Search Google for how to improve tennis footwork and you will normally find a familiar collection of advice. Practise your split step. Work on crossover steps. Use adjustment steps. Recover to the middle. Run through an agility ladder. Put some cones on the court and rehearse different movement patterns.
I used to think about tennis movement in much the same way.
The assumption is logical. Tennis players move their feet, so if we want better footwork, surely we should train the feet. Teach the correct movement patterns first, rehearse them until they become automatic and then bring those movements back into the game.
The problem is that tennis does not ask players to reproduce predetermined footwork patterns.
Tennis is an incredibly variable game. The ball can arrive at different heights, speeds, depths, widths and angles. It can spin, skid, bounce high or stay low. Players themselves have different heights, strengths, speeds, movement capacities and levels of experience. Then add different court surfaces, weather conditions, opponents and tactical intentions.
Against all that variability, the idea that good tennis footwork can be reduced to a collection of universal movement patterns starts to become difficult to defend.
That does not mean the split step is useless. It does not mean crossover steps are bad. It does not mean strength, speed or agility training has no place in tennis.
It means we may be starting in the wrong place.
Good tennis movement does not start with the feet.
It starts with information.
What actually is good tennis footwork?
I no longer think good footwork should primarily be judged by whether a player performs a particular sequence of steps.
The more useful question is whether their movement helps them solve the problem in front of them.
At its simplest, tennis involves trying to put the ball somewhere that makes it difficult for the opponent to return while maintaining enough stability to deal with whatever comes back. Movement helps us get to the ball, organise an effective action and position ourselves for what might happen next.
What that looks like is completely context dependent.
A beginner playing with a slower ball on a smaller court does not face the same movement problem as a professional player returning a 130 mph serve. Their absolute speeds are completely different, but both players are trying to regulate their movement relative to the demands of their own playing environment and their own action capabilities.
This is why I am increasingly uncomfortable when coaches talk about one version of “correct footwork.”
The real question should be: does this movement solution help this player solve this problem?
Your feet are often the end of the problem, not the beginning
When a player arrives late to the ball, gets jammed, loses balance or fails to reach a wide shot, coaches often immediately diagnose “bad footwork.”
But what we see at the feet may simply be the visible symptom of something that happened much earlier.
Imagine a beginner waiting for the ball to bounce before they start moving. Once the ball has bounced, they suddenly realise it is travelling away from them. Now they have very little time available. They rush, reach, lose their balance and eventually miss.
We could tell them to move their feet faster.
But faster feet may not actually solve the underlying problem.
They may need to become sensitive to information earlier.
Initially, that information might simply come from the flight of the ball: its direction, speed, height and trajectory. With experience, players can become increasingly sensitive to other information, including their opponent’s position and movement, what happened on the previous shot and the opportunities their own shot has just created.
Research in tennis has repeatedly shown that skilled performance involves more than simply reacting once the ball is travelling towards us. Skilled players differ from less-skilled performers in how effectively they use perceptual information, and perceptual training has been shown to improve anticipation in tennis players. At the same time, we have to be careful with some of this literature because much anticipation research has historically used experimental tasks that do not fully represent what actually happens on a tennis court. A systematic review by Avilés and colleagues found that only eight of 62 studies they examined met their criteria for a high degree of task representativeness.
That matters because perception and movement belong together.
Players do not simply perceive the world, make a decision somewhere inside their head and then instruct their feet what to do. Movement itself helps players pick up information. As we move, the relationship between ourselves, the ball, the opponent and the environment continually changes.
This is why I would rather think about tennis movement as a perception-action problem than simply a footwork problem.
Why ladders and cones are not the same as tennis footwork
This is where some traditional footwork training becomes questionable.
Agility ladders, cones, shadow movements and predetermined footwork cycles can certainly make players move. They may also have value for developing physical capacities such as coordination, rhythm, speed or general agility.
But that is not automatically the same thing as developing tennis movement skill.
Think about the information available when somebody runs through an agility ladder. The pattern is known in advance. The ladder is stationary. The information being used is close to the player and highly predictable. In many exercises, the player is looking down towards the ladder while reproducing a predetermined sequence.
Now compare that with tennis.
The player is interacting across a much larger area. The ball is moving. Their opponent is moving. The ball may bounce differently each time. The player does not know exactly where they will have to move next, how far they will need to travel or what movement solution will be required when they arrive.
