Image Of A Young Female Athlete And Her Coach High Fiving After Training

The Biopsychosocial Edge: Coaching From a Human-First Perspective

BY Andy Kirkland

Data will only get you so far as a coach — especially in the face of AI. Here’s why implementing a biopsychosocial approach to coaching can give you and your athletes a performance boost.

The boiling frog apologue suggests that if you place a frog in a pot of water and then heat the water gently up to boiling point, the frog will not hop out (despite having the opportunity to do so). Humans are similar. We rarely respond to slow changes in our environment, only prioritising action when there is an immediate threat. 

AI Is Turning Up the Heat

I’d suggest that many coaches in the endurance coaching market are already in hot water. This is because the majority of our coaching market is reliant on the ability to coach remotely. We use technology — including online coaching platforms such as TrainingPeaks, wearable GPS devices, and power meters — to prescribe training, collect, and analyze data. However, Artificial Intelligence (AI) is turning up the heat. 

AI algorithms are already capable of building evolving training programmes and providing basic feedback without the need for a ‘real’ coach. Whilst no commercial entity has quite yet mastered these algorithms to the degree of market monopolization, this time will come and it’s probably not too far away. 

The good news is that effective coaching involves far more than building training plans and workouts. It’s about building strong social relationships with your athletes, considering factors relating to subjective experience (which often can’t be measured), and taking a holistic view of performance. This article summarizes a paper that my colleague Joe Cowley and I recently published in Frontiers called An Exploration of Context and Learning in Endurance Sports Coaching. This paper highlights our perspective on the state of endurance coaching and this article focuses on communicating to you what we learnt to prevent you from becoming a boiled frog.

The Study

Upon entering the academic world, as a lecturer on the MSc. Performance Sports Coaching programme at the University of Stirling, I quickly learnt more about what I didn’t know rather than what I did. Despite having a Ph.D. in sports science, years of experience as a coach and coach developer, I wasn’t the expert I thought I was. Rather, there were huge gaps in my coaching knowledge. Furthermore, the academic literature surrounding coaching and coach learning did not reflect my experiences of what we actually do in the endurance coaching world. 

Therefore, I set about to conduct a research study to fill the knowledge gap and to confirm or refute my beliefs. Specifically, I wanted to help people like you to be better coaches and to encourage coach educators to design better programmes for you. Whilst I was delighted that around 10,000 coaches as well as coached and non-coached athletes participated, it took years to get the paper across the finishing line. My reflections on publishing a paper may be of interest to some of you. But what did we find?

Our Findings

Firstly, there is no ‘one-size-fits-all’ conceptualisation of endurance coaching. Rather, we work in a cottage industry, where planning and prescribing training is viewed as the most important thing we do. Despite my belief that supporting athletes to be more mentally prepared for competing should be the priority for coaches, this was seen as far less important for many. 

Furthermore, most coaches valued learning from experience more than formal learning programmes. This was presumably because the data confirmed that coach education had not prepared us well for our coaching roles. When seeking educational resources, we found that coaches tended to engage with websites and books which were commercial in nature (i.e., focused on selling products) and were created for athletes rather than for coaches. Many coaches also engaged with academic sources of information; however, these sources tended to relate to physiological ‘stuff’ rather than information on how to be a better coach. 

We concluded that coaches have a clear bias towards prioritizing readily quantifiable physiological factors over psychosocial factors. These biases were likely to be influenced by powerful market-driven agendas, which potentially leaves many coaches poorly prepared for their coaching role. 

Taking a Biopsychosocial Approach

When coach-athlete relationships are conducted primarily online, it is harder to develop the sophisticated relationships required to understand the biopsychosocial (i.e., biological + psychological + social) nature of performance enhancement. Streamlining processes that enhance communication between you and your athletes (like with TrainingPeaks’ built-in communication and feedback tools) can help reduce the virtual void.

In turn, learning to be a good coach requires social interactions with athletes and an ability to deeply reflect on coaching processes. The capacity to do so online without an appreciation of the psychosocial nature of coaching can limit a coach’s capacity to learn from experience. 

Top ultra-running coach Robbie Britton did a complementary study. His findings suggested that the best metric to measure coaching effectiveness was the duration of the coach-athlete relationship. Social factors like the coach believing in the athlete, shared values, and good communication were key to such effectiveness. 

However, the data from coached athletes in my study suggested that coaches could be better at providing clear expectations of what feedback they require and in asking better questions surrounding training and performance. In short, a gap in knowledge and understanding of psychosocial elements of coaching limits many endurance coaches’ abilities. 

How Not to Become a Boiled Frog

I started this article by discussing how AI was posing an existential threat to many coaches. My hope is that I’ve heightened your awareness of the dangers of placing your lilypad in a physiologically-biased cooking pot. Our research suggests developing a more holistic coaching approach through engagement with wider sources of knowledge is vital to becoming an effective coach. In a previous TrainingPeaks article, I posed the question where are you on the Dunning-Kruger wiggle?. Self-assessment of your coaching capabilities after reading this article may help. 

Finally, I suggest the best way to self-assess coaching effectiveness is by considering client retention rates and the duration of your coach-athlete relationships. Improving the quality of engagement with clients through effective feedback loops takes time. However, without investment, it will shortly mean that AI can probably do a better and cheaper job for your clients than you can.

Note: If you would like to discuss any aspects of our research, please feel free to email me at Additionally, I am part of a team which delivers a leading Online MSc. in Sport Performance Coaching programme at the University of Stirling in the UK. I am delighted to chat with any coaches who may be interested in studying with us.   


Britton, R. Promoting longevity in the coaching relationship: a coach and athlete led discussion. [Unpublished MSc. Dissertation]. University of Stirling (2018).

Kirkland, A. & Cowley, J. (2023, April 17). An exploration of context and learning in endurance sports coaching. Retrieved from

A Coach Using Trainingpeaks On Her Laptop In Her Home Office

Start Your 7-Day Free Trial

Explore all TrainingPeaks features and more with a free 7-day trial for coaches. Tools built by coaches to make coaching easier.

Avatar1538085439 7
About Andy Kirkland

Andy Kirkland Ph.D. is a Lecturer in Sports Coaching at the University of Stirling, delivering on the MSc. in Performance Coaching programme. He is an expert in endurance performance an active triathlon coach and a Chartered Scientist. He worked for British Cycling for 6 years as a coach educator and is a TrainingPeaks certified coach. You can find out more about Andy on his website.

Visit Andy Kirkland's Coach Profile

Related Articles