Stop Dragging Your Toes: Part VI

Stuart McMillan

Stuart McMillan


Welcome back to the final part of this series on the ‘toe-drag phenomenon’.  

Thanks for sticking with me here!

What was supposed to be just a short post on this somewhat controversial topic, has ended up being a six-parter – which honestly, could go on forever!  I tried writing this succinctly, but the reality is that much of what I was writing to start off with required further context – which led me down this current path.  

This sixth and final part might be a bit of a handful.  I have introduced a number of concepts that deserve multiple-part articles on their own – and I will perhaps endeavor to do so in the future. 

Firstly, I felt it was important we appreciate that block clearance technique does not exist in a vacuum; like any movement, it is a manifestation of the on-going interactions between the athlete, the task, and the environment – as described by Newell’s ‘Constraints-led Approach’. 

Secondly, framing the movement technique as a ‘solution to a problem’ allows us to appreciate that movement is ‘goal-directed’ – i.e.  goals drive our actions; and our actions are driven by a desired future state.  

As we recall, these solutions are governed by a number of internal and external factors, including ‘structural constraints’ (height, weight, etc.), and the athlete’s current understanding of the task. 

Remember from part III, that some of these factors are relatively easily modifiable (weight, shoes, surface, etc.), others are more difficult to modify (fibre type, length-tension relationships, etc.), and others cannot be modified at all (e.g. height).  One theme I hope you have taken from this series is that we should not try to fit square pegs into round holes – there is no single technique that is common to all athletes.  

In order to devise our overarching technical models, we utilize a framework through which to identify the key performance indicators (KPIs) of the task, as well as the underpinning physiological, biomechanical, anthropometric, and neuromuscular factors of these KPIs.  

Once we have gone through this process, our goal is to identify the athletes’ ‘unique abilities’  – in effect, asking “why is this athlete good at what she does?”  We answer this question through a process of observation (first qualitative and then quantitative), encouraging the athletes to move in their ‘most-natural’ way, and reducing verbal communication to a minimum.  We adjust task and environmental constraints to challenge the athlete’s technique – in so doing, testing their technical and skillful ‘stability’ in a variety of different contexts.  

An athlete cannot show us what they can do until we stop telling them what to do.

Nick Winkelman 

Once we have a better appreciation for how the athlete moves, we chunk all of this complex information as best we can by categorizing the athlete into various ‘movement biases’ – or mailboxes.  Recall that I offered examples in the scientific literature, such as force- or velocity-dominant, stride-length or stride-frequency reliant, and aerial- or terrestrial-dominant. Within our athlete population, I have also categorized by single-leg or double-leg dominance, and anterior-chain or posterior-chain dominance (pushers and pullers). 

In this final part of this series, I will introduce my two primary block-clearance mailboxes:

  1. TYPE A – linear & volitional 
  2. TYPE B – torsional & reflexive

Before describing these mailboxes in more detail, it is important to understand that this is not a straight dichotomy; it is not a binary choice between one or the other.  Rather, each athlete will exist on a continuum between TYPE A on one end, and TYPE B on the other. 


I have previously described the TYPE A athlete as ‘muscularly-driven’ – but to be honest, I feel this is a bit limiting, and it seems that too many folks get caught up in the terminology, rather than the description. 

The TYPE A athlete will tend to bias towards: 

  • Force production
  • Pushing (i.e. anterior-chain dominance)
  • Concentric strength abilities

The movement expression of a TYPE A sprinter will generally show smaller shapes and faster rhythms, while manifesting in more linear movement relative to TYPE B. 

The TYPE A athlete will generally accelerate with lower heel recovery relative to the TYPE B, with a ‘scooping’ motion from back to front.


I have previously described the TYPE B athlete as ‘fascially-driven’, these athletes tend to use more torsional and reflexive strategies, relative to TYPE A athletes, with movement expressions that include larger shapes and slower rhythms.  The TYPE B athlete will generally have a slightly more cyclical heel recovery, relative to TYPE A athletes, and tend to bias towards:

  • Force transmission or amplification 
  • Pulling (i.e. posterior-chain dominance)
  • Eccentric strength abilities 

Let’s now dive into the details of these two mailboxes. 


