Some sports scientists stay in one sport for their entire career and carve out a deep niche. Yet in the increasingly high-paced realm of human performance analytics, more organizations are seeing how expertise in one domain crosses over into another and hiring accordingly. It’s this dynamic shift in the industry that has taken Jesse Green halfway around the world from Australia’s Gold Coast to the East Coast, where he currently serves as the Pittsburgh Penguins’ director of performance and sports science. Along the way, he has seen how academic studies and practical applications are mutually beneficial, why learning on the job in different high-performance environments broadens and deepens sports science skillsets, and in what ways analytics, medical, coaching, and front office personnel are collaborating more closely than ever before.
Feeding a Fast-and-Slow Approach to Data Analytics
There can be an either/or divide between academics and professional experience in sports and data science roles but from his first job after college onward, Green has shown that the two are actually complementary rather than mutually exclusive. While working with the Brisbane Lions, he was also studying for the honors component of his BSc in sports science at Bond University, which allowed him to see the day-to-day application of research in high performance sport. “There was always an ongoing connection between the academic side and a more pragmatic approach via the students that were embedded within the program,” Green said.
Green also observed how sports science students can further their own studies by working closely with a club’s HP analytics group to gather and interpret certain data sets and simultaneously help the performance staff make more objective decisions using the same information. The practice is commonplace in Australia, the UK, and Europe, and Green believes it would also be beneficial to college athletic departments in the US.
“It’s a symbiotic relationship between the PhD student and the program, where they become embedded within that performance support staff,” Green said. “The research that the student is doing is facilitating some day-to-day processes that the organization may be utilizing, while they get access to the data and an inside view of the sport.” At the Brisbane Lions, the PhD candidate Green worked with was Nick Murray, who’s now head of performance for Melbourne Football Club. Back then, he was a PhD student focusing on the application of GPS data. “Not only was he benefiting the club by analyzing the information and using a bit more of an analytic process to answer his own questions, but he was also providing immediate insights for the coaches to utilize that GPS data,” Green said.
Much like Green’s own AFL experience of combining academics with practical work experience, Murray’s GPS project was a potent example of how the more methodical side of human performance analytics can feed into the frenetic environment of pro sports, and vice versa. Utilizing PhD candidates could also provide another option for resource-constrained organizations to expand their sports science capabilities if they cannot afford to hire more full-time HP analytics staff. “The data is coming in, and that student is providing the immediate insights that are required – like reporting and monitoring – while also working in the background on some of these bigger ticket items that maybe two or three years later can change those fast processes,” Green said. “It’s this fast-and-slow approach and a lifecycle that can continue student after student and project after project.”
During his four years with the Brisbane Lions, Green got to see firsthand how the performance, medical, and analytics staff within an AFL team worked collaboratively to serve athletes and coaches. One example was the return-to-play process when an athlete was injured. “We had really sound practitioners working on the strength and conditioning side and sports scientists on the data side who facilitated decisions being made on the return to performance and rehab side,” Green said. “That caught my eye again and again and confirmed that data can be very helpful when used appropriately in high-performance sport.”
Gaining Multi-Sport Experience
When Green left the Lions in July 2017, his next role was 9,000 miles away with the University of Louisville. He initially served the Cardinals as a Performance fellow, which was a dual role as a coach and analyst for the Olympic sports performance program. The switch from working with a single team in one sport to serving multiple squads across several different sports meant that he was constantly helping prepare athletes for the pre-season, in-season, or off-season. This empowered him to create strength and conditioning programs and player monitoring methods that were applicable to different disciplines.
“At Louisville, we had a primary, a secondary, and a tertiary coach for a number of different sports, and there were only two of us in the performance analytics division – the coordinator and me,” he said. “So at any one time, I was working with three or four sports – which could be a court- and field-based team sport and an Olympic one like track and field – during a given season,” Green said. “That’s where I see the benefit of working in a collegiate environment from an analytics and strength and conditioning perspective – you get exposure to a number of different sports, which can really accelerate your growth from a preparation and monitoring standpoint.”
Looking back on his time at Louisville with the added perspective of later working with the Sacramento Kings, Green recognized the disparities in the budgets and staffing between college and professional sports teams. With this in mind, sports scientists and coaches at the college level have to be highly strategic in the technology they select and decide what will provide the greatest bang for the buck in terms of potential performance improvements. They also need to think about what it will take to implement, scale, and maintain new systems across a large athlete population.
