tl;dr

Movement variability is the often imperceptible change in movement execution from repetition to repetition. It is less obvious for submaximal movements with minimal fatigue and more obvious with fatigue or technique changes.

You could say that no two repetitions of the same movement are ever identical at any point, for multiple reasons discussed below. Despite how they may look. Movement variability is linked strongly to effective motor learning and fatigue management.

You could also attribute it to the opposite of specificity, or at least at the opposite end of the spectrum. If you're not specializing in a movement, then you are generalizing and having a high degree of variability is desirable.

This ultimately applies very well to similar but not identical movement patterns. An overhead squat changes a squat to a significant degree from a front squat, is different from a back squat, is different from a goblet squat or a bodyweight squat.

They are all 'squats' or 'lower body push' movements but the load positioning of the load changes the movement variability considerable more than continually doing one but not the others.

Movement variability is important for:

  • Addressing gaps that prevent progress
  • Breadth of strength/ movement expression
  • Improving and maintaining adaptability
  • Preventing overuse injuries
  • Enhancing movement interpretation
  • Neuromechanical Matching (training different muscle fibers/ units even within the same muscles)
  • Improving performance (multiple exercises still tend to improve singular exercise ability because of the complementary nature of the nervous system)
  • Preventing is a static training environment and boredom.
  • Improving quality of life

This is something I’ve rarely seen discussed in any meaningful detail in books (that aren’t motor learning textbooks) or on the web but why should you vary your training programs periodically and by how much?

A client recently asked me;

“What’s the point of all these changes in my program? What are we working towards by changing this to this and that to that? Why not keep this the same until I’ve really learned it?”

All good questions.

I’ve read more than a little research and seen my share of presentations to be able to justify my answers scientifically (I’ll do a little bit below) but the simple answer to all of these questions is: Movement Variability.

The more complicated answers are:

  1. To maintain joint health, we want to utilize some movement variability, develop transferability, prevent plateaus, to keep things interesting for you…
  2. We’re working towards movement competency in each of the major basic movement patterns (knee bend, hip hinge, push, pull, etc…), so that we can get into more advanced patterns or combined patterns…
  3. If you keep hammering a nail that’s bent, is it going to set well? Or is it easier to pull it out and start over again with a new nail? Will it set better on the second attempt with a new nail? Probably… If you’re training well, 3-8 weeks is more than enough time for most movements to plateau, at that point most people are better off modifying the movement. Keeping things similar maybe, but different. You need a new movement challenge to really learn something and then be able to apply it.

What Is Movement Variability?

It’s the normal variation that occurs across multiple repetitions of any given task.

And there is always variation, even when it looks like there isn't.

For example, you do a set of bodyweight squats for five reps. All five reps will likely look much the same. However, imperceptible to the naked eye, there were subtle changes in muscle firing patterns, fibre utilization, joint moment arms and ultimately the path of movement.

Now five bodyweight squats is pretty easy for most people to execute. Without fatigue or proximity to failure you get less movement variability. This is in part why fatigue is the enemy of motor learning. However, some proximity to failure necessary for muscle building, creating a bit of a conundrum, depending on your training goals. I'll come back to this.

To make the variability more visible to the naked eye, just add load. Take the same five reps but this time add something closer to your five repetition maximum (5RM) and you'll see more obvious changes from rep-to-rep.

As fatigue sets in, you'll notice more natural variation within the movement. The first two reps might look pretty similar, but then next two you might notice some knee movement changes, the hips might shoot up earlier or later depending on where your weaknesses lie. By the last rep, the squat will be really slow with the worst looking technique of all the reps.

The objective and the result of both squats still the same in both instances: Squat five times. How the body got from point A to point B changed based on the addition of load. It's important to note that variability can also change based on the location of the load – front squat vs back squat for example – and the number of repetitions too.

Human movement lies on a continuum between too much variability and too much specificity, depending on the task or collection of tasks and the goals of the individual.

As a general rule specificity is only a real requirement in the elite/advanced...

A Spectrum of Need

This example shouldn’t be viewed as doctrine but it illustrates a decent point.