TYPE A sprinters are your ‘shufflers’.  You will recall that the TYPE A athlete will generally produce more force, have slightly quicker rhythms, and sprint relatively more linearly than Type B athletes.  It is these types of athletes where a low heel recovery starter might be appropriate. Before we delve a little more into that, however, let’s remind ourselves of what coaches are actually looking for when they prescribe low heel recovery, in the first place: this can be described through a three-step process:

  1. Once the foot leaves the ground on the back-side (toe-off), a low-heel recovery is prescribed (the foot should recover ‘as close to the ground, without touching it’ is what many coaches will suggest)
  2. Once the foot has traveled past the center of mass, it should lift up from the ground at sufficient enough height such that the athlete can apply the appropriate amount of force, in the appropriate direction 
  3. The athlete should push back and down as violently as possible, striking under – or slightly behind – the center of mass 

The above seems reasonable enough – and I don’t disagree with any of these steps – as long as they are consistent with how the athlete moves naturally. If it is not – and the athlete has to spend months working to ‘master’ this technique, for example – or if the athlete’s governing constraints are such that this technique requires ‘over-riding’ the self-organizing system to such a degree that the technique looks and feels awkward, then chances are we are going to negatively affect both their performance and their health – in both the short-term and the long-term.  

There are two primary justifications for the low heel recovery start (or ‘shuffle’ – or whatever other terminology you want to use):

  1. “the shortest distance between two points is a straight line”
  2. “you can’t apply force when you are in the air”

Again – seems reasonable enough.  

But neither of these justifications survive scrutiny (at least not across the board).

Let’s discuss them in order:

“the shortest distance between two points is a straight line”

Firstly, this is not the right question to ask – what we really need to be asking is not what the shortest distance is, but what the fastest is.  Even more than that, we need to ask what is the most-efficient – and it is simply not true that it is always a straight line.  

For example, if one point is at a different elevation than another (as in gait), then – due to gravity – the fastest route is not a straight line at all – but a curve, known as a brachistochrone curve (the curve of fastest descent), and although it may occasionally look otherwise, sprinters are not immune to the effects of gravity

Keep in mind also that a low recovery means that the length of the lever is significantly longer – and therefore will take longer time to swing through. Younger athletes, especially, will have difficulty pulling through a long lever, often scraping their toes on the ground, and struggle to project their CoMs vertically enough to give them the time to apply large amounts of force. 

In addition, heel recovery of the ankle is multi-planar: it involves multiple muscles, multiple joints, and multiple bones working to flex-extend, abduct/adduct, supinate-pronate, dorsiflex-plantarflex, invert-evert, protract-retract, depress-elevate, circumspect, and rotate. We cannot simply carve away one of these movements (in this case, flexion-extension) without considering the others.  Even if we did reduce movement to simple flexion-extension, we still have multiple joints to consider – from toe-off, the hip, knee, and ankle joints all flex simultaneously – but in a highly coordinated manner that is totally independent of conscious control. 

Consider also the role of reflexes, ‘preflexes’, sensory feedback, proprioception, and variability – all affecting this so-called shortest line, and we begin to appreciate the complexities of this movement. 

 “you can’t apply force when you are in the air”

Sure – at its core, this is true: we can only apply force into the ground while we are on the ground.  But the amount of force, as well as the direction to which it is applied, is dependent upon what happens in the air: i.e. more space = more force

This is not difficult to understand: make a fist, and hold it 1” away from the wall – now, without withdrawing your arm, punch the wall as hard as you can.  Chances are, if your name is not Bruce Lee, you probably were not able to punch the wall hard enough to hurt your knuckles. Now bring your fist back to 12”, and do the same.  At 12” away from the wall, you can do some serious damage to your knuckles and-or the wall. In addition, at this distance (or any other distance), do you apply the same amount of force as your friends?  Obviously not, right?  

