The Constraints-Led Approach: Principles for Sports Coaching and Practice Design

The aim of this chapter is to provide guidelines to implement the four principles discussed previously into the design of a CLA session. We will provide a step-by-step planning framework that will enable a coach to build his or her own CLA sessions. As part of this approach we will introduce new resources that coaches can use and provide some ideas to enable the consideration of the level of representativeness of practice and deliberate designing-in of affordances, variability, and instability. We will introduce representative learning design (RLD) and variability ‘dials’, which require coaches to pay attention to these core ideas in their session design and provide a tool that coaches can use to systemically manipulate constraints. We will conclude by providing a debriefing template to emphasise that adopting a CLA is a dynamic, iterative process where the coach and performer coach create an emergent learning and performance programme. We begin by adopting the ideas of the GROW model (Whitmore, 2017) to pose a series of questions that can form the foundation for CLA learning design.

Planning a CLA session

Step one: the current landscape

When designing any session, it is important to understand the current landscape for the performer(s). This requires collecting as much information as possible about the individual(s) and the performance environment. In line with Principle 1 introduced in the previous chapter, we have adapted Whitmore’s (2017) GROW model to frame this process. GROW is an acronym to describe a process that involves four key steps. Goal: what do you want? Reality: where are you now? Option: what could you do? Will: what will you do? In our adaptation, we have used Way Forward to replace ‘Will’ as we wanted to emphasise that this process is in line with our goal in this book, to move coaching forward and provide new alternative ways to frame coaching practice. As such, it is important to recognise the link between the GROW questions to the how, what, why and when (and whom) questions we raised in the earlier discussion. Adopting this more systematic approach to CLA session design highlights how constraints-led coaching can be much more nuanced than simply adding in different rules seen when using SSCGs, for example. As shown in Figure 6.1, we have used GROW and added questions that frame it a CLA to coaching.

Figure 6.1 Adapted GROW model for constraints-led session planning.

Goal

What is the goal or intention for this session?

Unless this is the first session a coach is going to run with an individual or team, he or she should have a pretty good idea of what needs to be worked on in this session. Ideally, this decision should be informed by some form of performance analysis, which may be as simple as watching how they performed, or at higher levels based on some form of data analysis. It is important that the goals or intentions of the coach are matched with those of the performers. As highlighted in previous chapters, a mis-match between what coaches want to see and what the performers can do can lead to frustration and failure for the coach and performer. Just because the coach wants to see the performer reproduce the perfect movement model shown in coaching books, does not mean that they have the current capability to achieve this desired goal and repeated attempts that lead to failure may have a detrimental effect and lead to negative consequences such as lower confidence, reduced interest and, ultimately, decreased motivation. Consequently, apart from with very young performers, a joint approach to session design is recommended to ensure that there is buy-in from the performers. This does not mean that everyone should sit down and plan the detail of session together (which, of course, they could do), but it does mean that the coach should at least talk to the performers to get their views as to the key areas to focus on moving forward. In elite sport this process might take place in the post-match review, with the coach using the feedback to determine the goals for the session.

How does this link to the overall goal?

In this book we are assuming that the coaching is geared towards improving performance and that the individual sport performer or team are taking part in or preparing for some sort of competitive programme. Consequently, the coach and performers should be working towards an overall goal. These goals can be short (e.g. to win the next competition, to qualify for a tournament), medium (e.g. to be selected for a representative squad) or long-term (e.g. to become a professional performer). The session that is being planned should, therefore, sit in a connected space (i.e. practice design and competition performances should be linked) and form part of a process of attempting to achieve an overall goal. It is worth noting that this does not imply that the overall plan needs to be rigid and fixed a long time in advance (i.e. a detailed, session-by-session, periodised plan). Rather, each session’s goals need to emerge as the performer progresses (or not). Of course, this includes within as well as between sessions. This is comparable to planning a journey; we know the destination and what time we must arrive by. We may have a rough idea of the options available to get us to the destination, but actual decisions about which road to take at a point in time might change in the moment due to local conditions. In our coaching, this means that we are adaptable and flexible in the design of our overall programme and even in defining the ultimate overall goal.