“In a college sports setting, you could have 300, 400, or 500 athletes, so if you were to get GPS units, that’s an enormous cost versus an NHL team, where there are only between 24 and 28 guys. In college, there’s a mindset that if we’re going to get it for someone, we have to get it for everyone. But in pro sports, it’s a more individualized approach because you can get technology for a small subset of players. From a resources standpoint, do we have the ability to buy it for everybody, or is it more of a bespoke approach? And do we actually have the time to roll it out? These considerations don’t just apply to technology, but also staffing, investing in your current team, and other aspects of your program.”
As Green’s role at Louisville progressed, it gave him the opportunity to ask and answer such questions as he played the role of both practitioner and sports scientist for women’s basketball and field hockey, men’s soccer and tennis, baseball, and several other disciplines.
“I started off in performance analytics, but then flowed into more of a 50/50 role of strength and conditioning and sports science,” he said. “I was able to refine my skills a little bit from the sports science and the analytics standpoint, embedding more systems and processes within our respective sports where the technology could actually save us time and enhance that program, as opposed to being a burden of collection analysis. That’s where I confirmed for myself that the data side of things can facilitate making decisions and help systemize informing some of them, especially from a planning standpoint. We were always using this information to plan an off-season program, monitor in-season and make subtle adjustments, or ramp up in a pre-season. So that really accelerated my development.”
Acquiring T-Shaped Skillsets
As Green has transitioned between various roles – including strength and conditioning coach, performance analyst, and head of sports science – he has seen that no matter how beneficial professional development resources can be, there is no better teacher than gaining practical experience in a variety of HP analytics settings. “Having worked in a college setting and pro basketball and Australian rules football, I feel more prepared to tackle, analyze, and solve problems in my current role based on my experience that I’ve had in multiple different sports. You can certainly provide a better product when you’re working with just one individual team and that’s my preference, but from a developmental standpoint, when you’re working across multiple sports, there’s no substitute for that.”
He has also become a firm believer in acquiring deep expertise in a singular area of focus while also acquiring a broad set of complementary skills. “Make sure you’re a T-shaped employee, where you have one deep specialty whether that’s strength training, conditioning, sports science, or data analytics – and also a broad skill set and knowledge of all of the neighboring disciplines as well,” Green said. “When I can hire an extensive sports science or data analytics group, it’s certainly going to comprise of T-shaped employees because a team like that is greater than the sums of its individual parts.”
To build greater breadth (the top of the T-shape), Green thinks that anyone in an HP analytics role would do well to build relationships with colleagues across the organization. “The data engineer has to know what the sports scientist, the full stack developer, and the data analyst are doing, at least to the level of being able to have a conversation and understand the tasks they’re completing,” he said. “If you’re in a high-performance environment with neighboring disciplines, spend time with the physical therapist, trainer, and coaches to understand the language they’re speaking and try to bring people together. As sports scientists, we should be facilitators of insights between roles and the decision-making process so that we bring value and make everyone else’s job easier.”
Bringing Sports Science and the Front Office Together
Green’s four years with the Sacramento Kings not only helped develop his skillset as a performance director and prepared him for a similar role with the Pittsburgh Penguins, but also gave him a glimpse into the future of sports science. While the discipline has already expanded and blurred the lines between analytical, medical, performance, and coaching staff, it sometimes struggles to get a handle on the sheer volume of data coming in. “The explosion of collecting and accessing all this information is far superseding the ability for us to interpret, understand, and derive insights from it,” he said.
Green believes the next evolution of HP analytics is for the sports science team and front office/operations analysts to work on some of the same data sets. Doing so will yield new explanations for why certain changes occur from game to game, which variations are the most meaningful, and how athletes’ in-game stats relate to offensive and defensive outcomes. Insights can then be communicated to the coaching and performance staff with the help of systems like Smartabase to inform player preparation.
“From the performance and monitoring side of things, adding context from the game itself and looking at very specific front office metrics has potential,” Green said. “For example, maybe in a game we see that a player has had a 30 percent increase in high-speed running meters or accelerations and decelerations. Without looking at the metrics that analysts collect, we’d think that was a significant increase. But then we’d maybe take into account that there were a lot of transitions because of the style of play employed, so the increase may have been a function of a tactical change. Without this extra information from the analytics staff, we would’ve lacked context. That’s what we try to do as sports scientists – see what has changed, why, and if it’s something to be alarmed about or just a normal response to a planned tactical, or positional change for example.”