Those are very different information environments.
Research outside tennis demonstrates why this distinction matters. In cricket, Pinder and colleagues found that changing the informational source from a live bowler to a bowling machine altered young players’ timing, step length and movement coordination, even when delivery speed was controlled. It does not prove that ladders or ball machines are useless, but it illustrates a broader principle: change the information and you can change the movement that emerges.
So I would not say agility ladders have no value. I would ask a different question.
What are we trying to develop?
If the goal is coordination, speed or a physical capacity, perhaps they have a role. If the claim is that running through a ladder will automatically improve tennis-specific footwork, I think we need to be much more cautious.
Transfer should not be assumed simply because an exercise looks athletic.
What about the split step?
The split step is probably the most coached movement pattern in tennis.
We teach it to performance players, recreational adults and sometimes children who are four or five years old. Players jump over lines, off steps or around cones. Coaches shout “split!” before feeding the next ball. Sometimes the player is asked to call “split” when their opponent makes contact.
There is a good reason the split step receives attention.
Research suggests that it can help players initiate lateral movement more quickly. In a simulated tennis task, Uzu and colleagues found that collegiate players reached the target faster when using a split step than without one. Crucially, however, its effectiveness also depended on the timing of the landing.
That timing point is important.
Studies of higher-level tennis players show that the split step is not simply a fixed technical action performed identically every time. Its timing changes in relation to the unfolding game situation. Filipčič and colleagues found differences in split-step timing across different situations, while work with world-class players found that the action was organised around racket-ball contact and early ball-flight information. The timing of subsequent movement was also related to players’ neuromuscular capacities.
In another representative study of expert players returning first serves, Navia and colleagues found that players scaled the timing of their return movement to the server’s unfolding action.
That gives us a very different picture from “everyone must learn to split step.”
The split step can be a highly functional solution when the task requires rapid movement in uncertain directions. But its value depends on the movement problem.
Consider red-stage tennis. The court is small, the ball is slow and many players can cover most of the relevant space within a few steps. A child may rarely face a movement problem that requires the same preparatory action as an elite player returning serve on a full-sized court.
Yet we sometimes teach the split step simply because we believe it is a fundamental of tennis.
Instead of asking, “Has my player learned the split step?”, I would rather ask, “Does the game currently require it?”
When the movement becomes genuinely functional, we may see it emerge anyway.
Stop turning tennis players into dancers
Another movement pattern I see regularly, particularly in junior tennis, is what I jokingly call the crab dance.
The player hits, shuffle-steps back towards the middle, split steps, shuffles out again, hits and repeats. The movement can look beautifully organised.
It can also look nothing like the movement demands of the point.
Humans can run. A running step will often cover more distance than a shuffle. A crossover movement may allow a player to travel quickly across a larger space. Sometimes very small adjustment steps are useful. Sometimes the player hardly needs to move at all.
The problem appears when descriptive movement categories become a choreography that players believe they must reproduce.
Tennis is too variable for that.
Even two apparently similar balls are not identical. The player’s starting position is slightly different. Their momentum is different. Their opponent is positioned differently. The previous shot has changed the available space. The player’s intention may also have changed.
There will obviously be recurring movement solutions because we share broadly similar human anatomy and are performing the same sport. Elite players will therefore show common movement patterns.
But common does not necessarily mean compulsory.
Rather than trying to make every player move identically, I want to develop adaptable movers.
Recovery is not simply “get back to the middle”
Recovery position is another example of a useful coaching rule becoming too rigid.
We frequently tell players to hit the ball and recover towards the middle.
Sometimes that is perfectly sensible.
But the middle of the court is not automatically the best place to stand.
Every shot a player hits changes what is available next. In ecological dynamics we might describe these as nested affordances, but the practical idea is straightforward: your previous action creates the next set of opportunities and problems.
Imagine I hit a deep cross-court ball that stretches my opponent. Depending on their position and capabilities, a down-the-line reply may become difficult. I may not need to run immediately back towards the centre. Staying slightly wider could leave me better positioned for the more likely cross-court response.
If I have genuinely destabilised my opponent, I may need to move forwards rather than sideways because I have created the possibility of a shorter ball.
Now imagine I leave a short ball through the middle. My opponent may have considerably more freedom. They could attack into either side of the court. Under those circumstances, recovering towards a more central position may become much more functional.