Sports like powerlifting or olympic lifting exist at an end of the continuum where we almost want to minimize movement variability to focus on specificity.

There will still be some variability still, but for success at elite levels of competition you'll need the competition movements to look as close to the same every time we train. Or as similar as humanly possible. We need to specialize to to be successful at these sports at elite levels and often that specialization will come at the expense of variability.

Bodybuilding will need more variety than powerlifting because we want to attack muscles from different angles and use multiple exercises to accomplish this. There are simply too many useful muscle developing exercises to be doing at all times, so you have to cycle more of them in and out. i.e. we can't do a hundred exercises within a month (1 phase) of programming, but we might be able to integrate 100 within a macrocycle.

Now there are sports where creativity and variability are more desirable due to chaotic opponent driven circumstance. Team sports being among them with many martial arts representing the other more extreme side of the variability spectrum. Yes, you'll still lift weights but you need considerably more strength, agility, conditioning and power in a bigger variety of positions than bodybuilding, powerlifting or Olympic lifting.

General population 'health and well-being' individuals tend to be somewhere in the middle of this spectrum. There is very little need to specialize in only a handful of resistance training exercises because the end goal isn't elite performance in 1-5 lifts. On the other hand, many will like to participate in activities above and beyond basic sitting and walking. Activities like playing with their kids, gardening, handy work around the house, helping a friend move or amateur sport participation.

Long story short, the more movement you have exposure to in your life, the greater need for movement variance in your exercise programming. The more specialized your movement needs to be the less exposure you probably need.

Within reason, and I’ll discuss that further below.

In a Practical Context

If your movement were perfect, you wouldn't need the practice. And yet, as I just told you, even without fatigue, there is no such thing as a set of perfect repetitions.

I won’t let my clients bang out 5 ugly looking squats. That’s too much poor variability under fatigue that is likely to reinforce bad movement habits, but I’m okay with the odd ugly looking one, every once and awhile. We want to reach outside our comfort zone a little from time to time.

Instead of letting anyone just bang out ugly squats, we find a way to make them look pretty most of the time. At the same time if your squats are always pretty, you're likely never challenging yourself outside your comfort zone.

It's a balancing act.

The need to groove good movement patterns and at the same time prepare people for new challenges that look or feel familiar to what they already know and have experienced.

When the body comes across a new a situation that looks like a squat, feels like a squat and just might be a squat, but is not exactly a squat you’ve done before, you’re ultimately prepared. i.e. you travel to Indonesia and the toilet is just a hole in the room.

Throughout a good program we want to challenge the body to approach things differently periodically. To prepare the body for unique variation. This is fundamentally why you want to incorporate some degree of movement variability.

You Can't Completely Specialize

Looking to the spectrum above, it’s easy to think that powerlifters should just squat, bench and deadlift but it turns out that this still yields less than ideal results.

Most elite powerlifters have long known this and introduced movement variability into their training. Instead of merely back squatting, they’ve introduced many slight variations like box squats or safety bar squats. Instead of merely strict bench press, they'll utilize board presses or floor presses.

Powerlifters (and Olympic lifters) have also long added accessory movements that resemble bodybuilding-esque training. Isolation movements that increase muscle mass and work on weak links directly.

No competitive modern strength sport athletes exclusively train the competition lifts. Specialization is necessary, but will only get you so far. If you do what you've always done you get what you've always got.

The lesson being to push through plateaus even in highly specialized environments you need to try some new things periodically. Not too similar but not too different either. We still have a squat and a press, but they aren’t the same competition squat and press. Only slight variation and supportive isolation exercise.

The Problem With Too Much Specificity

Notably: Strength Expression.

Of any quality you can readily train, strength is one of the easiest to improve, notably enhances performance and decreases injury potential. It’s a true game changer but it can still be misapplied.

In more highly variable team sports or martial arts competitors I’ve seen too many instances, where I take on an athlete with too much specialization in a lift. This is why the pervasive myth that weightlifting decreases performance in these sports still exists.