The focus for learning is . . .

Once the goal for the session is determined, it is useful to summarise what the focus of learning is going to be for the session. The focus of learning is on closing the gap between the current skill level and the level required to succeed. Essentially, this means identifying the most relevant ‘rate limiter’ (i.e. the key factors that are acting as a handbrake) on the emergence of higher levels of performance. Traditionally, the term ‘rate limiter’ is associated with motor development and is typically used to describe how the relatively slower rate of development of a specific subsystem can act to prevent a new behaviour from emerging. For example, a lack of strength can delay the onset of walking in babies (Thelen & Smith, 1994). In sport coaching these ideas can be broadened to include environmental and task constraints such as limited opportunities to practise in specific performance settings or against highly skilled opponents. Of course, given the above definition, the focus of a session can relate to any aspect of performance and could involve a specific focus on individual skills, team co-ordination, emotional control, perceptual skills and so on. It is worth noting, that while the session will have a specific focus for learning, the interactive nature of constraints will likely result in some other expected and perhaps unexpected learnings that coaches could exploit in their practice.

Reality

What is the current skill level?

• Co-ordination?

• Adaptability?

As mentioned previously, the goals and intentions of the coach need to be matched to those of the performers. Therefore, a careful assessment of the current skill level of the performers is required to ensure alignment to the goals set for the session. Recently, in an attempt to provide a more workable categorisation of skill level for teachers and coaches, Renshaw and Chow (2018) proposed an adaptation of Newell’s (1985) three-stage learning model (Co-ordination, Control, Skill) (see box below).

Key concept: Newell’s (1985) skill framework

Stage 1: Co-ordination. Is focussed on the assembly of a suitable, functional (i.e. it works) coordination pattern.

Stage 2: Control: This phase is concerned with building on the established co-ordination patterns to enable greater attunement to dynamic performance environments. Gaining control of the co-ordination pattern is therefore typically characterized by subtle, refined variations in movement patterning to strengthen its adaptability in different circumstances, allowing them to function more adaptively in many situations (Liu, Mayer-Kress, & Newell, 2006).

Stage 3: Skill: The final stage of Newell’s (1985) model refers to the optimization of skill. Essentially, this means gaining even more control of the coordinative structure through the learned ability to exploit intrinsic dynamics and more flexible and open to exploit environmental information sources. Performers therefore become more flexible, efficient and have increased control as they become adept at exploiting forces from the movement, including instantaneous adaptations to sudden and minute environmental changes (Newell, 1996).

A more complete review of Newell’s (1985) framework and the ideas behind can be found in the original paper and in Davids et al. (2008).

Describing a performer as being at ‘Skill’ level was felt to be confusing and not helpful when talking to practitioners about skill learning. Additionally, as highlighted by Newell himself (1985), the framework only fits well with discrete (e.g. hitting a ball, throwing an implement) and serial tasks (e.g. a gymnastics tumbling routine, penalty flick in field hockey) and not continuous tasks (e.g. running, swimming, cycling) where strategy ‘over rides any single aspect of individual co-ordination within the context of the activity (p. 296). Given the focus on developing flexibility and adaptability in the second and third stages, Renshaw and Chow (2018) reduced Newell’s model to two stages. Stage 1 was categorised as a Learning to Co-ordinate phase (as per Newell) with a second stage being classed as a Learning to Adapt phase (a collapsing of Control and Skill from Newell’s model).