There is no single recovery rule that captures all of these situations.
Players have to become sensitive to what their previous shot has done.
Is my opponent balanced or stretched? Have I created an attacking opportunity? Have I created a problem for myself? Which parts of the court can they realistically access next?
Once we ask those questions, recovery stops being a footwork routine and becomes part of playing tennis.
How would I actually improve a player’s movement?
Suppose I have a player who struggles when their opponent sends the ball towards the outside of the court.
One approach would be to stand with a basket and repeatedly feed balls towards the corners. The player can rehearse the movement pattern and get lots of repetitions.
I would probably approach it differently.
I might make the playable court wider by bringing the tramlines into play. That immediately increases the amount of space the player needs to defend.
However, simply making the task harder may overwhelm them, so I could simultaneously slow the ball down. Instead of a yellow ball, perhaps we use an orange ball.
Now two constraints are working together.
The wider court increases the spatial demand and makes larger movements necessary. The slower ball gives the player more time to pick up information and organise their action.
Quite often you will then start to see different movement behaviours emerge. A player who previously used small adjustment steps may begin crossing into a run because the task now requires them to cover more distance. A preparatory movement resembling a split step may appear because the player increasingly needs to move quickly in either direction.
I have not told them exactly which steps to take.
I have changed the problem.
The same principle can be used in many different ways. If I want a returner to develop their ability to cover more space, I might deliberately start them away from their normal returning position. If I want players to explore different recovery positions, I might change the geometry of the court or scoring conditions.
This does not mean the coach disappears.
If a player continually fails to find an effective solution, I can alter the constraints again. I can ask a question. I can demonstrate something. I can explicitly draw their attention towards useful information.
Ecological coaching is not simply setting up a game and hoping learning magically happens.
The skill is designing a task in which the information and movement problem invite useful exploration.
Develop adaptable movers, not perfect footwork patterns
One of the easiest ways to increase movement variability is simply to let players experience different versions of tennis.
Play on different surfaces when possible. Change the court dimensions. Make the court wider or narrower. Use different balls. An orange ball, green ball, new yellow ball and old yellow ball all behave differently.
Every change alters the movement problem.
A slower ball gives the player more time. A wider court changes the distance that needs to be covered. A different surface changes bounce and movement demands. Different opponents generate different speeds, angles and intentions.
That variability gives the player opportunities to discover different movement solutions.
This is much closer to what I mean when I talk about developing adaptability.
I am not arguing that anything goes. Some movement solutions will clearly be more effective than others for a particular task. Physical and biomechanical constraints matter. Humans cannot move in infinitely many ways.
But I do not believe the coach needs to prescribe one movement solution in advance simply because that is what an elite player appears to do.
The goal is not perfect footwork.
The goal is functional movement that can adapt.
Where does strength and conditioning fit?
This is an area I have discussed a lot with Howard Green from Tennis Super Movers.
There is an important distinction between improving someone’s capacity to move and improving their skill in regulating movement during tennis.
Strength and conditioning can increase action capacity. A player can become stronger, faster, more powerful or better able to decelerate and accelerate. That can increase the number of solutions available to them.
As I get older, heavier and slower, there are balls I physically cannot reach that I may have reached years ago. My physical capacity constrains the movement solutions available to me.
At higher competitive levels, those differences become increasingly important. When the ball travels faster, opponents direct it more effectively and the physical demands increase, strength and conditioning can expand the player’s potential solution space.
But that does not mean a stronger player automatically becomes a better tennis mover.
They still have to learn to use that capacity in relation to tennis information.
That is why I think S&C coaches and tennis coaches should work closely together. The physical coach can help increase the player’s action capabilities. The tennis environment then needs to provide opportunities for the player to exploit those new capabilities while interacting with the ball, opponent and court.
The transfer between those environments cannot simply be assumed.
Do recreational players need footwork training?
This is where I would make a clear distinction between different players.
If you play tennis once a week and want to improve your movement, my first recommendation probably would not be an agility ladder or an additional footwork session.
I would play more tennis.
More exposure gives you more opportunities to become sensitive to ball flight, court position, opponents and space. As your level improves, the game itself normally places greater demands on your movement. Opponents hit faster, move you wider and create different problems.
Your movement has to adapt.
