In the age of advanced data analytics, sports statistics abound. Take professional sports, for example. It may be common knowledge that basketball player Michael Jordan is revered as the greatest scorer of all time – averaging a record 30.12 points per career NBA game. We know, by virtue of statistics, that Drew Brees is the most accurate quarterback of all time sporting the highest career completion percentage on pass attempts in the NFL. What about Baseball hitting metrics? A great hitter scores high on per-season statistics such as batting average, on-base percentage, and slugging percentage (total bases achieved divided by number of at-bats). Regardless of sport, tradition suggests that statistics are best consumed as a summation of effort over a season or career, and rarely do we compare year-over-year or game-to-game statistics with any value. Further, results analysis is generally Pass/Fail in nature. Did the athlete succeed? Is this number successful? Why or why not? By the time these questions are answered, it’s too late to do anything about it!
In the world of quality, we know better! We understand that any process – even sports – can and should be considered as a process over time if we want to continuously improve. After all, professional athletes are surely striving for their numbers to go up over time. It would be interesting to focus less on “athletic doing” and more on “athletic learning.” What’s important in Statistical Process Control is that we ask the right questions and understand the variation in the process actually being monitored.
As a coach of a highly competitive club volleyball team, I was afforded a terrific opportunity to test-drive this new way of thinking. The ultimate goal was easy: improve consistency and performance per athlete over the course of an 8-month club season. The process, however, is what we’re going to do to achieve our goal. This is where I noted that the decisions made by the coach are equally as important as the effort of the athlete. Are we emphasizing the right skills in practice that will give athletes the best chance for success? Are we providing the right feedback to our athletes to guide them to skill improvement? It’s not enough to just ask; “did the athlete get better?” and then distribute blame to the individual for failing to improve. Perhaps it’s the fault of the coach for generating insufficient practice plans. The following was the ultimate focus of my SPC in sports development study:
Are we effectively preparing in practice and seeing statistical improvement that we can expect to translate into improved match performance? Players’ performances were graded daily in practice on the four essential skills for success in the sport of volleyball, and decisions for future training sessions were made following each practice based on the trends (or out-of-control signals) present on the charts. The players (and even better: their parents!) would walk home with a weekly recap like the following:
What questions does this chart answer?
With a basic understanding of SPC, we know the answer. First, this tells us whether the player’s performance exhibits stability. In other words, is his or her future performance in match play predictable based on data from prior practices? Some of the best athletes ever were the most consistent players. Second, trend lines can help identify if player performance metrics are increasing or decreasing as the season progresses.
We see by the red boxes in the example above that this athlete’s performance is not stable in two of the four skills being measured. In both cases, a series of consecutive practices resulting in scores lower than the athlete’s overall average occurred. When a process is unstable, it would be unwise to take any value from the trend lines. However, the bottom two charts exhibit stability and positive trends, so we may be able to conclude that we’re doing the right things to see increased performance in those two skills.
This exercise highlights the importance of the roles everyone plays in the learning process. Players, coaches, and practice plans are only three of the variables involved with athletic learning. It is up to us, as coaches, to identify the controllable variables and best interpret what statistics over time are telling us. After all, that’s why they pay us the big bucks!