As with most industries, technological advancement and the use of data is becoming a necessity in order to keep up with the ever-changing world. However, in the athletic performance realm, we are just now beginning to see the application of data becoming more apparent. There is no question that data and the use of statistics has already infiltrated its way into athletics. When listening to any sports reporting or broadcast, you hear and see examples on the screen all the time. What we typically see and are used to seeing in an athletic sense regarding data are examples like:
The Capital’s Alex Ovechkin trailing Wayne Gretzky’s chart-topping career goals by 164 goals before the start of the 2021-22 season but having a higher goal per game average (0.6 vs. 0.61).
While at Clemson the prolific Travis Etienne having rushed 686 times for 4,952 years with 70 rushing touchdowns and caught of 102 passes for 1,155 and eight receiving touchdowns in 1,852 career snaps over 55 games.
Zack Greinke, who in 2015 posted a 1.66 ETA over 222.2 innings (6.94 IP/G) and kept batters at a .185 AVG, easily giving him one of the best pitching seasons
With the advancement of technology and data analytics, we can collect, gather, and analyze so much data in game like pitching speed, angle of a golf ball, where in the goal a keeper is least likely to block, percentage of 3-pointers likely to be made given certain circumstances. The list goes on and on…
However, why don’t we see huge advancements of using data to influence the actual performance of the athletes? With the stronghold that athletics has on so many people’s lives, why hasn’t the “science” followed suit with the great stride seen in technology, medicine/health, and the comfort of making objective decisions through data analysis. That’s where the new and growing field of “Sports Science” comes into play.
Let’s start by answering the question—what is Sports Science? Sport Science is the application of scientific theory to sport, or the study of sport using scientific methods of inquiry in the fields of human performance, athletic endeavors, and sporting competition. To rephrase— it’s changing data from just random numbers to valuable information to support decision making and influence outcomes of human performance capacity for the specific purpose of maximizing performance. Maximizing performance could mean less injuries, more wins, being bigger, faster, stronger, smarter. I refer to the application of Sports Science data usage as working smarter, not harder to WIN.
Lots of teams have a plethora of data, but are unsure of what to do with it, the questions to ask, or the problems to solve.
Some cool questions I have been able to answer using data:
- Which Functional Movement Screen pattern is leading to the most injuries? Where are these injuries occurring?
Example: A Football program I worked with I noticed that a high percentage of our hamstring tears had the same Functional Movement Screen scores/patterns—mostly having failed in ankle mobility. Sharing this objective data with strength and conditioning staff, a quick ankle mobility and stability exercise was added a few days a week in the warmup. At the end of the spring ball season, hamstring injuries were significantly reduced.
- How do we know how “ready” an athlete is before a heavy squat day? If they are fatigued, should they be lifting the same amount of weight? How do we quantify this?
Example: Jumping on a force plate before their squats, I was able to look at changes in softball athlete’s average eccentric and concentric force production to note how they were feeling. If feeling fatigued and sore, we would see decreases in their force output, but if energized and feeling good, we would see increases. In order to optimize the day and not put too much stress on the body that lead to injuries, we would then adjust the athlete’s one rep max or their bar velocity goal range in order to make sure that they hit appropriate intensity for the day.
- Is there a way to predict how fast a pitcher will throw a fastball? How do we individually train each athlete in order to optimize their fastball velocities?
Example: Again, using a force plate as a tool to monitor and track fatigue, force output, and readiness, I would have baseball players jump on a force plate before pitching/practice. Correlating changes in their force output and changes in their pitching velocities, we easily created a prediction model to see roughly what velocity we could expect that day. For games, this is essential as it quantifiably gives more insight into who is best prepared at that given time, especially if coaching staff is between two different athletes to play.
- Following an injury, is there a way to ensure that an athlete is back to 100% and ready for full-practice and games?
By using data in a “Return-to-Play” plan with Sports Performance Staff, it is very easy to find trends in the data that could have led to the injury (therefore, you can prevent it in the future), but also to compare the athlete to pre-injury to post-injury. If Athlete A tore their right ACL, we would want to make sure that all the measures are back to the same or better than their baseline. If we know that their Max Speed is 18mph, however now they are only reaching 15mph before feeling pain, we know this athlete is not ready for play.
One of my personal favorite things to track and monitor is Training Load (i.e., Player Load, Training Volume, etc.). A key component for both programming and monitoring an athlete’s progression or regression is the quantification of training load. Training Load can be collected in multiple ways (one of the most common being wearable devices) and gives an output of the amount of practice for the day. A harder, longer, more intense practice is going to have more volume than a light, short practice. One common method to provide insight into training is Acute to Chronic Workload Ratio (ACWR). ACWR works by using an athlete’s “fitness” and “fatigue” by taking their average training volume for the current week divided by their average volume for the last 4 weeks. This gives coaches a way to quantify practice volume and make increases in safe progressions. 80-120% is the “sweet spot” for an ACWR. While certain practices in a week might be significantly higher in volume than others, this weighted moving average helps to accommodate for that and even gives wiggle-room for the natural ebbs and flows that happen with training. The sweet spot of 80-120% is particularly important to note as going outside of this range increases an individual’s risks of injury—dropping below 80% leading to under training, and then going above 120% leading to over training.
Think of the following situation, you have been consistently working out for a year, and then you take a 2-week vacation in which you do no physical exercise at all. When you return from vacation, you begin the same workout routine as before. The next few days, you will probably feel extremely sore. This would come from the acute load now being much less than the chronic load in our ACWR. A situation where we might and commonly see ACWR rise above 120% in college athletics is when student athletes begin the “Training Camp” period of their seasons. While the athletes may have already been working out during the allotted NCAA allowed 8 hours, when the season goes into 20 hours a week, volume is more than doubled and injury risk increases.
A real-world example of how Training Load philosophies have been implemented and shown immense success has been with the Los Angeles Rams. The Rams, which have been one of the NFL’s top-five least injured teams since 2015 and have gone to two Super Bowls since then, places an extreme focus on monitoring players’ workloads to prevent unnecessary physical strain on their bodies. If a player has even walked too much, the coaches modify or have even been known to go as far as cancelling an entire practice. Preserving player health using workload analytics has become critical to the training regimens of many NFL teams. This is also especially important on an economic end as well—if you have Matthew Stafford contracted in at $40 million a year, I think you would do everything in your power to keep him injury free.
In conclusion, the use of data in athletics is not a new concept, however using it on a performance side is a huge breakthrough and quickly rising as a necessity for teams everywhere. Using performance data, analysts can dive in deep to see trends as to what is influencing game outcomes, what training leads to more success, how to decrease injuries, and much more. Implementing the concept of data science into the performance aspect of teams and individual athletes leads to working smarter, not harder for gameday success.
Comments