Performance Engineering for Team Sports

As Ryan Weisert, writer for the Phoenix Suns has put it, the cat is out of the bag.

Science Makes Hoop Dreams Reality

Friday morning, I gave a talk at the MIT-Sloan Sports Analytics Conference, presented by ESPN. In it, I detailed how the various models I work with can be applied to team sports such as basketball, football and soccer to engineer peak performances for both athletes and teams. These concepts are embodied in PhysFarm’s  Mercury System software and algorithms. At the bottom of this article, you’ll be able to click on a link to see how it is applied to our endurance athletes using the Apollo System.

Season Planning

Every athlete is a complex system, representing a unique combination of biology and talents,  learned skills and psychology. However, athletes are not machines. Each of these things adapt and change on their own time course, and this time course can be calculated. Moreover, how and when you train the athlete alters this time course.

What you end up with is basically an engineering problem: How can we best plan training to minimize the risk of injury, and maximize both recovery and ability to perform? How do we get the greatest number of players into the best condition for the most games?

Our techniques can help coaching, management and medical staff answer these questions.

Game Management

All athletes use two separate but related energy systems during exercise. There is one system that dictates performance over long time periods and distances (think of a marathon). There is  a second system that dictates the ability to perform sprints and explosive exercise (think of a fast break).

This “sprint” system can be thought of as a battery. It is easy to drain through sprints and jumping, because it recharges rather slowly. When the battery is empty, athletes slow down, get sloppy, and have an increased the risk of injury. By monitoring athletes during exercise, it is possible to know just how full or empty this battery is at any point in a game.

In other words, it is possible to calculate when the athlete will get fatigued. This can help coaches plan when to sit an athlete, and when to put them back in the game for a critical play.


The rehab of an injured athlete is a special class of strength and conditioning problem. When asked how long it will take to get a player back to practice and game play, the medical staff is often forced to make an educated guess based upon clinical evidence.

The same models that can be used to direct training and recovery can be used to understand the speed of the rehabilitation process and help provide both medical staff and management with important information.

PhysFarm’s Mercury System Answers These Questions, and More

The take home message from the MIT Conference is that we have entered the era of Big Data and Analytics, but that no one really knows what to do with the data or how to present it to coaches on the floor. We’re here to help.

The beauty of our approach is that we don’t dictate what you do. The Mercury System is adaptive. You tell us what measurements and metrics are important to you, and we customize the software to work the way you already think as a sports professional. The system doesn’t tell you what to do (although it can help). Rather, it helps you decide when to execute your strategy, and how to plan for that moment.

If you would like to read a bit about how we apply this to endurance athletes using the Apollo System, check out this link. The functionality and feature sets of Mercury is quite similar, but adapted to the needs of the sport.

Interested in learning more? Contact Dr. Skiba for more information at, or by phone at (908) 463-5292.