The bottom line is this: sprinters require sufficient vertical impulse to create enough flight time to reposition their limbs, “whilst all other strength reserves are applied horizontally in order to maximise acceleration” (Wild, 2019).  The key here is that some athletes require more space and time than others – even in the elite ranks. Consider, for example, the variance in flight time between athletes in the following table from the World Athletics Biomechanical report from the 2018 World Indoor Championships.  

If you are not familiar with these statistics, the key takeaway is that the difference in 1st step flight time varies two-fold between competitors (i.e., more than 100% difference between 0.027 – Taftian,  to 0.060 – Xie). That difference of 0.033s may not seem like a lot – but it is actually substantial variance. 

Flight time variance is higher still with sub-elite sprinters, as shown by Slawinski, 2010 and Otsuka, 2014

What this research is telling us – again – is that all athletes are different, and use different movement solutions, leading to different movements expressions; some sprinters need to lift their feet up high, and hammer it back into the ground. Others (mainly elites) don’t need a lot of space, and can apply such high forces so quickly that they can, in effect, shuffle.  

This does not mean that this is the way to do it for all sprinters, however!!  

All this means is that some outliers of outliers are able to do it  (and by ‘outliers of outliers’ – I would suggest that there might be a dozen guys on the planet who can effectively ‘shuffle’ – and even fewer females). 

Simply put, athletes need to be in the air for long enough to reposition their limbs in time to strike the ground in the appropriate place, and in the appropriate direction.  Some can do this very quickly; others cannot.  

Additionally, it is important to remember that while force generation doesn’t begin until the foot contacts the ground, the initiation comes while the foot is still in the air – peak power generation at the hip is observed prior to initial ground contact (Debaere, 2012). 

The way Dan Pfaff described it was if you were to hammer a stake into the ground, would you use a sledge hammer lifted over your head, or would you use a ball-peen hammer, and just tap away? 

The higher you raise the hammer the harder you can hit the stake

We found that the fastest athletes all do the same thing to apply the greater forces needed to attain faster speeds. They cock the knee high before driving the foot into the ground, while maintaining a stiff ankle. These actions elevate ground forces by stopping the lower leg abruptly upon impact.

Dr. Peter Weyand

Remember that the height of the swing leg foot is a result of the amount and direction of force applied by the stance-leg.  The ‘shuffler’ will volitionally cut their stride short in an attempt to increase frequency and positively affect the ratio between ground contact time and flight time. However, when we cut the stride short, we don’t take advantage of the elastic components (stretch reflex) of the muscles. 

After the stretch, there is nothing the athlete can do that would make the movement faster.  A good example would be a person, after loading a slingshot, trying to make the projectile move faster by moving the hand with the projectile forward. This action would cause a reduction of the elastic force, resulting in slower movements.  There is an illusion of speed when the levers do not go through their full range of motion, but the objective is not to move the lever fast but rather to transport the mass down the track in the least time possible.

Coach Tom Tellez

Now – the heuristic “more space = more force” does not mean that we ask all our athletes to pick up their feet (or their knees) as high as possible (or push back as far as possible).  There is a trade-off between the amount of time in the air and the amount of force applied to the ground. As hopefully you understand by now: how we want an athlete to move depends upon ‘how that athlete moves’ – i.e. the internal and external factors that govern the ‘movement solutions’ that are available to them. 

Bruce Lee might be able to knock a man over with a 1” punch – but no one else can.  


The TYPE A sprinters are good because they can apply large amounts of force.  They tend to be muscular, powerful, and seemingly efficient – with minimal rocking from side to side, or rotating back and forth.  They tend to have high stride frequencies, and can accelerate very fast.  

These athletes can be excellent 60m specialists, but often have trouble maintaining high velocities for very long, and can perhaps tighten up in later stages of the race. 


Before I introduce this mailbox, we need to take a quick step back to better understand our sprinting biomechanics. Firstly, we must acknowledge that the most important aspects of movement may not be measurable, or even easily observed in the first place.  Our eyes are incredible organs, but while we live in a 3D world, we can only see in two dimensions.  

When we observe a sprinter exiting the blocks from the side, we can see extension and flexion fairly clearly, but can only infer adduction-abduction, spine and thigh rotation, and foot inversion-eversion, etc.  

