Are all sporting decisions complex?
It’s the final game of the season and the ball is passed to the star new recruit. They display poise under pressure to complete the winning play and are showered with adulation by fans and media pundits. The general manager and scouts behind the decision to bring the player to the organisation are hailed for what is regarded as a recruiting masterstroke!
A player is deemed fit to return after 12 months out injured with a serious knee injury. Five minutes into their first game back they fall awkwardly, re-injuring the knee and are stretchered from the field. The morning after the match, the medical team come under public scrutiny, with various media pundits suggesting the player was returned to competition prematurely.
On first glance, these two stories appear to have little in common. However, they’re both just two examples from sport whereby we can be too quick to praise or criticise a decision maker with respect to their apparent influence on an outcome, which in reality they have little control over. More often than not, there are many other influences impacting the outcomes of such situations which have nothing to do with the decision maker.
What if the ball rolled the wrong way and the play had not been completed? To what extent would the decision to recruit the player be considered less successful?
Or, what if there is some undiscovered combination of genetic markers that meant the player would have been re-injured no matter when they returned to competition? One could spend hours generating any number of scenarios.
Complexity is everywhere
I’ve been contemplating for some time the notion that most, if not all, sporting decisions are complex. Almost certainly, much more complexity exists in sport than we account for in our systems and methods of decision-making. Whether it is in the board room or on the coaching field, what this means practically is that no single correct solution to many of the problems that we face exists.
So what is complexity? Complex problems are characterised by a large number of elements, interacting non-linearly often with minor changes producing major consequences. Such problems are dynamic and difficult to predict, with outcomes emerging seemingly at random. Considerable recent work has aimed to increase the recognition of complexity, in order to help organisations enhance their decision-making practices. One example, from the Centre for the Evaluation of Complexity Across the Nexus (CECAN), summarises complexity as 16 features, such as change over time, tipping points, feedback loops, unpredictability and emergence
Another is the 2007 Harvard Business Review article by Dave Snowden and Mary E Boone, A Leader’s Framework for Decision Making. In this work, problems are described as falling into one of four quadrants: Simple, where cause and effect relationships are well established; Complicated, which refer to scenarios where one or more correct courses of action exist, but which may not be easy to identify without certain knowledge or training; and Chaotic, where cause and effect is impossible to determine, things are unknowable and the only immediate decision relates to how to (re)establish order.
Complex problems on the other hand, in addition to those mentioned by CECAN, are represented by characteristics such as nested systems, path dependency, self-organisation and adaptability: all traits synonymous with humans. In differentiating complicated from complex problems, Snowden and Boone use the example of a Ferrari. Although difficult, a mechanic could completely take the car apart and build it back again piece by piece; it is static. A rainforest on the other hand is complex in that it is much more than simply the sum of its parts, it is in a state of constant change and interactions meaning that no exact situation is ever repeated exactly. That very much sounds like a description of sport to me.
So it begs the question: How many problems in our jobs do we treat as simple, which in reality are actually complex? I won’t give examples; I’ll let you think of some from your own environment. It’s common when things are going well for us to think in simple terms. That things are going well because we’ve done our due diligence and have control of the situation, safe in the knowledge that one factor in a problem directly causes another and thus is easily manipulated.
Complex problems too can also be wrongly categorised as complicated. For example, it’s tempting to think that if an athlete follows the exact instructions of their coach then they won’t be injured. That despite being complicated, there is nonetheless a process that can be followed prescriptively to avoid the negative outcome. But despite our increasingly advanced knowledge of the human body, there is still much that we don’t know. If we did, then much less injuries would occur. Similarly, a coach does not have the ability to change the outcome of a match due to the substitutions and tactical changes they make, no matter how adroit they may be.
The point in both examples is the same. However skilled the coach or practitioner; they cannot directly cause the outcome to change, because the problem they are working on is complex.