It’s not that weightlifting actually decreases performance in sports, it’s that some people try to train these athletes with the specificity of a powerlifter (or Olympic lifters). You see similar problems with general population trainees who have latched onto one approach too strongly as well.

You cannot turn rotational and chaotic team sport athletes into powerlifters and expect them to still excel in their sport. Where pushing, pulling, jumping and other movements are expressed from a variety of different angles and strength expression needs to happen within varying degrees of stability.

For example, a football athlete who can crush bench press but still somehow struggles to do push ups. I took on an athlete once who was benching almost 4 plates a side — 365-385 lbs, something like that — yet could not manage 5 pushups with decent form.

How is this possible?

Two Words: Opportunity Cost

"the loss of potential gain from other alternatives when one alternative is chosen."

There is a cost of doing business, so training with a high degree of specialization tends to happen at the expense of variability, and vice versa.

They are both fundamentally a horizontal press but the push up has suffered as a result of too much bench press specialization. Strength expression has thus become limited to the bench press, which is not how any athlete will ever push anything in sport.

Take a different example of a hockey player that can absolutely crush back squats but fumbles all over the place the moment you put the foot behind them into a split squat position or a staggered stance position.

They are all fundamentally lower body presses but the split squat has suffered as a result of too much squatting. General lower body push strength suffers as a result and strength cannot be expressed outside the specificity of a squat.

I can't tell you how frequently I've seen this amongst new clients but it's likely as common as merely being completely untrained – granted many people who seek my services value fitness and come to me with some degree of fitness already.

Why Movement Variability Matters

Researchers used to attribute movement variation in biomechanical analysis as just being noise in their measurements but it’s become clear in the recent years that it’s an essential part of how human movement works.

Instinctually, it's natural to assume that the more time you can dedicate to practicing one movement, the better you'll be at it. Yet, research conflicts a great deal with this assumption.

For one, movement variability exists within each repetition of "same" movement already. You cannot escape it with perfectionist tendencies or by only practicing one movement at a time. The outcomes of such an approach trend to worse.

What's the saying, "if you can't beat 'em, join 'em?"

That's not to say that you should embrace doing something different every time you train either. The answer to one extreme side of the spectrum isn't to go to the other extreme.

If you change exercises too frequently, there is no way to know if you've gotten better at them. Some kind of consistency is needed and some kind of variability is needed. It's about finding the right balance.

Near as I can tell, here are some extremely valid reasons to incorporate some kind of movement variability into your programming:

  1. Increases the breadth of strength expression
  2. It improves motor learning/skill
  3. Increases adaptability
  4. Likely reduces the risk of injury (especially repetitive strain injury)
  5. Neuromechanical Matching
  6. Provides Balance to the system

Meaning that movement variability is also the purposeful manipulation of variables within an exercise program or similar group of exercises to improve each of those points. It should include more than the movement variability you only see within the same exercise.

Strength Expression

There are over 600 muscles in the body, it's impossible to adequately train all of them with even a few dozen carefully-selected exercises.

If you want to excel in something highly specific, you need to specialize to a degree. If not, your everyday life exposes you to a variety of movement situations.

Movement puzzles if you will.

  • How to move a couch around a tight corner with a friend?
  • How to efficiently throw a child in a pool without hurting them.?
  • How to maneuver safely around a big boulder on a hike?
  • Etc...etc...

There is a good chance your body will be called upon to do a lot more than just squat with a barbell or snatch a barbell overhead.

Both are good exercises to use periodically if you can, but there is zero reason to specialize if you don't compete in either. And plenty to gain by regularly manipulating the tools, reps, and exercises you use when you train.

With my general population clients, I want to maximize their ability to express strength from a variety of positions because they are likely to deal with a variety of movement puzzles in everyday life. A degree of variety and a breadth of strength expression will more adequately prepare them for that. Rather than trying to master only one lift of many.

Motor Learning

Evidence has been accumulating to support the idea that some degree of movement variability improves motor learning and function.

Multiple studies have suggested that neural variability is conducive to motor learning. This is in line with other reinforcement learning theories, like interleaved practice.