The limits of our eyes biases us as to what is really happening, and we act surprised when someone shows us photos or a video from the front or back – or worse yet – when we ourselves watch athletes from in front or behind, and deem any ‘non-linear’ movement as inefficient. 

graphic from @RUGBY_STR_COACH Twitter account

This couldn’t be further from the truth, and coaches who are attempting to constrain any non-linear movement of their sprinters are doing them a disservice. 

Pelvic and lumbar rotation during walking gait has been clear for decades; Saunders’ seminal work (1953) included pelvic rotation as one of the ‘six determinants of gait’, for example.  

The running and sprinting world might have been a little late to the party, but it has been at least 30 years since coaches have understood the significance of shoulder,  pelvis, and spinal rotation to sprint speed. Coaches Charlie Francis, Tom Tellez, and Dan Pfaff, for example, were communicating the relevance of rotation as long ago as the early 1980s.  

The flexion and extension ranges and timing of the knee joint must be countered by the upper extremities in order to maintain structural and postural balance in gait. If this counterbalance scheme is efficient, we see a harmony of hip and shoulder transverse axes oscillating and undulating with rotational fulcrum of the spine being in the lower thoracic region.

Coach Dan Pfaff

This ‘functional neural coupling’ – the involuntary coordination between arms and legs – and the considerable oscillation and undulation (as well as tilt) of the pelvis and thorax observed by Coaches for decades has only recently begun to be better defined in the research.   For example, Ryu Nagahara in 2014, was the first to investigate the thorax and pelvis movements during the entire acceleration phase of sprinting. 

The counter-rotation of the shoulders and pelvis – along with the adduction of the hip allows sprinters to strike the ground with their feet relatively inside their base of support “minimizing the effect of the lateral foot positioning on the medio-lateral velocity of the centre of mass.” (Gabaere, et. al, 2012)

The details behind the complex network of rotation, torsion, reflexes, co-contractions, fascial slings, etc. are beyond the scope of this already too long series; but suffice it to say, it is not yet well-understood (at least in elite sprinters), and is thus ignored by many.  

That said, I will attempt to briefly explain a couple of under-appreciated principles:

During gait, the lumbar vertebrae flexes, extends, rotates, and side-bends. Gracovetsky’s ‘Spinal Engine Theory’ hinges on the concept of the coupled motion of the spine, whereby a lateral bend of the spine induces an axial torque driving the pelvis.  

The lower extremity can be completely removed without interfering with the primary movement of the pelvis … it is obviously preferable to have legs, but they only amplify the movements of the pelvis, and their functional role remains secondary.

Serge Gracovetsky

One way to help us better appreciate this is by understanding the relationship between proximal and distal muscle-tendon ratios.  The ratio of muscle to tendon around the pelvis is very high, while the ratio of muscle to tendon at the extremities is the opposite – where there are significantly more tendinous and elastic structures.  

So what does this mean?

Athletes with thick erectors, big gluteals, large abdominal musculature, etc. will tend to be better ‘force-producers’ (TYPE A); while athletes with less muscle mass around the hips, and longer Achilles’ tendons tend to be better ‘force amplifiers’ (TYPE B). 

The first group is what I have termed ‘muscularly-driven’ athletes, while the second is what I have termed ‘fascially-driven’

When it comes to block clearance and acceleration, the first group might be the successful ‘shufflers’ – they are generally more square; they ‘hit, push, and recover’ with very little rotation.  The second group rely more on torsion – winding and unwinding their fascial slings – more reliant on their reflexes; and we will thus observe significantly more rotation. Sometimes, you will find a hybrid – an athlete with huge ‘force-producers’ and efficient ‘force-amplifiers’; these sprinters tend to be the best starters in the world (Bolt and Coleman are maybe the two best examples, and if you don’t think Bolt was a good starter, check out his 10-20 and 20-30m splits!). 

Again – it is important to point out that all elite sprinters take advantage of their rotational capacities; the difference lies in where they sit on the continuum of dominance. 

Consider the video below of Olympic long jump gold medalist Greg Rutherford (an elite sprinter in his own right, with a PR of 10.26).