In reality, the age, experience and standard of competition (i.e. international, national, 5th Division), may lead coaches to assume the current skill levels of their performers; however, it is still worth the coach paying particular attention to each individual performer’s status, particularly in sports where there are a range of skills that need to be utilised (e.g. backhand and forehand shots in racquet sports, creating different flight characteristics in golf, or kicking a ball with both feet and different surfaces in football). As many coaches understand, in any squad or team, performers are at different stages and levels in their development. Additionally, a performer may be at the adaptable stage in some skills, but still at co-ordination level in others. Unfortunately (in our experience and opinion), a lack of understanding of the stage of learning leads many coaches to only focus on the co-ordination stage. Hence, even with elite performers, coaching is often focused on breaking down deeply learned complex moment patterns in attempts to re-model the co-ordination pattern. While at times this may be necessary, it has significant consequences for performance and the psychological state of the individual. We would caution such a focus and suggest that coaching time might be more effectively spent helping the performer to become more adaptable in their performance. For those at Stage 1, session goals should be more about creating learning opportunities that promote searching and exploring to facilitate the development of intra-individual-environment or inter-individual-environment co-ordination patterns (Renshaw & Moy, 2018). Those performers who have developed basic co-ordination patterns move into the Adaptation Phase. This phase is concerned with optimising performance through developing stable, yet flexible, co-ordination patterns based on the emergent ability to exploit affordances available in the individual-environment system (i.e. passive, inertial, and mechanical properties of limb movements (Renshaw & Chow, 2018; Davids et al., 2008). Essential to this process is providing learning opportunities to educate; (i) intentions (the process of setting up the perceptual or perceptual-motor systems to detect a particular informational variable), (ii) attention (the process of coming to attend to the more useful variables) and to calibrate actions of performers (the process by which the single-valued function itself becomes adjusted). Attention is said to be optimally educated if perceivers detect a variable that specifies the property that they intend to perceive (Jacobs & Michaels, 2007). To support this goal, learning environments must be highly representative of the performance environment in order to create high levels of transfer in terms of knowing what to do and how to do it in specific situations. Ensuring that session design has a high level of RLD has the potential to result in greater adaptability as performers are able to demonstrate an ability to adapt to, and ultimately instantly exploit, minute environmental changes leading to greater smoothness and fluidity in movements (Davids et al., 2008; Newell, 1996).

What affordances of the performance environment do you want to design-in to practice?

• Which?

• Why?

• When?

Once the stage of learning of the performers has been determined, specific decisions can be made about the design of the learning environment. Of particular relevance is deciding which affordances need to be designed-in, knowing why they should (or should not) be available and when they should be included in practice task design. For example, if a four-year-old child is learning to hit a tennis ball with mum or dad, the key affordance that needs to be provided is balls thrown to him or her on the bounce at an appropriate pace and in a place to invite swinging a racquet at. In contrast, a performance-level tennis player who is learning when to play a passing shot or a lob, needs the affordance of an opponent who is charging the net and inviting a return hit into the space either to the side or behind. Clearly, the decision as to when to design-in this affordance is determined by need, that is, when being able to execute passing shots and lobs becomes a significant requirement for success.

Option

What PRACTICE ACTIVITIES will bridge the gap?

Sport coaches will have a range of practice activities available to them to create learning environments that meet the goals of the session. The skill of coaching is identifying which activity is best matched to the needs of the performers. The traditional coaching session typically follows the same format. That is, a warm-up (usually non-related to the goals of the session), a decontextualised technical practice, followed by a generic game with a concluding non-related warm-down. The majority of these sessions result in the development of skills and abilities that lead to minimal transfer to the performance environment. We need to move away from this formulaic approach to choose tasks that will result in effective transfer. Coaches need to design sessions that spend the majority of time undertaking learning activities that improve performance effectively and efficiently. Our goal is to support this process.

Which PRACTICE ENVIRONMENT will you use?

Where a coach can run a practice session is a key constraint that impacts the design of coaching sessions. Learning and practice settings can include using the actual performance environment (completely, parts of and with or without modifications) as well as ‘other’ venues. It is important for the coach to identify all possible options and make informed decisions when designing the learning environment. Figure 6.2 provides an example of an environment selector for golf. This figure shows that the golf coach may be able to choose from a range of possible ‘off-course’ options such as the driving range, a putting green, a short game area, an indoor trackman ‘room’ or virtual golf simulator. Alternately, on-course options may be available, which might include the ‘competition venue’, similar courses, par three courses or just ‘courses’. A key decision for the coach is how important the representativeness of the environment is to meet the session goals. This can be understood with reference to the concept of affordances discussed above and in the theoretical chapters. The coach may therefore ‘slide’ up and down the ‘level of representativeness continuum within sessions.

Figure 6.2 An example of an environment selector for golf.

What constraints will you use and how will you manipulate them?