I have an article in the works on that topic, but basically it's an idea that sits contrary to popular belief on learning. Specializing on a topic (to learn) for blocks of time yields poorer learning outcomes than interleaving multiple topics or elements of the same topic. Mixing topics together and adding a small amount of variety tends to yield better learning outcomes.

There is a paper somewhere that I learned of from a motor learning textbook and always struggle to find. The conclusion was something along the lines of young/ new tennis athletes learned to serve with better proficiency in the long-term by varying the location of that serve than by staying in one place. They struggled more initially but rapidly overtook the group only serving from one location on the baseline.

The explanation was improved motor pattern adaptation, as even from the same location, no serve is ever exactly the same anyway.

When it comes to movement variability and motor learning it appears that the fluctuations (or the "noise") in neuron activity is actively regulating and shaping motor variability to augment or enhance learning. It would explain why we see so much noise within the same movement execution.

The specific details still need to be teased out and established, but there are sophisticated ways that neural circuits develop and are maintained.

Adaptability

In the same vein as motor learning and probably strength expression. If you want to be more “athletic” or more easily express a variety of daily living movements, you need to be adaptable to a variety of (often 'unique') circumstances.

Some movement variability produces adaptability within the nervous system to various patterns of coordination. This adaptability is necessary to secure positive outcomes in unfamiliar but similar movements. i.e. deadlifting that awkward couch that you only move once until you get a new couch.

The old you get and the more specialization you develop, the less transfer outside the specialization we see and the less adaptability to new situations is the result. A 30 year old all-star baseball pitcher won't necessarily make an all-star quarterback, despite both needing excellent overhead throwing ability.

You see this when trying to fix a new clients movement pattern. It's generally harder to undo an established pattern of movement and replace it with something new than it is to build a new movement.

It's also why the long-term athletic development protocol prioritizes a lot of movement variety initially (beginners) and only encourages or promotes specialization when an athlete is at a level that warrants it.

To quote one review:

“…variability in movement systems helps individuals adapt to the unique constraints (personal, task and environmental) impinging on them across different timescales.”

Optimal movement variability exposure leads to complex neural mapping, plasticity, and performance, depending on the environment and the amount of variability required.

Injury

And to a degree balance too.

Photo by S O C I A L . C U T / Unsplash

As is the case with many of my clients, most (if not all) are knowledge workers and by extension spend most of their days at a desk. i.e. desk jockies! I say that with love...

Regardless of the typical strained postural positions you find yourself in, your exercise programming should attempt to provide an appropriate counterbalance stimulus to positions you find yourself most often in.

It's not that slouching once and a while is bad for you, it's that eventually any form of specialization opens the door for injury due to repetitive strain. And a variety of postures or movement is the remedy.

Too much specificity likely open the door for injury, but the research in this department is limited currently to youth athletic development.

Neuromechanical Matching

Warrants a new article probably as I'm still learning about it. This is a pretty new concept, when I started writing this article, this wasn't a term on my radar.

To quote a recent/introduction review to the topic:

Motor tasks involve the recruitment and discharge of motor units, both within and across muscles, which work in synergy to develop force. A motor unit is defined as a motoneuron, its axon and the muscle fibers it innervates. The firing pattern of a motoneuron is determined by the conversion of synaptic inputs into action potentials, which depends on the intrinsic properties of motoneurons, neuromodulation of the biophysical properties of motoneurons, and the summation of multiple excitatory and inhibitory synaptic inputs to the motoneuron (1).

A lot of big words in there, but in lay terms movement requires various subunits of muscles (called motor units) to fire in synergy to develop force. That firing pattern is ultimately determined by a lot of things but it's likely that previous exposure, positioning, load, and training history are huge determinants of what fires in what order.

Bodybuilders have long known and attested to the ability of the human body to make slight tweaks to a movements with the end result being more hypertrophy in a given location.

There is a great deal of electromyography (EMG) research that supports this claim, we see different electrical activity within the same muscles depending on positioning and the movement trajectory. Implying different motor units within a muscle are activated according to the demands of the position.

