If a tennis drill has 8.8 million views on Instagram, does that make it a good tennis drill?

That question has been sitting behind a lot of my work recently because social media has completely changed the way coaches discover practices. We scroll through Instagram, see a player moving beautifully, watch a drill that looks organised and impressive, save it to our phones and perhaps use it with our own players later that afternoon. The practice looks good, the player looks good and thousands, sometimes millions, of people appear to approve of it.

The problem is that none of those things actually tell us whether the practice is helping a player become better at playing tennis.

Over the past few years I have become increasingly interested in how we assess tennis practice design, particularly through the lens of Representative Learning Design and skill acquisition. I regularly get messages from coaches asking whether a particular tennis drill is good, whether I would use it or whether I think the skill being developed will transfer into competition. The more I have worked in this area, the less interested I have become in simply answering those questions with my opinion.

I don’t know your players. I don’t know what happened before the clip was filmed. I don’t know the problem the coach was trying to solve, and I certainly don’t know whether a practice that works with one player will work with another.

What I do think we can do is give coaches a better way of assessing practice for themselves.

That is ultimately why I developed my simplified Tennis Practice Assessment Scorecard, inspired by the Representative Practice Assessment Tool developed within the academic work of Ross Pinder and later applied specifically to tennis through Lyndon Krause’s research. Rather than me telling coaches what to do, the scorecard asks ten relatively straightforward questions that encourage you to look at the information available within your own practice and consider how closely it relates to the competitive game. 

The question isn’t really, “Is this a good drill?”

It is much more useful to ask:

“What is this practice developing, and how much potential does that learning have to transfer into the game?”

Practice Can Lie to Us

One of the reasons this matters so much is that tennis practice can be incredibly convincing.

Players can look brilliant.

They can hit ball after ball into the same target, reproduce an impressive movement repeatedly and become incredibly consistent within a particular drill. As coaches, that gives us something very tangible to look at. The player couldn’t do something at the start of the lesson, and forty minutes later they appear able to do it. It feels like evidence that our coaching has worked.

Then we start playing points.

Suddenly it disappears.

The player who had a beautiful forehand during the drill doesn’t move in the same way once another player is trying to beat them. The serve that looked fantastic from a basket falls apart when there is a returner. The player who could hit twenty consecutive balls into a target struggles to make the same decisions under score pressure.

I experienced this constantly throughout the first twenty years of my coaching career. Players would look excellent in practice and then appear completely different when I watched them compete. At the time, I would often explain the difference through confidence, concentration, mental toughness or simply the need for more repetition.

What I wasn’t looking closely enough at was the environment I had asked them to learn in.

The practice court and the match court contained completely different information.

That matters.

If the player spends most of the session receiving predictable balls from a coach, with no opponent, no meaningful score and very little consequence attached to their actions, they are learning to coordinate their behaviour around the information available within that environment. When we suddenly introduce an opponent, uncertainty, genuine consequences and score pressure, we haven’t simply made exactly the same task a little bit harder. We have changed the problem.

This is where Representative Learning Design becomes so useful for tennis coaches.

What Does Representative Practice Actually Mean?

Representative practice can sound unnecessarily academic, but the underlying idea is relatively straightforward.

If we want learning from the practice court to have the potential to transfer into competition, we should think carefully about whether the player has access to the information they will actually use when competing.

That does not mean every tennis lesson should simply consist of playing full matches.

The competitive game is probably the most representative environment available to us, but coaching exists because we often want to work on something more specific. A player might struggle with their serve, fail to recognise opportunities to attack, continually hit back through the middle of the court or become too passive when defending. The role of the coach is to take that problem and create an environment where the player has more opportunity to interact with it.

The challenge is doing that without stripping away so much information that we create an entirely different skill.

An opponent is an obvious example.

Take the opponent away and we lose information about their position, movement and behaviour. We also remove the consequences created by their response. If I hit an effective ball that puts my opponent under pressure, their next action will change and therefore the problem I face with my next shot changes too. If I hit an ineffective ball and give them a comfortable opportunity to attack me, the next situation changes in a completely different way.

That interaction is tennis.

Score matters for similar reasons. A player does not necessarily perceive and behave in exactly the same way at 40-love as they do at love-40. Score carries consequence, emotion and risk, all of which are part of competitive sport.

Variability matters because the game does not repeatedly deliver exactly the same ball to exactly the same location at exactly the same speed.

None of this means a practice has to contain every possible piece of match information. It means the coach should understand what has been removed and consider whether removing it helps or harms the purpose of the practice.

Assess the Practice, Not the Coach

This is one of the biggest misunderstandings I have experienced since I started publicly assessing tennis drills.

If I give a practice a low representativeness score, I am not saying the person who designed it is a bad tennis coach.

Equally, if I give a practice a high score, I am not declaring that the coach is brilliant.

Those are completely different questions.