A primary requirement when implementing a CLA is to identify all possible constraints that can influence performance. To this end, we suggest that coaches should formulate their own table of constraints by separating them into individual (and team), environmental and task constraints (see Chapter 3). Once a comprehensive list has been created then the coach can design-in the ‘constraints to afford’ that will help meet the session intentions. Ideally, the choice of constraint is based on an understanding of its likely impact on performance. That is, we can use a constraint as a control parameter (see theoretical chapters) to create a re-organisation of co-ordination or use a range of ‘random’ constraints (e.g. different types and sizes of tennis racquets, different ball types in invasion games) to encourage exploration to solve task problems. However, given that empirical research on specific examples of control parameters or the effectiveness of adding in random constraints in specific sports is still rare, it is often up to the coach to use his or her experiential knowledge or intuitive thinking to select which constraints to design-in. The next key decision is when to add in the constraint, how long to leave it in for, or when to manipulate it, with the caveat of knowing ‘how much’ to manipulate it by. For example, a long jump coach who wants to improve an athlete’s vertical impulse at take-off may place a small hurdle 1 m in front of the take-off board that requires a higher knee lift to get over. Once the athlete has got this ‘feeling’ then the hurdle could be removed. The coach would then check that the ‘new’ behaviour was being maintained after the hurdle had been removed. This is an important point for all coaches, as any artificially ‘added’ constraint that is not present in the performance acts as an affordance to shape behaviour. We, therefore, should only leave the constraint in place long enough to create an established co-ordination pattern. Of course, this length of time may vary from individual to individual, but coaches will become better at making decisions about how long to leave a constraint in place through experiential knowledge. As such, choosing which constraints to use, how much to manipulate them by and how long to leave them in place is an exploratory learning process for the coach in the same way as we are encouraging in the performer.

Figure 6.3 An example of a constraints builder for golf.

How will you measure performance?

An often-overlooked aspect of coaching behaviour during practice is systematically and objectively measuring performance to move beyond the often-used intuitional level of ‘gut feel’ of the coach. What to measure is determined by the goals that are decided by coach and athlete(s). While many coaches below elite levels may not have access to support staff such as performance analysts, biomechanists, conditioning staff and sport psychologists who can objectively collect data, access to modern technologies (e.g. tablets, smart phones, drones) give the everyday coach access to ways of recording athlete behaviours and responses to session design. Additionally, injured players or parents can be asked to help. A good way to frame performance related to the set goals is to decide in advance of the session what a ‘successful’ outcome looks like. The following question frames this idea.

We will know we have been successful if

• We see . . .

• The data shows . . .

• The coach says . . .

• The performer feels . . .

To ascertain whether a coaching intervention has had any observable effect, a baseline is needed, preferably over a number of competitions, to ensure that a poor (or good) performance or sub-component was truly reflective of the ‘norm’ for performance. Determining in advance what is defined as ‘success’ in terms of a coaching intervention is necessary and can be framed in terms of ‘we see . . .’, ‘the data shows . . .’, ‘the coach says . . .’, ‘the performer feels . . .’ Table 6.1 provides some exemplar examples to illustrate these ideas.

Table 6.1 Exemplar for ‘success’ measures in CLA

It is worth cautioning here, that development is nonlinear and just because instant qualitative or quantitative changes cannot be observed, it does not mean that some learning has not taken place. Often coaches, need to be persistent but also patient in pursuing goals.

Way forward

Use of instructional constraints

How will you prepare the practice environment?

While still important, preparing the practice environment involves more than just setting out the equipment, writing up the session outline on the white board and paying overall attention to the mechanics of the session. It also refers to the creation of the cultural and social environment (cf. with the form of life as discussed in Chapter 4) that is developed and sustained by the coaching and support staff, and which athletes ‘buy into’. It is essential that coaches and support staff view themselves as a key part of the practice environment and recognise the significant impact of their communications (both verbal and non-verbal). It is, therefore, important for the coach to plan what to say at key moments of the session and to anticipate the likely reactions of individuals in response to coaching interventions. The coach should therefore carefully plan the language used and ensure the expectations, values and beliefs of the session are clear. An additional consideration for elite coaches who may have a coaching and support team is the significant danger that the messages given to the players are not consistent. The head coach must ensure that everyone is ‘singing from the same hymn sheet’.

