How should the foot be lifted from the ground?

When running on a level surface in the absence of wind resistance, momentum carries us forwards. According to Newton’s first law of motion, maintaining a constant velocity does not in itself require any input of energy. However, we do have to put energy into moving our legs forward quickly enough from one stance to the next to avoid a face down crash.

The process of moving the leg forward from stance to take up stance again with the foot beneath the body about 300 milliseconds later can be subdivided into three segments: lifting the foot off the ground, swinging it forward and allowing it to fall to the ground. These three phases merge into each other, but nonetheless, each has its own characteristic role in the gait cycle, and it is worthwhile to try to examine the requirement of each phase.

Because it is wasteful of energy and potentially injurious to land any more than a slight distance in front of the centre of gravity (COG) the swing forwards should be as passive as possible (though maybe when we want to run very fast, we do need to put a bit more energy into the swing). Because the foot needs to be going backwards relative to the COG at foot-strike, it is also best if the foot-fall is a passive as possible, though of course, substantial muscular contraction is required to stop the knee collapsing on impact.

The involuntary muscular effort required to prevent the knee collapsing is not directed at moving the leg forward, so the major muscular effort to move the leg forwards must be provided in the lift off phase. In this post we will consider only this phase. In later posts I will return to the question of what should happen in the other phases; though from the point of view of conscious effort, the answer is probably: ‘very little’.

There are various different ways we might use our muscles to lift our foot from the ground. Let us start with the issue of how we might lift the left foot from the ground while standing stationary. It is instructive to consider three ways in which we might do this. I will describe the intentions, sensations and results of each of the three ways of doing it:


(1) Start with feet side by side; then think ‘Lift the ankle to the hip’. The left foot travels almost perfectly straight up, with only a slight initial swing backwards, so that the mid-point of the left instep passes just behind the medial malleolus of the ankle of the right foot. I feel a contraction in my hamstrings, and also a weaker amount of activity is hip flexors, especially sartorius (a thin strip of muscle that runs down and across the front of the thigh.)


(2) Start with feet side by side, think ‘lift the knee’; The hip flexors, especially rectus femoris, the large muscle on the front of the thigh, contracts; the foot comes almost straight up, but with slight forward swing, and the instep passes just in front of medial malleolus on the right ankle.


(3) Start with the active foot about 12-15 inches behind the support foot, with the ankle in a neutral position; think: ‘Lift foot to hip’. The foot travels upwards just a little behind the direct line to the hip, as the hamstring contracts. This feels as if it is an almost pure hamstring action. I have no awareness of contraction of either sartorius or rectus femoris when I do this.


Which of these is nearest to the required action when running? We can rule out action (2) immediately. First of all, several studies in which electrodes attached to the muscles have been used to record muscular activity during running of runners reveal relatively little activity in the quadriceps during the lift-off (see for example, Pilutsky and Gregor, Journal of Experimental Biology, vol 204, pages 2277-2287, 2001); although some studies such as the ingenious study by Modica and Kram (Journal of Applied Physiology, vol 98, pages 2126-2131, 2005) do indicate that the hip flexors might play some role. While it may be that the runners who participated in these studies were not using the most efficient style, we would be unwise to try anything that is too far different from what runners typically do. We can reasonably assume that at even in modern society we are not so far removed from the natural human condition that modern man (or woman) has made a really radical shift from the optimum running style. Our goal in seeking an efficient style is largely to remove any relatively minor inefficiency that might have crept in as a result of modern life style (and shoes). Therefore, it seems sensible to assume that hip flexors play at most a relatively minor role.

So we are left to consider methods (1) and (3), both of which rely largely on the hamstrings. Brief consideration of the consequences of spending at least the minimum reasonable time on stance suggests that (3) is most appropriate.

When on stance, the body will necessarily rotate in a ‘head forwards and downwards’ direction, as a result of forward momentum that carries the body forwards while the foot is anchored. If we are moving at 4 metres per sec, immediately after foot-strike the COG will continue to move forwards at 4 metres per sec, while the foot is stationary. Assuming the COG is approximately 1 metre off the ground, the line joining COG to the foot must start rotating at 4 radians per second (a radian is the ‘natural unit’ of angle and is approximately 57 degrees). Assuming a cadence of 180 strides per minute, the total duration of each stride is 333 milliseconds. Unless we are prepared to suffer ground reaction forces exceeding 3 times our body weight, we must remain on stance for at least one third of the gait cycle in order for our weight to be supported, so we must spend at least 110 milliseconds on stance. During this time, our COG will have moved forward about 0.4 metres (somewhat more than a foot.). That is why when I described method (3), I had specified starting with the left foot about 12-15 inches behind the right foot. In fact, most runners will remain on stance longer than 110 milliseconds; the foot will be even further behind at lift off and the hip will be further extended. However, the crucial point to make at this point is that if we land with the foot under the COG (which is desirable to minimize wasteful braking), by time of lift-off from stance, the hip must be moderately extended. A pure hamstring pull without hip flexor action will now bring the foot up towards the hip without any action of the hip flexors.

In general, action of the hip flexors would increase the risk that we will over-stride at the next foot fall, unless we actively employ the hamstring in the late swing phase to arrest the swinging leg. It seems inefficient to use hip flexors at lift off, and then have to apply hip extensors in late swing to correct for this. Perhaps if we are aiming for maximum speed, we might have to be prepared to spend a bit more energy to get the swing moving forwards more quickly, at the price of needing to do extra work later in the swing to arrest it, but this appears to be an expensive way to run.

