Carousel Coaching: Efficient or Effective?

The question is not simply whether players rotate. It is whether the learning travels with them.

Carousel coaching is easy to understand. Divide the available space into stations, allocate a coach to each area, split the players into groups and rotate them at predetermined intervals. From an organisational perspective, it can work extremely well. Large numbers become manageable, coaches have defined responsibilities, facilities can be maximised and players remain active. In grassroots football, camps, introductory programmes, pre-season testing or sessions where participation and exposure are the priority, those benefits should not be dismissed.

My concern begins when the carousel moves beyond being an organisational solution and becomes the default model for long-term player development.

If we describe an environment as developmental, I think we have to ask more of the session than whether the players were active, engaged and exposed to several good practices. We should be able to explain what they were learning, how one part of the session supported the next, how that learning connects to the wider curriculum and, ultimately, how it contributes to the football we are trying to develop.

That is where I increasingly favour deliberate, connected session design over the traditional carousel.

The case for carousels

There is an obvious danger in criticising a coaching format without recognising the problem it may have been designed to solve.

Space is finite. Coach-to-player ratios vary. Attendance changes. Some organisations have several coaches available simultaneously. Grassroots coaches may have one session each week with a large group and limited facilities. In those circumstances, dividing the environment into smaller working areas can increase involvement and make delivery considerably easier.

Rotation can also introduce practice variability. Motor-learning research has long examined contextual interference: broadly, whether repeatedly practising the same task or varying the order and nature of practice influences retention and transfer. Evidence suggests that more variable or random practice can sometimes benefit later learning even when immediate practice performance looks worse (Brady, 2004; Farrow & Buszard, 2017). However, the effects found in controlled research are not always reproduced as strongly in applied sport settings, particularly with younger performers (Brady, 2008; Czyż et al., 2024a).

That nuance matters. Variety is not inherently bad. Neither is changing the problem players encounter.

The mistake would be to conclude that because variability can support learning, a collection of unrelated activities automatically constitutes a development programme.

It does not.

A common theme is not a connected session

Imagine three coaches working simultaneously under the same weekly sub-principle.

All three might genuinely be coaching the same broad idea. Yet Coach A could interpret it through an unopposed technical practice, Coach B through a possession activity and Coach C through a directional game.

On paper, everything aligns.

But what does the player experience?

If each rotation introduces different terminology, different information, different conditions and a different interpretation of the objective, the player may effectively restart the learning process every fifteen or twenty minutes.

That is the distinction I think matters:

A common theme does not necessarily create a connected learning journey.

The individual practices may all be excellent. The individual coaches may all be excellent. The problem sits above them: session architecture.

Who owns the learning journey across the whole session?

Scaffolding rather than collecting activities

For me, a session should have a thread.

That does not mean every activity has to look the same or that football should become scripted. It means the activities should have a deliberate relationship with one another.

If the objective is, for example, to create and exploit an overload, the session might initially help players recognise the information that indicates an overload is available. The next practice might increase opposition and require players to create that advantage. Later, the environment may become increasingly representative and ask players to exploit it under greater pressure.

Eventually, much of the coach support can disappear.

The football problem remains; the scaffolding changes.

This is important because skill acquisition is not simply about reproducing actions. Practice design should help performers become attuned to relevant information and adapt their actions to changing situations. Contemporary constraints-led approaches therefore emphasise manipulating task, environmental and individual constraints rather than prescribing a single solution (Dehghansai et al., 2020).

Representative practice is also concerned with maintaining meaningful relationships between the information players perceive and the actions available to them (Pinder et al., 2011). Research into perceptual-cognitive skill development has similarly emphasised the importance of similarity between training information and the competitive environment when transfer is the objective (Broadbent et al., 2015b).

So rather than asking:

What activity comes next?

I would rather ask:

What does the player need to understand next?

That subtle change can fundamentally alter session design.

Progression does not mean changing the exercise

This is another area where I think coaching can become unnecessarily activity-driven.

We sometimes demonstrate progression by changing the practice.

But learning can progress while the basic activity remains intact.

We can manipulate space. Change numerical relationships. Alter opposition behaviour. Adjust starting positions. Change the scoring system. Remove a condition. Introduce another source of information. Give players greater autonomy.

The football problem remains recognisable while the level of challenge evolves.

This is close to the idea of repetition without repetition: players repeatedly solve a related problem without simply reproducing an identical action. Ecological approaches to practice design explicitly advocate representative environments that expose performers to variation while preserving the information that matters to performance (Pinder et al., 2011).

That seems particularly relevant in football.

Matches do not present the same picture repeatedly. Players need adaptable solutions.

But adaptable does not mean disconnected.

The final game should not abandon the session

This is perhaps where I have the strongest preference.

A common structure is to complete the practices and then finish with a normal game.

There is nothing inherently wrong with playing football. Nor do I believe the final game needs to be overloaded with rules until the behaviour becomes artificial.