How will you prepare the performers for the session?

An important consideration is how the coach will prepare the player’s mentally and physically in advance of the session, emphasising the intertwined relationship between cognitions, intentions perceptions and actions (see Chapter 2). To maximise the impact of the session, performers may need to be primed in terms of framing the task in the context of the bigger picture. That is, by framing and setting intentions and clarifying expectations. For example, it is essential that the performers are made aware if the main intention of the session is learning or performing. This links to creating a climate where ‘mistakes’ are framed as being part of a process of exploration and the overall goal of developing adaptability, or where the goal is clinical execution. Consequently, coaches may use pre-practice meetings with teams and individuals accompanied by the production of materials to support the delivery of the session. For example, highlights packages of previous competition footage, short notes to individuals, or perhaps referrals could be made to other media (e.g. podcasts, interviews and online videos). Overall, the coach needs to know that the performers are totally prepared for the upcoming session.

Is there anything else you need to do to be ready?

It is important for coaches to step back and consider if there are any other elements that they need to do to be ready. While this question can only be answered by the coach in specific contexts, we would encourage consideration of operational issues, consideration of the way the coach will prepare themselves for the session, as well as any ‘ancillary’ support such as bringing in opponents to ‘represent’ an opposition. Essentially, we suggest that the coach goes through a process of considering the ‘what ifs’ related to the session.

Step 2: putting the principles into practice

Now the ground work for the session design has been completed, prior to designing the session, the coach needs to develop two resources to simplify the process. As highlighted previously, session design is built on choosing the appropriate learning environment and the appropriate constraints to design-in. Below are the two resources for you to complete for your own coaching.

Design your own environment selector below:

Design your own constraint builder below:

We also provide these and templates of the tools used in this chapter in the Resources Section at the end of the book.

To help coaches frame the session, key features of a CLA session are summarised below. The aim is to provide a ‘one-page’ resource for coaches that visually brings the key concepts into one place and enables them to connect the key ideas, so they can begin to build the session.

As well as providing key information for each principle, the resource is also interactive as it contains a ‘constraints to afford’ section, which provides a place for coaches to decide on the constraints they wish to design-in to the session. While all three categories of constraints are considered, coaches should only pick the constraint (or interacting constraints) from the constraint builder that are mapped to the session intentions. Thinking about why and which constraints to choose is assumed as given once the coach ‘fills in’ this section, but the coach is also asked to consider when she will utilise the constraint. A key message is that manipulating constraints is the mechanism for creating variability and any subsequent (in)stability, therefore, answering this question should be undertaken alongside decisions about how much variability is going to be designed-in to the session. We will cover this shortly.

Step 3: completing the session planner

Below we provide a template that you may wish to use when designing your CLA session. The planner can be understood in terms of four vertical panels and the horizontal ‘tasks’, which are designed to cover key phases of the session. A key feature of the design is the introduction of dials to explicitly consider the level of representativeness and variability designed in to the different phases of the session. Essentially, the purpose of the dials is to encourage coaches to manipulate the levels of task representativeness and adaptive variability during different phases of the session by considering the affordances that are designed-in and manipulating constraints. The top panel focuses on the session goals and can be populated from the GROW process undertaken previously. Note that the session goals should link to the level of skill that connects to the ‘method’ which is effectively an expression of the intentions of the session (i.e. to explore to develop co-ordination, to learn to adapt, or to focus on executing the skill in the performance environment).

Figure 6.4 An overview of the key aspects of a CLA session.