Regardless of whether you agree with my assertions, it’s clear that there is a growing adoption of complexity in sport, with the topic gaining considerable attention in problems relating to injury, training and football playing positions. In both research and practice, I believe this increased awareness has largely been driven by a change in our collective perception, which can be represented as a four phase cycle.
The first phase, Question, relates to our continually increasing ability to measure and record the world around us, largely thanks to developments in technology. If we consider the most challenging problem we’re faced with in our jobs and ask whether we’re acting on all of the information currently available, most of the time we would probably answer negatively. We realise that there is more to know, potentially a better solution at hand. When we’re forced to make important decisions without all of the available information, it can often result in us feeling a lack of control of the situation at hand. Sometimes though it can also invoke a deep need to learn more, in order to improve next time. Other times we want to know more merely out of curiosity, which of course we all possess by nature. Technology is not the stimulus for this curiosity, but it certainly is the fuel to the fire.
So, in Phase 2 this drives us to better Inform ourselves, not just through collecting new types of data but also better versions of what we already have. But this phase is more than just compiling more data. It also relates to the tools that transform data into usable information: computing and algorithms for example.
By the time we reach Phase 3, we are in a position to Intervene based on the new insights provided by this information. From something as simple as generating a new report through to implementing a new insight that leads to a fundamental shift in practice; the extent to which the intervention leads to solving the problem is only part of the story. The act of intervening itself causes our sense of control to somewhat return, however fleeting it may be.
In the final Phase 4, we Assess the intervention and the new information with respect to the new light it has shed on the problem. Even with a seemingly satisfactory solution, the desire to re-iterate remains untamed. New and better data sources emerge. Our competitors learn something that we don’t. Technology continues to fuel the fire.
What does it mean for practice?
This cycle means that our perception of problems that we perhaps once viewed as simple, is now constantly moving to one of complexity. Whilst this may be an appropriate outcome, it does come with potential downsides. Trying to collect every piece of information on a given problem is fraught with danger, even if it is helping to better understand the problem at hand. Researchers have written extensively on the implications this has for our cognitive capacity as well as our abilities to use incomplete information sets in various forms of decision making. Fortunately modern computing is helping with this by doing a lot of the hard work for us.
Nonetheless, paralysis by analysis is a very real phenomenon; if we focus on always having perfect knowledge prior to acting then we would never get anything done. And then of course there are times when our motivations for collecting more data are questionable themselves; perhaps we’re merely trying to keep up with the latest technology our competitors are using.
Further, as complexity becomes more apparent to us on a given problem, existing in this uncertainty can also be uncomfortable. Thriving is even more difficult. In jobs that have strong roots in evidence-based practice, this acknowledgement can be completely revolutionary, causing us to question our very training. As a result, so often it is tempting to revert back into simple ‘cause and effect’ mode because it feels better, thereby creating an illusion of control.
And that perhaps the biggest concern of all in this. The more we collect, the more opportunities to create manipulated narratives from a selection of the data. And as we so often see, that creates a race to the bottom, as the narrative most linked to risk aversion often wins out. The current preoccupation with measuring athlete training load in sport is a perfect example of this. With every new metric collected on the health and wellbeing of an athlete, so too emerge new reasons to justify a restriction to their activity. But increased complexity should provide more opportunities for creativity. Why, so often then, doesn’t it? One reason is simply that it is easier to control and constrain, then set free and create.
Smart organisations realise that the more factors they’re aware of which can potentially impact a problem, the less certain they should be about the ability to manipulate the outcome. It’s like the old saying, ‘the more we know, the more we realise that we don’t know’. And this of course leads us back to the start of the cycle, realising that we still need more data – and that we may never have enough. Again, the good organisations plan for this with agile technology infrastructure, integrated decision making systems and a culture of growth and humility; an idea meritocracy if you will.
So next time you read that article or piece of research that says we need more data, stop and pause. Whilst enhanced sources of information are shedding new light on the same questions we asked 50 years ago, maybe it’s more than just the data? Perhaps in order to make real progress, we also need to change the lens by which we’re viewing sports biggest problems.