I spent around ten years working in coach education and assessing coaches against different qualification frameworks, so I understand how quickly assessment can feel personal. Coaching is tied closely to identity. We invest enormous amounts of time developing our beliefs, our methods and the practices that we use, which means criticism of a drill can easily feel like criticism of us.

That isn’t what the assessment is designed to do.

The only question I am interested in when using my Practice Assessment Scorecard is something along the lines of:

How much relevant information from the competitive environment is available within this practice, and therefore what might its potential be for skill transfer?

There could be a perfectly legitimate reason why a coach has temporarily chosen a low-representation practice. The player may be returning from injury. The coach may want to bring attention to a particular sensation. The clip may show only thirty seconds of a much larger session.

Context matters.

The score simply tells us something about the practice in front of us. It doesn’t tell us whether the coach is good, bad, experienced, inexperienced or successful.

That distinction is important if we are going to have useful conversations about tennis practice design without immediately becoming defensive.

Why Traditional Tennis Practice Often Removes Information

Much of traditional tennis coaching has been built around a relatively logical-looking progression. We simplify the skill by stripping elements away, develop the movement in a controlled environment and then gradually add complexity back in.

The player might begin with a hand-fed ball, progress to a racket feed, then rally cooperatively, then eventually play points.

With the serve, we might stand a player on the baseline with a basket and ask them to repeat the service action into an empty box. Once the technique is polished, we introduce the returner.

The underlying assumption is that we can take the skill out of the game, polish it and then put it back in later.

Contemporary skill-acquisition research gives us good reason to question how straightforward that transfer really is. Representative Learning Design asks us to consider what happens when important sources of information are removed, because the movement that emerges is related to the environment in which the player is acting. 

This is where the difference between task decomposition and task simplification becomes really important.

If a beginner cannot yet cope with a full tennis court, the answer doesn’t automatically have to be removing the opponent and feeding balls to them. We can simplify the game while keeping important information.

We can reduce the court size.

We can use a slower ball.

We can change the scoring system.

We can enlarge a target.

We can reduce some of the variability.

What we are trying to avoid is unnecessarily removing the things that make tennis tennis.

A beginner does not need to play the same version of tennis as a professional player, but they can still interact with an opponent, use a racket, experience consequences and solve meaningful problems.

A Simple Serve Example

Imagine a player practising their serve at the end of a traditional lesson.

There is a basket beside them containing fifty balls. They stand at the baseline and repeatedly serve into an empty service box.

From a technical perspective, the practice might look fantastic.

From a representative perspective, however, we need to ask what information is missing.

There is no returner.

There is very little consequence beyond whether the ball lands inside the service box.

The next shot does not depend on what happened with the serve because there is no next shot.

The server does not have to detect the returner’s position.

They do not experience how the returner reacts to different serves.

The Practice Assessment Scorecard would highlight several of those things very quickly.

The answer is not necessarily to throw away the entire practice.

Perhaps the first change is simply placing somebody at the opposite end of the court.

Now the server can see a returner.

Their serve produces a consequence.

A strong serve might create a weak return. An ineffective serve might allow the returner to attack. The server begins receiving information about whether the action they selected was actually useful.

One relatively small change has altered the informational environment considerably.

This is exactly how I would encourage coaches to use the scorecard. You do not need to completely rebuild your coaching philosophy tomorrow morning. Download the scorecard, take one practice you already use and assess it honestly.

If one of the ten areas scores red, start there.

Why Coaches Find This So Difficult

There is another layer to all of this that has very little to do with tennis drills and a lot to do with us as human beings.

Path dependence is powerful.

Most coaches initially coach in ways that resemble how they were coached themselves. We complete qualifications that introduce particular ways of thinking, work alongside coaches who reinforce those ideas and gradually develop routines that become familiar.

Then confirmation bias kicks in.

We remember the players who succeeded.

We see a professional player doing a basket drill and use that as evidence that basket drills must be effective.

We see an internationally successful coach using an isolated technical practice and assume that the practice explains the player’s success.

The difficult question is whether we can actually establish that causal relationship.

Professional players also play enormous amounts of representative tennis. They compete. They practise against opponents. They play points. They experience years of unpredictable ball flight, changing environments, pressure and consequences.

Yet social media often shows us the five-minute technical exercise.

We then attribute the player’s expertise to the thing we can see.

That doesn’t mean the practice had no value. It means we should be cautious about claiming that because an elite player performs a drill, that drill produced their expertise.

The same applies to our own coaching.

“I’ve done this for thirty years” tells us that we have done something for thirty years.

It doesn’t automatically tell us that another practice couldn’t have been more effective.

That is why assessing practice is uncomfortable, and also why I think it is important.

You Don’t Need to Score Five out of Five

Another misconception is that every tennis practice should achieve a perfect representativeness score.

I don’t think that is realistic, and I am not even sure it would be desirable.