The next panel refers to the level of representativeness of the practice task. While the ideas of representative learning design have been misconceived by some who suggest that adopting such ideas means that performers should only ever practise by playing the real game, this is not the case. Adopting the ideas of RLD means that the coach considers the performance environment and decides how much representativeness he or she wants in the practice task. Indeed, as discussed earlier, for very good reasons, he or she may decide that high levels of representativeness are not needed at a specific point in time, particularly when working with beginners. However, creating practice tasks that are low in representativeness does not mean that the key principles of a CLA are ignored and concepts such as perception–action coupling, co-adaptation and self-organisation are still essential requirements. The colouring of the dial to include a red and green zone is designed to emphasise this point with the ‘red zone’ representing any task where these principles are not considered. For example, a task would be placed in the red zone if performers were required to hit balls off a tee, when the sport requires learning to hit a moving ball. We would add, that learning to couple hitting to hit a moving ball is the key principle to underpin RD here and modifying tasks by using different balls and bats or racquets would still be classed as using a representative task, even though it may rate low on the scale. We will illustrate how these dials may be used to ‘turn-up’ or ‘turn-down’ the level of representativeness of practice within a session in examples in the practical chapters.

The third panel relates to the constraints that will be utilised in the session. In the constraints section, TIE, refers to task, individual and environmental constraints. Here, the coach simply puts a tick (check mark) alongside the category of constraints that will be used in the session and writes the name of the constraint along the top of the box. The separation into tasks is designed to allow the coach to modify the constraints within the session. Hence, Tasks 2–4, would reflect adding in, removal or manipulations of the designed in constraint. Of course, the session may have fewer or more than four phases; template lesson planners are provided for two, three and four tasks in the Resources section. Within each ‘constraints’ box, the coach could summarise the task or if she has a name for the practice task, simply write the name in here. The organisation section allows the coach to illustrate the layout of the task and any equipment needs.

Figure 6.5 The CLA session planner.

The final panel considers the level of variability that is to be designed-in to each task. Again, a dial is provided to invite coaches to manipulate the level of variability to match the current ability level of the performer during the session. The zones here are framed to reflect levels of variability that are likely to result in; (i) a strong likelihood of success in the task but wherever possible promote exploration and enhanced dexterity and hence adaptability (i.e. the green zone), (ii) instability that will require the leaner to search for new solutions (amber zone) and (ii) too much variability that will result in total failure (i.e. the red zone). Here, the coach will often need to make ‘best guesses’ as to which zone a task will be in, but over time and with systematic reflection should be able to become more accurate. When designing-in variability, it is recommended that the initial task sits in the green zone with a shift to the amber zone followed by a shift back into the green zone. This structure would enable the performer to feel confident and successful at the start of the session and leave the session in the same frame of mind (Renshaw et al., 2012).

Step 4: systematic reviews and reflections

Building systematic reviewing and reflection is an essential part of CLA coaching practice. Given that learning is a non-linear process, it is difficult to plan whole programmes in advance as the response of performers to coaching interventions can often be unpredictable, with sudden jumps in performance and indeed regressions being possible. A connected coaching programme is therefore emergent and co-created by the coach and the performers. Below we provide a debriefing resource that coaches can use to formalise this process. A five-step process framed around five questions is provided to learn from every aspect of the design and execution of the CLA practice session. Question 1 frames the reflection in terms of the original goals and expectations and pays specific attention to the affordances that were designed-in to the session. Question 2 invites the coach to reflect on what actually happened and breaks this into three aspects; the design of the session, the coaching and the actual learning that took place. Question 3 drills deeper into these areas and asks what went well and why. Here, the coach is asked to consider the learning design and the impact of what was said, with specific focus on the instructional and informational constraints provided during the session. Questions 4 and 5 focus on the future, with question 4 asking what can be improved and how, before Question 5 invites the coach to consider the most important findings from the reflection process by considering the three tings to continue doing and the three factors to change in future practice.

Figure 6.6 The CLA session de-briefer.

Summary

The last two chapters have focussed on ‘bridging the gap’ between the theory and practice of CLA with the aim of providing the tools to enable coaches to build their own CLA practice sessions. By following the guidelines provided, we hope that coaches will feel more confident in designing CLA sessions that are tailored to their own contexts and needs. In the final part of the book, we will attempt to bring CLA to life by providing examples of CLA sessions in athletics, invasion games (specifically in hockey) and golf. We will use real-life case studies to model the processes outlined in this chapter. We would point out that each case is not based on any one individual or group we have worked with, but the cases are often an amalgam of ideas.