But if we have deliberately spent the preceding part of the session developing a particular behaviour, I struggle with the idea that the most game-like part of the session should suddenly become detached from that learning objective.

There is considerable space between an over-conditioned game and simply saying:

“Off you go. Play.”

Suppose the objective has been creating and exploiting overloads.

The game can remain largely normal while giving additional value to a goal that follows the successful creation and exploitation of an overload.

Or perhaps the opposition's restart is manipulated.

Perhaps the pitch dimensions encourage the relevant problem.

Perhaps no additional condition is required at all, but the coaches have agreed beforehand exactly what behaviours they will observe and review.

The important point is that the learning objective remains alive.

Player autonomy should increase as the session progresses. Coach intervention may decrease. Conditions may disappear.

But player freedom does not require the learning objective to disappear with them.

Poorly structured game-play without specific goals has itself been criticised in skill-acquisition literature alongside the opposite extreme of repetitive practice divorced from game context (Farrow & Buszard, 2017).

For me, the final game should therefore provide an opportunity to showcase, test or evaluate the learning, not simply reward the players for completing the training.

Where does the session sit in the bigger picture?

The discussion becomes more important when we move from an individual session to long-term development.

A session should not exist in isolation.

Ideally, I want to see a recognisable relationship:

Game Model → Coaching Curriculum → Principle → Sub-principle → Development Block → Weekly Objective → Session → Practice → Transfer → Reflection

Not every organisation will use those exact terms. That is not important.

The connection is.

If we claim to have a game model but it does not influence the curriculum, or a curriculum that does not meaningfully influence session design, we risk creating impressive documents above the pitch and disconnected activities on it.

Long-term athlete development is inherently complex. Development emerges from interacting individual, task and environmental factors across different timescales rather than through a simple linear pathway (Dehghansai et al., 2020).

That should arguably make joined-up thinking more important, not less.

Where contact time is limited, every session becomes particularly valuable.

We should be able to ask:

What did we observe today?

What evidence suggests players understood the problem?

What needs further development?

How does that inform what comes next?

That creates a cycle:

Plan → Deliver → Observe → Reflect → Adapt → Plan again

Without that connection, a programme can easily become:

Theme → Activities → Game → Home → New theme.

Those two models may look remarkably similar from the touchline.

Developmentally, they are very different.

Could a carousel still achieve this?

Yes.

And this is why I would not argue that carousel coaching should simply be removed from football.

A carousel could be designed around one connected problem.

For example:

Station 1 — Recognise

What information tells the player the opportunity exists?

Station 2 — Create

What movement or action manipulates the opposition and creates the advantage?

Station 3 — Exploit

Can the team use the advantage before the opposition recovers?

Game — Transfer

Can those behaviours emerge within the game?

The players still rotate.

The coaches still have individual areas.

The organisational advantages remain.

But now every coach understands what came before their station and what should come afterwards. Language can be aligned. Observations can be shared. The final game can provide evidence of transfer.

The carousel has stopped being a collection of activities and become a scaffolded session delivered across multiple spaces.

That distinction is significant.

What about variety?

There is another argument in favour of carousel structures: players may benefit from having to adapt continually to different tasks.

There is evidence behind that principle. A recent systematic review and meta-analysis found an overall transfer benefit associated with higher contextual interference, although the effect in applied settings was smaller and statistically non-significant, and negligible in younger participants (Czyż et al., 2024a). Another recent meta-analysis found benefits for retention, particularly in laboratory conditions (Czyż et al., 2024b).

That should make us cautious about simplistic conclusions in either direction.

Blocked, repetitive practice is not automatically superior because it looks organised.

Random, variable practice is not automatically superior because it looks representative.

And a carousel is not automatically developmentally effective because it exposes players to multiple activities.

The appropriate question is:

What type of practice best serves this player, this problem and this point in the learning process?

That requires coaching judgement rather than allegiance to a format.

My preference: deliberate before different

My own preference remains clear.

Where the objective is long-term player development, I would rather see fewer disconnected experiences and more deliberate connections.

I want coaches to understand the wider game model.

I want the weekly objective to mean something.

I want the physical preparation to connect where appropriate with the football demands that follow.

I want practices to build upon one another rather than merely occupy consecutive time slots.

I want coaches to know what their colleagues are developing.

I want the final game to provide evidence about whether learning has transferred.

And I want what we observe today to influence what we design tomorrow.

None of that requires a particular coaching methodology.

It requires joined-up thinking.

A carousel can potentially deliver it.

But rotation alone certainly does not.

The question worth carrying forward

Perhaps, then, the debate is not:

Should we use carousels?

It is:

What are we trying to develop, and does the architecture of the session actually support it?

If a carousel provides the best organisational solution and every station contributes deliberately to a coherent learning journey, there is a strong case for using it.

If players simply rotate through independently designed activities before finishing with an unrelated game, I think we should be willing to ask whether efficiency has been mistaken for development.

Because ultimately, the objective is not to fill every quarter of the pitch.

It is to develop the player.