Greg is an extremely fascially-dominant athlete – and might be placed at the extreme end of the fascial-muscular continuum, if it wasn’t for his impressive absolute strength as well!  Like Bolt and Coleman, I would describe Greg as a hybrid. That said, it is plain to see his natural torsional solution (somewhat amplified by his bowlegs), and most coaches will be familiar with this type of movement.  While we tend to think of rotation, etc. in terms of upright sprinting, it is actually more pronounced during acceleration, as we can see in the below videos.  Firstly, this is Greg once again from behind. Substantially more rotation than upright! (BTW – you will not find an elite jumper whose primary strategy is not fascially-driven)


The below video is very interesting, as it shows 4 very unique movement solutions. For reference, the 100m PRs of each of the four athletes on the left (from left to right) are 10.16, 9.97, 10.11, 10.06.  

It is interesting to note that I coached all of these athletes, and their coaching instruction was quite similar.  However, it is plain to see that each athlete clearly has his own technical solution – the intention is the same, but the outcome is very different.  

Now – here is the key: there is a ‘Goldilocks effect’ between the amount of rotation and effective-efficient sprinting.  Too little, and we are constraining the effective use of our fascial slings, reflexes, etc.; while too much, and we end up blowing right through the end-points, and losing stability; not only is excessive rotation not performance-enhancing, but it is also a potential risk factor for running-related injuries. 

I should also point out that all of the above is a natural outcome of the sprinting process; just as we should not attempt to constrain it by cueing athletes to stay square, nor should we attempt to amplify it, by cueing athletes to rotate.  


The TYPE B athlete, then, is the sprinter who relies not as much on volitional force generation and frequency, but on reflex, rotation, and length.  They are good not because they can apply large amounts of force, but because they can apply large enough amounts of force in a short time. They are very efficient ‘force amplifiers’, and tend to look like they are bouncing down the track.  

They may not be the greatest ‘starters’, but they finish races well.  


So where does all this leave us?  

Essentially, when coaching technique, we have four options: 

  1. Do nothing
  2. Apply one model to all athletes, despite their inherent differences
  3. Design an individualized model for every athlete
  4. Mailbox athletes into categories based upon the commonalities of how they ‘most-naturally’ solve movement problems

In this article, I have introduced the two mailboxes I use to categorize sprinters.  Understand, however, that most elite sprinters share far more commonalities than differences, and most will bias towards the TYPE B end of the continuum (developmental sprinters, slower sprinters, or still powerful athletes who sprint in others sports – such as running backs in American football – may bias more towards the TYPE A end of the continuum). 

The key is not necessarily coaching athletes into different solutions, but in appreciating that different outcomes will manifest from similar intentions. Individualization and categorization is not permission to disobey mechanical principles. 

“All you used to hear a coach say was, ‘Go, go, go.’ Running was thought of as voodoo. One guy was faster than another because of some mysterious power he had,” he says. Eventually [Coach Tellez] noticed that the fastest runners had certain things in common, and it occurred to him that running freestyle was haywire: “I wanted to create a scientific model every runner could use.”

As I alluded to earlier, the intention remains consistent across athletes, and in this case, can be described by the three ‘first principles of acceleration’:

  1. PROJECTION: project your CoM maximally (note that projection is task-dependent. In block-clearance, it is maximal – but in other acceleration tasks, it may not be (for example a 400m start)
  2. RHYTHM: increase the rhythm of your steps successively 
  3. RISE: raise your CoM with each step

For a TYPE A sprinter, we would expect faster rhythms, lower recoveries, and less rotation.  These athletes will generally respond better to ‘push-type’ cues (e.g. “push the track back”).  For a TYPE B sprinter, on the other hand, we would expect slower, more patient rhythms, with increased air time, and more whole-body rotation.  These athletes will generally respond better to ‘air-based cues’ (e.g. “drive the thighs forward”).

The outcome of these 1st principles – regardless of athlete – should have far more commonalities than differences.  I tend to focus on the shapes an athlete creates first and foremost.  The shape is “the stable state an athlete occupies in space”.  The stable states in acceleration are at the ‘toe-off’ and ‘touch-down’ positions. 