For instance rotating the toes inward, straight-ahead or outward during a calf raise yields proportionately different rates of calf (gastrocnemius) hypertrophy. When you turn the toes outward, you see more hypertrophy on the fibres towards the inside (medial part) of the calf. When you're toes are pointed inward, you see more hypertrophy in the fibers on the outside (lateral part) of the calf. When your toes are straight ahead you see more hypertrophy in the middle of the muscle.

Muscles are 3-D, so only doing one movement for each muscle from the same position is detrimental to complete development in the long-term.

Use only one toe positioning, and in the long-term you risk over-developing one part of the muscle at the possible expense of other parts of the some muscle. The motor units within the muscle not involved in the one exercise you've chosen. Without some variance in foot positioning, you'd never provide an adequate balance for developing the entire muscle.

Balance

Overlaps with reduced injury risk to a degree but also prevents weak links from forming and provides a gauge for where other training can or should be incorporated.

You might assume that only doing one thing leads to better outcomes but the limited research we have suggests otherwise.

For example, in one study, they compared five groups performing resistance training with differing exercise selections and intensities.

  • Constant Intensity, Varied Exercise (CIVE)
  • Constant Intensity, Constant Exercise (CICE)
  • Varied Intensity, Constant Exercise (VICE)
  • Varied Intensity, Varied Exercise (VIVE)
  • Control Group (they did no training, obviously...)

The constant exercise groups only the Smith Machine Squat, which has even less movement variability than a barbell squat. The varied exercise groups did leg press, deadlift and lunges on top of the Smith Machine Squat. All groups did the same total number of reps and sets, just spread over one movement, or over four.

In effect, you use a variety of movements on one side and no variety of movement on the other side. With or without constant intensity – changes to the rep range.

You might expect that people doing only the Smith Machine Squat would develop the most strength in that exercise, after all they did more specific training and volume on that exercise.

However, the group that did 4 exercises instead of one, displayed the most improvement in strength for this single exercise, despite doing 25% of the volume.

They also experienced more complete leg hypertrophy, which isn’t surprisingly because more muscle in the legs were challenged with four exercises than only one.

One explanation might be that the 4 exercises better manage to improve weak points of the movement that one exercise alone cannot effectively train. The Smith machine might not target the hamstrings effectively for instance and if that is a weak point, the deadlift would shore that weakness up significantly better than the Smith Machine ever could alone.

Keep in mind that the study design isn’t perfect, and was done on recreationally trained (at best) people so we need more research like this. Nevertheless it supports the notion that some variety is needed, just not necessarily a lot of it.

The other balance consideration is that not every exercise maximizes different parts or elements of any lift. Yes, exercises are important but they can be broken down further into their requisite parts. Static strength, eccentric strength, and concentric strength for example.

A Romanian Deadlift (or hip hinge) will train considerably more of the lowering action (eccentric phase) of the hamstrings than a conventional or sumo deadlift. This change in muscle action can change the training effect dramatically creating more hypertrophy in series and better eccentric specific training adaptation. This could be very desirable in a sprint athlete, but you can't train all of these qualities all the time.

Movement variability allows you to programmatically address weak areas of staple lifts and relevant movement patterns.

How Much Variability Do You Need?

That's a good question, one I've been trying to determine myself for quite sometime.

What appears clear is that beginners do better with a ton of variety and interleaved learning upfront than more advanced trainees with more specific goals. You want a broad focus when you're new to something, one that possibly narrows as your goals change.

That's why plenty of 'simple programs' online like Starting Strength fall way short of long-term value with real beginners.

I understand the desire for simplicity and it can improve adherence but it's equivalent to telling a C student they'll never amount to anything. It pigeon holes your motor skill.

The calf study aboves suggests that 3 positions per muscle group is a good starting point. The smith machine squat suggests that 4 exercises can also work.

Anecdotally 3-5 appears to be a decent range of variety most of the time. The more elusive answer is how do we optimally distribute the variety? Do you really need 3-5 exercises per phase (or month) of programming?

That seems unlikely. Depending on the objective.

