital game film and break it down into some of the most advanced metrics that youth basketball has ever seen.
Vasa Kulkarni, the founder of Krossover, realized that in an era of cheap, high-quality digital video cameras, obsessive parents and coaches would be filming virtually every sporting event already. So for the maximum rate of $2,000 per season, Krossover will scrutinize these films and send back analyzed game film within 24 hours, employing a program that helps his firm’s human analysts quickly break down game film.
When Krossover returns the report, often the morning after a game, the coaches can click on a player's number and view video of every play he was involved in, as well as intimate measures of his performance that include a breakdown of where his assists were received: close to the rim, in the paint or out near the three-point line, for instance. For each unit of players, there is something called the pace of play, which can estimate how many possessions they'll have per game. Another statistic Krossover compiles is effective field-goal percentage (or eFG%), a weighted formula that adjusts for the difficulty of the shots taken.
Basketball isn’t the only sport that Krossover can analyze. This year they have unveiled a video-based analytical system for men and women's lacrosse that's getting some traction with college programs. Services for football will be available in summer 2012 and they have announced on their website that soon they will support volleyball and hockey as well.
Krossover also provides services for athletes. On their website
, they say that players using their services can:
- Create their own highlight reels of themselves, their teammates or both.
- Share their highlight reels via Facebook and Twitter... or email them to their friends and families.
- Improve their chances of getting recruited by sending out official game footage to recruiters.
- Get their stats for each game and see how they performed over the entire season.
Some may think the analytical service that Krossover is offering to high schools is not a good idea. They may make arguments such as the following:
Isn’t this Moneyball comes to high school? Moneyball
Doesn’t this send the wrong message – that winning is everything?
What about sports being about competition and fun?
Can you really build a team by focusing on data?
What about the crucial intangibles like heart and passion?
Where do instinct, ingenuity and guts appear in stat tables?
is a reference to a recent movie starring Brad Pitt. The central premise of Moneyball
is that the collected wisdom of baseball insiders (including players, managers, coaches, scouts, and the front office) over the past century was subjective and often flawed. So Oakland Athletics baseball manager Billy Beane (played by Pitt) altered the sport by employing data and analytics to compare and evaluate the success of player skills. His rigorous statistical analysis demonstrated that on-base percentage and slugging percentage were better indicators of offensive success, and the A's became convinced that these qualities were cheaper to obtain on the open market than more historically valued qualities such as speed and contact. Beane’s approach was seen as controversial because it flew in the face of conventional baseball wisdom and the beliefs of many baseball scouts and executives. However, Beane’s approach worked. By re-evaluating the strategies that produce wins on the field, the 2002 Oakland A’s, with approximately $41 million in salary, were competitive with larger-market teams such as the New York Yankees, who spent over $125 million in payroll that same season.
Basically, Beane took emotion and tradition out of his management decision-making and replaced it with rational management that relied on statistical regression models and probability comparisons. His player evaluations focused on factually important skills, not on looks or tradition. He simply used information from the past to guide decisions in the future, and it paid off.
Research has shown that statistical models serve better than unaided judgments for personnel decisions. In Moneyball
, Billy Beane analyzed data on the performance of baseball players and used the model’s output to select players for the Oakland A's. In this process, opinions and prior experiential knowledge may have provided inputs to the model, but they were not used to revise the model’s recommendations. Essentially, evidence-based procedures were used to make personnel decisions.
If high school basketball coaches adopt a Moneyball
/Krossover approach to their sport, they can spend less time securing data and more time coaching and educating their players, become more realistic in understanding players’ capabilities and performance, and be able to use data and logic to address the concerns and doubts of second-guessing, emotional parents and fans. Regarding that last point, keep in mind that use of statistical models can actually eliminate potential bias due to variables that are not related to performance, such as height, weight, race or religion.
Some may argue that the high schools that secure Krossover data are gaining an unfair competitive advantage. Well, if they are, the best way to counteract it is to subscribe to a service like Krossover as well, because (all things being equal) the advantage disappears when other teams also use the model. Of course, some may argue that their school system may not be able to afford the service. That’s understandable, but they should also investigate the possibility that the service may actually save the school money when they determine how much the existing coaching and scouting activities are currently costing them.
Most high school coaches are collecting data anyway, but they are doing it the hard way. They drive to one another’s teams’ games to observe them, they watch game films, they create their own stats, and they evaluate all this within the framework of their own passionate, nuanced expertise, honed over the years. Whether the data is tacit (in the coaches’ experience) or explicit (on their clipboard), the coaches need the data and they are using it! And it is my guess that coaches understand that things like heart and passion do not necessarily translate to effective performance (e.g., Tim Tebow). So using a service like Krossover is really not changing the coaches’ job, but making what they already do easier and more transparent.
and Krossover are also doing is providing stimulating anecdotes to help to promote the tangible value of statistics and predictive analytics
in decision making to a young audience. In education, the key instructional elements for promoting student interest and achievement in a topic are clarity, relevance, stimulation and organization. Demonstrating how business intelligence
can be used to evaluate and improve athletic performance at the high school level is clearly an effective instructional tool because it can dramatically affect something that many high school students really care about, as either a participant or a fan. And maybe, if we’re lucky, some of those students will see the value of business intelligence and want to learn more about it in college.
Recent articles by Richard Herschel