Navigating the Complexity of Learning & Performance in Sport

Mark Upton
Jul 29, 2015

The following is really just “thinking out loud” (hat tip Ed Sheeran) in the hope something coherent emerges. It emanates from recent discussions, experiences and reflections on the complexity of player/athlete development and performance at all levels, and the influence of coaches, sport scientists, parents, administrators and others.

In previous posts we have conceptualised players, coaches, teams and sporting organisations as complex systems so we won’t go back into detail on that now.

“For every complex problem there is an answer that is clear, simple, and wrong” — H. L. Mencken

Yes and no. For coaches that have little experience, base their practice on “what has always been done” and without critical reflection, then I would say this potentially rings true. For others that have embraced, navigated and “marinated” in these complex problem spaces for a period of time, the “clear and simple” solution may in fact be very much effective. This respects the nature of dealing with complexity, where relatively simple solutions (generated from rules of thumb & principles) can be highly effective. A recent example of this was an experienced manager opting to “do nothing” for the time being in relation to the next action in dealing with a complex (uncertain) problem. Unfortunately we often confuse complex with the need for complicated solutions and drift off into a futile search that, as a coach I used to work with would say, has us “flat out doing f***k all”, and often accompanied by a significant time & resource cost.

There are also social and cultural misconceptions about dealing with complexity and the inherent uncertainty that accompanies it…

The above sentiments were also echoed by a sport psychologist espousing the need to remove complexity from decision making processes. This desire to remove or reject complexity, in cases where it truly exists, is not helpful.


Many Sport Scientists struggle to deal with complexity and fail to respect that good coaches have developed (tacit) knowledge and expertise that enables them to function effectively without having to quantify everything and prove “cause and effect”. Coming from a culture of Newtonian Science where deterministic thinking reigns, young Sport Science practitioners can be “paralyzed” when data and a textbook fail to provide THE answer as to the most effective next action.

From the late-17th Century until the early 20th, the Laws of Motion and other linear, mechanical principles discovered by Isaac Newton dominated the understandings of science and filtered down into every aspect of the Western world. This view of reality over time penetrated our education system, our culture, our language, our organizations and our management practices so completely that it became taken for granted. Most people are not even conscious that they are using what is called the mechanical view of reality when they think and talk. This view of reality assumes:

* Things happen because something causes them to happen (cause and effect).

* We can understand what happened by reducing things to their components or parts and examining those parts (reductionism).

* The universe is orderly, follows natural laws, and works like an incredibly complicated machine.

* The best way to manage people is to organize them into a clear structure and control them with clear directions.

* The best results occur when work is streamlined to be as efficient as possible, with a minimum of wasted effort, producing the most output in the least amount of time (the “lean machine”).


Experienced practitioners have come to realise there are other ways of working. Nick Grantham has written a couple of excellent blog posts –Push the Button and Programme Design: The Answer. Nick describes an approach to dealing with complexity and uncertainty that is advocated by complexity experts — that is, pre-intervention analysis and data collection will provide limited value. The best way to better understand the dynamics of a complex system is to take action (ideally with small “safe to fail” experiments) and sense how the system (player/athlete/team/organisation) responds. This informs the next response, particularly whether to amplify the action or dampen it (try something different). Another caution when dealing with complex systems — “copy and paste” methods are likely to fall flat, ie what worked with one player/situation/environment possibly won’t with another.

An experienced sports nutritionist described this recently — a nutritional intervention that produced great results with one athlete was “copy and pasted” with the next athlete, but resulted in completely different (non-desireable) outcomes. We also see this with head coaches in team sports who think the “drivers” of success with one team/club can be replicated with a new team/club, only for things to pan out completely differently. The takeaway is that in a complex system, there are rarely single actions that are “drivers” – only modulators. It is an easy trap to fall into as we have a tendency for retrospective coherence – looking back on a specific situation and, with the benefit of hindsight, being able to identify the mechanisms that resulted in a certain outcome(s). We then risk “copy and paste”, claims of “best practice” etc.

Darren Roberts featured on a podcast recently and described the “performance playground” he tried to create for his extreme sport athletes. This exploratory space for trial and error (not to mention athlete autonomy) reflects a good way of working with complexity, in this case the performance and/or rehabilitation of these amazing athletes. The “performance playground” reflects the approach Al Smith and I are proposing in our quest to support people who want to create dynamic learning environments. Yet Darren’s was equally a cautionary tale, making specific reference to the fact that these environments do not lend themselves to easy forms of measurement and the pretty colours on spreadsheets that many are fond of.

This leads to my final thought on the escalating obsession with measuring and quantifying everything. Whilst there is clearly great value to be derived from data when used appropriately (narrative data is proving useful in complexity), I’m more concerned with the dynamics that are driving this obsession, which seems to stem from a desire to control everything and remove uncertainty, remove complexity. Attempts at Talent ID from a young age reflect this.

If we are genuinely committed to helping players/athletes be their best, we must embrace the fact that player development, coaching, applied sport science is messy…complex…uncertain. Our actions, methods, frameworks, policies, processes and organisational structures must then reflect this. Do this and we increase our chances of successfully navigating towards positive outcomes and rewarding experiences.

…“thinking out loud” complete 😉