There are obviously excellent reasons to strength train outside tennis. Physical activity, strength and cardiovascular fitness matter far beyond sport performance. But that is a different argument from saying that a recreational player needs separate footwork conditioning in order to improve their tennis.
For a serious competitive player, the balance changes. At regional, national and professional levels, physical qualities can become genuine performance constraints, and structured S&C becomes increasingly important.
The mistake is giving everybody the same prescription.
So how do you improve tennis footwork?
For me, the biggest change is to stop treating footwork as something separate from tennis.
We like to divide players into neat categories: technical, tactical, physical and mental. We do something similar with movement. We talk about racket work over here and footwork over there.
The player does not experience tennis that way.
The person, racket, ball, opponent and environment form one continuously interacting system. Movement emerges as the player tries to solve whatever problem the game presents.
So if you are a recreational player, play more. Experience more different balls, opponents and situations. Become more sensitive to what is happening before the ball reaches you.
If you are a coach, spend a little less time immediately correcting the feet and ask what information might be shaping the movement you are seeing. Has the player recognised where the ball is travelling early enough? What did their previous shot do? Where is their opponent? What space needs protecting? What are they actually trying to achieve?
Then manipulate the practice environment to help them discover better solutions.
For higher-level players, continue developing physical capacity as well. Strength, speed, agility and power can expand what the player is capable of doing, but those capabilities still need to become functional inside tennis.
Perhaps that is the biggest change I have made in how I think about footwork.
I no longer see the feet as the place where movement starts. By the time we see a player’s feet move, a huge amount has already happened between the player, the opponent, the ball and the environment.
If we really want better tennis movement, we need to stop decoupling footwork from the game and start helping players become better at perceiving, adapting and moving through the game itself.
References
Uzu, R., Shinya, M., & Oda, S. (2009). A split-step shortens the time to perform a choice reaction step-and-reach movement in a simulated tennis task. Journal of Sports Sciences, 27(12), 1233–1240. https://doi.org/10.1080/02640410903233222.
Filipčič, A., Leskošek, B., & Filipčič, T. (2017). Split-Step Timing of Professional and Junior Tennis Players. Journal of Human Kinetics, 55, 97–105. https://doi.org/10.1515/hukin-2017-0009.
Mecheri, S., Laffaye, G., Triolet, C., Leroy, D., Dicks, M., Choukou, M. A., & Benguigui, N. (2019). Relationship between split-step timing and leg stiffness in world-class tennis players when returning fast serves. Journal of Sports Sciences, 37(17), 1962–1971. https://doi.org/10.1080/02640414.2019.1609392.
Avilés, C., Navia, J. A., Ruiz, L. M., & Martínez de Quel, Ó. (2019). Do expert tennis players actually demonstrate anticipatory behavior when returning a first serve under representative conditions? A systematic review including quality assessment and methodological recommendations. Psychology of Sport and Exercise, 43, 16–26. https://doi.org/10.1016/j.psychsport.2018.12.015.
Navia, J. A., Avilés, C., Dicks, M., & Ruiz-Pérez, L. M. (2022). The spatiotemporal control of expert tennis players when returning first serves: A perception-action perspective. Journal of Sports Sciences, 40(1), 16–23. https://doi.org/10.1080/02640414.2021.1976484.
Pinder, R. A., Renshaw, I., & Davids, K. (2009). Information–movement coupling in developing cricketers under changing ecological practice constraints. Human Movement Science, 28(4), 468–479. https://doi.org/10.1016/j.humov.2009.02.003.
Pinder, R. A., Davids, K., Renshaw, I., & Araújo, D. (2011). Representative learning design and functionality of research and practice in sport. Journal of Sport & Exercise Psychology, 33(1), 146–155. https://doi.org/10.1123/jsep.33.1.146.
Williams, A. M., Ward, P., Knowles, J. M., & Smeeton, N. J. (2002). Anticipation skill in a real-world task: Measurement, training, and transfer in tennis. Journal of Experimental Psychology: Applied, 8(4), 259–270. https://doi.org/10.1037/1076-898X.8.4.259.
Williams, A. M., Ward, P., Smeeton, N. J., & Allen, D. (2004). Developing anticipation skills in tennis using on-court instruction: Perception versus perception and action. Journal of Applied Sport Psychology, 16(4), 350–360. https://doi.org/10.1080/10413200490518002.
