The Missing Manual for Constraint Design

You have read about the constraints-led approach, maybe even taken a course on it, but when you are standing on the field designing a session, do you actually know which constraints to choose and why? In this episode of Coaching Research To Results, your host, Beth Barz, digs into new research calling for more practical, research-informed guidelines for constraints-led coaching. If you coach using games-based or constraints-led methods and want more confidence in your constraint design, error recognition, and session planning, this episode is for you. Find out what the research says is still missing for coaches and how to start building your own applied playbook.

https://thecoachdeveloper.com/coaching-research-to-results-podcast-notes/

Transcript

SPEAKER_00 0:00

Here's a question I want you to think about for a minute. Have you ever read about constraints-led coaching, or been on a course about it, or heard a keynote about it, and walked away thinking, okay, that all makes perfect sense. And then you stood on the training pitch and thought, so what exactly do I do? You changed the field size, you added a rule, you took a player out, and you watched what happens. And sometimes it was brilliant, and sometimes it wasn't. And you weren't entirely sure why, in either case. If that sounds familiar, you are not the problem. Today's research has something important to say about that uncertainty, and it might make you feel considerably better about the fact that you've been improvising. Welcome to Coaching Research to Results. I'm Beth Barts, the coach developer. This is one paper, three ideas, and two actions for tomorrow in under 15 minutes. This is the podcast where coaching research gets off the shelf, out of your notes, and into your practice. Let's go! The paper is called Adding Texture to the Art of Constraints Led Coaching, a request for more research-informed guidelines, published in Sports Coaching Review in August 2024. The lead author is Andrew Kennedy from the School of Human Movement and Nutrition Sciences at the University of Queensland, and his co-author is Catherine O'Brien from Queensland University of Technology. Both are researchers who work at the intersection of ecological dynamics, skill acquisition, and sports coaching. This paper grew directly out of a gap they kept seeing in the constraints-led literature. Coaches are told that the approach works, and broadly it does, but the research has been slow to answer the specific applied questions that coaches face every single day on the training ground. Kennedy and O'Brien sat down to name those questions out loud and to ask the research community to start answering them. The big idea could be called the missing manual. Here's the situation. The constraints led approach, the idea that coaches should shape the training environment rather than dictate movement solutions, is genuinely compelling, and the research base behind it is growing. Coaches are supposed to become what the literature calls ecological designers, people who adjust task constraints, environmental constraints, and individual constraints to allow athletes to find their own best movement solutions. The three constraint categories, by the way, are worth naming clearly because they're the tools you're working with. Task constraints are the rules, the scoring system, the equipment, the size of the space, the number of opponents, the time pressure. Environmental constraints are things like the surface, the weather, the crowd noise, the physical layout of the training facility. Individual constraints are the athlete's physical characteristics, fitness and fatigue levels, emotional state, prior experience, and skill level. All three interact constantly. Now Kennedy and O'Brien's argument is this: the theory is excellent, but the practical guidance hasn't kept pace. The current advice for coaches amounts to manipulate constraints, watch what happens, reflect, and adjust, which is fine as a starting point. But it's not a manual. It's more like telling someone to just cook something and then expecting a gourmet meal. The constraints-led approach tells you what to cook, but it doesn't give you the recipe. So here's the question for your own coaching. When you design a session using modified games or constraint adjustments, how confident are you that you know why you're choosing those specific constraints? And if it's not working, do you know what to change? Idea two could be called noise or signal. Every athlete makes errors. A swimmer who keeps dropping their elbow in the catch phase, a football player who looks down when receiving the ball, a basketball player who tends to jump before looking to pass. Under traditional coaching, these errors are things to correct directly. You see it, you name it, you fix it. Under the constraint slot approach, errors are reframed as productive instabilities. The athlete is working out the movement problem, and errors are part of that success. The coach is encouraged to hold back and let the athlete self-organize. And there's real wisdom in that. Overcoaching errors can make athletes dependent, anxious, and mechanical. But here's what Kennedy and O'Brien point out. The CLA literature is very quiet about what happens when the error doesn't resolve itself. When is an error just a natural variation that will settle down on its own? And when is it a sign that the constraint design isn't working or that the athlete genuinely needs some form of guidance? Right now, coaches are left to figure that out alone. The research hasn't given them a decision framework. Think about a field hockey example, a player who dribbles with their head down and can't see their teammates or the space around them. Should the coach wait? Should they redesign the task so the player is forced to look up? Should they say something? The CLA doesn't give you a clear answer. That's the gap. Idea three could be called bad habit in a good disguise. Related to errors but distinct from them, sometimes athletes develop motor patterns that look like adaptation but are quietly causing problems. The constraints-led framework calls these undesired skills, and it does acknowledge that they exist. But again, it's not very clear about what to do with these errors. The reason that this matters is that once a movement habit forms, it's very difficult to change. And the process of self-organization