For the beginner simply needing to learn how to lift, 3-5 exercises within a phase could be good for learning the coordination of various lifts. Variety appears to improve outcomes in this instance.

For a strength objective, a plateau might indicate it's time for a change, or a deload if you need to specialize in a certain lift at least.

For hypertrophy example you might get the best outcome when an exercise is kept in the routine long-enough to plateau and then some. However, to accumulate an ideal amount of volume (# of sets) you're probably going to use at least 2-4 exercises that act on any given muscle per phase anyway.

There is still a ton of research to be done, and as I said above, it’s likely to be different depending on the person, the activity and their environment.

In my view good training is generally about optimizing an individual’s degree of freedom for optimal self-governance that can occur during the movement. A warm-up is a good opportunity to introduce a lot of movement variety at multiple thresholds too.

No one taught Michael Jordan how to do a reverse dunk, no one taught Pele a bicycle kick and I’d like to think that I don’t really teach people how to squat, deadlift, push, pull or twist. I’m just trying to optimize their experience with any given movement so that they figure out as much of it as they can on their own by exploring multiple options.

Sure I’ll toss out a cue or two, or write down a little blurb on how to execute it but don’t get too in your head about the specifics.

Generally speaking in all the movements I coach, I just want to see good joint positioning.  Most importantly the spinal position followed by joint stacking and I try not to obsess much over anything else.

Movement has variance from person to person and even spinal and joint positions will vary from person to person too. “Optimal” is more about probabilities than absolutes.

Summary

As usual that post got away from me, but I sincerely hope it provides a great deal of clarity into why you should include a wide variety of exercises into your programming. Especially if you're new to training. The more exposure you get early on, the better.

However, the degree of variability relates directly to your goals and actual ability in the l0ng-term. You can specialize, I simply wouldn't do that until you're expressing some mastery over particular lifts and trending in a competitive direction.

Athletes that specialize too early are prone to burn-out, and so are you. Not only that, premature specialization will likely lead to a:

  • Lack of skill/coordination development
  • Increased chance of injury
  • Lack of adaptability
  • Limited long-term potential
  • Inadequately trained muscles or motor units within muscles
  • A lack of balance in your programming

Do you want to sacrifice your long-term development simply to simplify your training to a point of inadequacy?

Einstein is quoted as saying:

"Everything should be as simple as it can be, but not simpler."

It's important not to forget the last part of that quote. Squat, Bench, Deadlift is incredibly simple, but it leaves a lot of holes too. Rectus femoris won't be trained, the biceps femoris won't be trained, the calves won't be trained, etc...etc...

Large muscle groups being excluded in your training won't prepare you for the movement variety that is life.

Movement variability is important to understand because few people talk about it. Strength and conditioning coaches, physical preparation coaches, whatever you want to call us, we put out these “recipes” online for people to do, without really instructing about the importance of movement.

I have one out there myself.

Just like any other recipe though, there is variability. They reason your souffle didn’t turn as well as a recipe indicated might be because you’re at a different altitude than the person that wrote the recipe.

Life is 3-D. Movement is thus 3-D. Most exercises are 1-D, maybe 2-D. Your body likes to move in a variety of ways.

Now get out there and move!

Further Reading

Creative Motor Actions As Emerging from Movement Variability
In cognitive science, creative ideas are defined as original and feasible solutions in response to problems. A common proposal is that creative ideas are generated across dedicated cognitive pathways. Only after creative ideas have emerged, they can be enacted to solve the problem. We present an alt…
The Propagation of Movement Variability in Time: A Methodological Approach for Discrete Movements with Multiple Degrees of Freedom
In recent years, theory-building in motor neuroscience and our understanding of the synergistic control of the redundant human motor system has significantly profited from the emergence of a range of different mathematical approaches to analyze the structure of movement variability. Approaches such …
The Role of Variability in Motor Learning
Trial-to-trial variability in the execution of movements and motor skills is ubiquitous, and widely considered to be the unwanted consequence of a ‘noisy’ nervous system. However, recent studies have suggested that motor variability may ...