These shapes are the ‘attractors’ to which athletes’ limbs ‘self-organize’ towards during gait. It is important that we coach to these stable shapes, and not the fluctuating paths between them (i.e. ‘dragging the toes’, or ‘low heel recovery’), which are substantially more variable, and require far more conscious control. 

Consider the following four silhouettes of elite male sprinters (block clearance toe-off), all of whom have ran sub 10 seconds in the 100m:

While there are some differences between these shapes, there are some obvious similarities:

  • the stance-leg has forcefully extended 
  • the swing-leg thigh has flexed forward and upward
  • the swing-leg ankle has flexed in anticipation of initial ground contact
  • the arms have flexed and extended to counterbalance the legs
  • the head is in alignment with the torso

Most coaches notice the dissimilarities between runners. I was looking for similarities.

Coach Tom Tellez

Consider now, the following silhouettes of these same sprinters at initial ground contact:

Again – the similarities in these shapes (the overall ‘posture’) are far more evident than the differences. 

  • the stance-leg foot is under the center-of-mass
  • the stance-leg foot is behind the stance-leg knee, and underneath the swing-leg knee
  • the torso angle is fairly perpendicular to the stance-leg shin 
  • the arms continue to flex and extend to counterbalance the legs
  • the head remains in alignment with the torso

Again – these commonalities are what we should be coaching towards.

Further, as the athlete continues to accelerate down the track, we should observe the following: 

  • Projection angles increasing with each step
  • Ground contact time decreasing with each step
  • Step-length increasing with each step
  • Flight time increasing with each step

(This is where my acceleration principles of projection, rhythm, and rise all interplay) 

It is important to note that most of the time, we will observe these characteristics happening quite naturally, regardless of the athlete and task.  

Athletes will find the most appropriate recovery parabolas based upon their individual morphologies. Some athletes can apply high forces through short distances – so they will naturally gravitate towards a lower recovery; others need a little more space/time, so will recover higher.  Why try to unnaturally influence this?

Part of our coaching objective is to teach athletes how to honestly express their ‘most-authentic’ movement – i.e. to facilitate the most-natural individualized solution. 

The coaching of the mechanics of sprinting can get stuck in the weeds quite quickly, and as an industry, we have tended to bias towards over-complication of an inherently natural ability, but we should remember that the best coaching is the most-simple.  It respects the natural efficiency that lies within all athletes. It is rooted in the appropriate manipulation of the coaching space so that the task is acknowledged as a problem to be solved by the athlete – not by the coach. 

Most-importantly, we must appreciate that all of the above is individual to each athlete: we must respect the unique abilities of the person in front of us, and in no scenario attempt to apply a technical solution that does not respect these abilities.  

If the athlete accelerates powerfully, with fast turnover, and doesn’t rotate a ton – as long as she abides by the 3 basic principles of projection, rhythm, and rise, and displays appropriate shapes and postures – then we should be ok with this. 

Conversely, if the athlete spends a little longer in the air, with a little more rotation, and increased knee flexion through swing – as long as she abides by the 3 basic principles of projection, rhythm, and rise, and displays appropriate shapes and postures – then we should be ok with this. 

We coach to the athlete’s solution – not our own.  

Thanks for reading,


Thanks again to all who contributed to this series – whether through feedback, editing, research, or guidance. Special shout out to Dr. Ken Clark, Kebba Tolbert, and Dan Pfaff


The most-challenging part in all of coaching: once we understand what an athlete does naturally, how do we know whether this is the most-efficient and most-effective way for this particular athlete to move?

I may attempt to answer this question in upcoming articles.  Stay tuned. 


Stop Trying to ’Fix’ Athletes!

I’ve never been comfortable with the term ”corrective exercise”. In this short post, I’ll tell you why.

Coaching Chat #3: Coach to Olympic Gold Medalist – Sue Humphrey

Let’s Connect

ALTIS 360 Logo


Only $9.99/mo

Gain exclusive access to 100s of ALTIS education videos and join a global community of coaches and athletes.