that the CLA relies on can strengthen bad habits just as easily as good ones. The athlete is self-organizing. They're just organizing themselves in the wrong direction. The basketball example from the paper is a good one. A player who jumps first and then looks to pass is making a movement that sometimes works. They might complete the pass successfully. So, by the framework's logic, it could be interpreted as a functionally adaptive response to defensive pressure. But more often it leads to turnovers, and the longer it's left, the more deeply grooved it becomes. Here's what's genuinely tricky about this. To tell the difference between the athlete is still finding the solution and this is actually becoming a harmful habit, a coach needs to know knowledge of the sport, the athlete's developmental stage, and the specific performance context. The CLE research doesn't yet offer guidance on how to make that call, or critically, a list of constraint manipulations that might specifically help unwind it. The constraints-led approach says trust the process, but what do you do when the process is producing the wrong answer? Now, let's bring in some contrasting evidence. The critique in today's paper could leave you thinking that the constraints-led approach doesn't offer coaches any practical tools at all, and that's not quite fair. Renshaw, Davids, Newcomb, and Roberts, whose 2019 book is essentially the practitioner bible for the CLA, developed the Grow Model, a structured reflection process that asks coaches to identify their goals, the reality of the athlete's current level, the options available in terms of constraints to manipulate, and the way forward. It's a genuinely useful framework. So the counterpoint is this: the tools exist, the grow model works. Representative learning designs works. The question Kennedy and O'Brien are raising isn't that nothing is available, it's that what's available doesn't go far enough for the specific, messy, real-time decisions coaches have to make. Recently, I was working with a coach who was relatively new to the constraints-led approach. He came to me frustrated about what he perceived to be ineffective use of it. He'd been to a CLA workshop and then he'd read a bunch of resources and genuinely wanted to implement in his practices. But when he sat down to plan his next session, he had a list of things he could theoretically constrain, but no way to know which constraints to choose. He related, I know the theory, but I just don't know how it translates to the session plan. So I helped him build up a menu of constraints, that is, a personal inventory of task, environmental, and personal constraint options organized by the outcomes each constraint was likely to produce. Then we did a brief analysis of his last game and identified two main areas to work on in the next practice. Based on the outcomes we wanted to chase, we went back to the constraint menu and selected two constraints to help direct the design of the game. In this case, we changed the physical environment by shrinking the size of the playing area to force more pressure on the attacking team, and we slightly changed the task by forcing attackers to score from a particular area. This forced the attackers to work out a new way to get into that area before being able to score. By all reports, the practice went well. It was just connecting the step between knowing the outcome and choosing the right constraint to chase the outcome that the coach was looking for. Your two actions for CLA are Action one, build your constraint menu. The paper's core argument is that coaches currently have to guess which constraints to use. Here's an action that begins to change that. Before your next session, take 15 minutes and write out a constraint menu. Not for the specific session, but for your sport in general. Organize it under the three headings from the framework: task constraints, environmental constraints, individual constraints. I've provided a list of constraint examples in the show notes that you can adapt to your sport and context. Your goal isn't to use all these today. Your goal is to have a searchable list you can go to when you notice something in a session and you want a constraint to shape it. Once the list exists, you can start tagging which constraints you've actually tried and what happened. Over time, that becomes your personal evidence based. Action two is to run a constraint audit on one session. At the end of the session, take five minutes to write down answers to three questions. What constraints did I use today, categorized as task, environmental, or individual? Two, was there a moment where I felt the need to intervene verbally because the constraint wasn't producing what I wanted? And three, if I could change one constraint for the next session, which one would it be and why? This is about building the reflective habit of noticing what you're actually doing and when you're designing constraint environments. The paper's argument is that the coaches need more applied guidance, and part of building that guidance is coaches like you documenting your own applied practice. So let's bring it back together. Kennedy and O'Brien aren't asking you to abandon the constraints-led approach. They love it. But they're naming three real problems the research hasn't fully solved. Coaches need better guidance on reading errors, on recognizing harmful habits before they become permanent, and on selecting constraints intelligently rather than just guessing. The big idea today was the missing manual. The constraints led approach is a genuinely powerful framework, but right now it asks coaches to improvise in situations where a bit more structure would go a long way. Here's the sharpest thing I want you to take away. The environment is the teacher, but someone still has to design the classroom. And designing it well is a skill that needs its own manual. That manual is still being written. Today's paper is one of the voices asking for it to be finished. I'm Beth Barts, the Coach Developer. You'll find show notes and the full library of episodes at www.thecoachdeveloper.com. If any idea from today lands for you, like, subscribe, and share it with another coach this week. That's how research actually travels. See you next time. This podcast was produced by Ann Reifenstein at RECERA services. Original music created and recorded by Sean Patterson and Vinyl Safari Studio.