If a practice reproduced every feature of competition perfectly, we would probably just be playing a match.

Coaching usually involves amplifying something.

We manipulate a rule because we want a particular behaviour to emerge. We alter the dimensions of the court because we want players to become more sensitive to space. We use an orange ball with an adult beginner because we want to slow the environment down. We might occasionally use two balls or allow an unusual rule even though neither exists in formal tennis.

Those manipulations may reduce the representativeness of one particular piece of information while increasing the player’s opportunity to explore something important.

That is why the scorecard is a thinking tool rather than a judgement tool.

A score of three isn’t automatically bad.

A score of four isn’t automatically good.

The score should make you ask why.

What information is available here?

What have I removed?

Why have I removed it?

What behaviour am I hoping emerges?

What would happen if I changed one constraint?

Those are coaching questions.

How to Use the Tennis Practice Assessment Scorecard

The most useful way to use the scorecard is before you walk onto the court.

Take the practice you have planned and work through the ten questions as honestly as you can. Each question uses a simple red, amber and green system, giving you an immediate visual representation of where important information may be missing or inconsistent.

Nobody else needs to see your score.

That matters because the exercise becomes useless if we simply give ourselves green on everything to prove that our coaching is already brilliant.

If you find a red, resist the temptation to redesign the entire session.

Make one adjustment.

Perhaps your practice has no consequence, so introduce a simple scoring system.

Perhaps the player never receives information from an opponent, so find a way to include one.

Perhaps every feed is identical, so create slightly more variability.

Perhaps players cannot choose between different actions, so loosen the rules enough for more than one solution to become available.

Take that version onto court and watch what happens.

The players may respond exactly as you hoped.

They may also respond in a completely unexpected way.

Good.

Now you have new information as a coach.

Adapt again.

That is much more useful than treating a practice plan as something fixed that has to be delivered exactly as it appeared on paper.

A Higher Score Does Not Guarantee Learning

This point needs to be made clearly because I do not want Representative Learning Design to become the next simplistic tennis coaching slogan.

A highly representative practice does not guarantee learning.

Human learning is complex.

Players are different. Their experiences are different. Their abilities to detect information are different, and the problem that one player needs to solve may be completely irrelevant to somebody standing beside them.

What greater representativeness gives us is greater potential for learning to relate to the context in which we ultimately want the player to perform. That is a very different claim from saying that a particular score guarantees transfer. The transcript behind this article repeatedly makes that distinction: representativeness is being used as a lens for considering transfer potential, not as a promise that learning will automatically occur. 

The coach still matters.

Observation matters.

Individualisation matters.

Practice design matters.

The scorecard simply gives us a framework within which to think about those decisions.

What Are We Actually Coaching For?

This may be the most important question in the whole discussion.

If we are coaching tennis, surely somewhere along the line the player wants to become better at playing tennis.

That does not mean everybody needs to play tournaments.

A beginner might simply want to play points against their partner.

A recreational adult might want to enjoy a club doubles match once a week.

A junior might want to compete nationally.

The competitive level is different, but the central problem is still interaction with another person across a net.

If we spend enormous parts of their coaching time developing behaviours that only appear when the opponent, uncertainty, consequence and score have been removed, we should at least ask how easily those behaviours will survive when all of that information returns.

That is the conversation I want coaches to have.

Not whether basket feeding is evil.

Not whether one coaching philosophy is morally superior to another.

Not whether somebody deserves to be called a good or bad coach.

The interesting question is much simpler:

Does the information available in my practice help prepare the player for the information they will experience when they actually play tennis?

Once we begin looking at practice through that lens, practice design becomes much more interesting.

Assess Your Own Tennis Practice

The easiest place to start is not by changing everything.

Take one drill.

Preferably choose one you use regularly and one you believe works particularly well.

Download my free Tennis Practice Assessment Scorecard, work through the ten questions and score the practice honestly using the red, amber and green system.

Then look for one area you could change.

Move it from red to amber.

Take the practice onto court.

Observe.

Reflect.

Adapt.

You may ultimately decide that the original version was better for that particular player and that particular problem. That is absolutely fine. The important change is that you now have a clearer reason for your decision rather than simply using the practice because you have always used it.

The scorecard also comes with a free mini-course where I assess two very different practices myself, including one that scores relatively poorly and another that scores much higher. The idea isn’t to give you another collection of drills to copy. It is to help you develop the ability to assess, redesign and adapt your own practices.

Because the future of tennis coaching probably isn’t going to be about who owns the biggest collection of drills.

It is going to be about who understands why a practice is designed the way it is, what information it provides, what behaviour it invites and whether what players learn has a realistic opportunity to survive when they step onto the match court.

And perhaps that is the question we should have been asking all along.

[Download the Free Tennis Practice Assessment Scorecard] and watch the video podcast below

Favorite