So these considerations suggest that a pure hamstring pull starting from a position of moderate hip extension is the most economical way to perform the lift off. Maybe active hip flexion is necessary if we want to run really fast, but this active flexion will come at extra cost.

I think that these considerations have led us to an action largely consistent with the proposal of Dr Romanov (founder of the Pose technique) who states that the pull should be performed by the hamstring. However, I am a little puzzled by one of the drills that Dr Romanov recommends in order to help us learn the correct direction of the pull. In the tapping drill, he recommends starting with the feet side by side, and as far as I can establish, the recommended action is very similar to what I have described in action (1). It should be noted that my knowledge of Pose comes only from Dr Romanov’s book ‘Pose Method of Running’ and from information on the PoseTech website ( I am looking forward to my first workshop with Dr Romanov in March of this year, and maybe I will understand the tapping drill better after that.

So for the time being, I think that if one wants to learn the correct muscle action for economical lift-off using a drill performed when stationary, it might be more sensible to start with one foot 12-15 inches (30-40 cm) behind the other. I have previously posted (in the side bar of this blog) what I describe as the ‘swing drill’ to encourage the appropriate action for moving the leg forwards from one stance to the next when running. Today, I have modified this drill to recommend starting with one foot about 12-15 inches behind the other. I must emphasize that this suggested drill is experimental. I am still working on its development. It has not been tested on anyone apart from myself and cannot yet be recommended as a safe or useful procedure. Nonetheless, I would value any comments on this proposed drill.


5 Responses to “How should the foot be lifted from the ground?”

  1. Simon Says:

    I think the reason that your version of tapping feels like it isolates the ham is because gravity is providing the hip flexion that would otherwise need to come from muscular activity. I’m not sure this mirrors the situation when we run. Once running, the leg will be moving rearwards relative to the torso which causes hip extension; that hip extension needs to be recovered through muscular activity and is done so as part of the pull (or lift). Thus the Pose tapping drill with emphasis on the hamstring but performed vertically means that the hip flexion is built into the movement – the drill thus ingrains the correct movement pattern.
    My interpretation of the Pose tapping drill is that it aims to have a short quick pull of the hamstring and gentle subconscious guidance with the hip flexors, typically only up to mid-calf of the support leg (30cm or so) followed by relaxation which allows the leg to move up some more via its momentum before descending. The perception ‘lift the ankle’ is thought to be sufficient to gain the right amount of hip flexor activity once the movement is properly initiated with the hamstring.

  2. canute1 Says:

    Thanks Simon. I think that it is inevitable that when running, the hip will be extended at the start of the pull, and therefore, virtually no hip flexion is required to ensure the foot moves towards the hip. On the other hand, in the tapping drill proposed in the Pose method, a modest amount of hip flexor action is essential if the foot is to travel directly towards the hip, Therefore whatever the cause of the hip extension at the commencement of the pull when running, I think it is more logical to practice a move that requires minimal hip flexion.

  3. Simon Says:

    The dynamics of the vertical pull are in my mind very similar to the dynamics of a running pull whereas an exclusive hamstring pull is significantly different and will not produce the desired Pose result (it may produce a result you are happy with though).

    Pose works well for several reasons, one of which is that the direction and intensity of the pull returns the foot to support under the COG without any further muscular activity. The direction is provided to a degree by the hip flexor involvement. If you pull exclusively with the hamstring, your foot will end further back and then if nothing is done to correct the position, it will not drop under your COG but behind if left to its own devices.

    Your drill and a dynamic running pull are not the same and the reason is primarily the rearwards momentum that the foot/leg has when running compared to the opposite in your drill where gravity provides hip flexion during the pull. It’s an interesting idea but not one I would encourage unless you seek to ingrain an exclusive hamstring movement. I believe this will then lead your running style to incorporate a mechanism to move your foot forward following your pull – either swinging the hip or using the quad to extend more forwards or possibly something else.

  4. canute1 Says:

    Simon, I agree with your statement that gravity will promote a tendency towards hip flexion in my ‘swing drill’ though I think the same is true when running. In the early part of the airborne time (ie as the foot is pulled towards the hip, the torso actually moves upwards, so gravity will tend to promote hip flexion. In fact, I believe that in some circumstances it is actually desirable to have some hip flexion quite early in the airborne phase when running. I think it is virtually essential if you really want to run fast. However, when developing a relaxed style with the goal of minimizing injury, I think it is best to minimize hip flexion at this time.

    However, I accept that this is all mere speculation, as I believe is your the assertion that the foot will drop behind the COG if the pull is done purely by the hamstring. The events occurring during the swing and foot-fall phases will also play a role. As touched upon briefly in my article on gravity (Jan 10th), these processes cannot be purely passive, as the torso is falling freely downwards in late swing so some force is required to make the foot fall faster than the torso. As I state, this should not be a conscious action, but should be left to the brain’s automatic adjustment mechanisms. I believe that the adjustments made at this stage play a large part in determining where the foot falls.

  5. Simon Says:

    It would take some complicated modelling to show if gravity would help hip flexion whilst running, so I guess we are left with educated speculation – one step up from mere speculation I hope! At cruise paces, you could be right I think. At sprint pace I doubt it very much, looks like we will agree to differ on this point though. I do agree that hip flexion is important in running though.

    My comments regarding leg return to support come largely from reviewing Pose running clips on the Pose site. A late, poorly powered or poorly directed pull tend to have the same effect of leaving the foot behind in mid air rather than under or close to the hips. It takes more effort to then recover the leg and get the foot down in time to land.

    I read yesterday’s blog (Jan 10th) and completely agree about the subconscious directing the movement following the pull and it being best suited for the job – well done and keep blogging 🙂

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: