Is there a magic running cadence?

The six posts in my recent discussion of running mechanics, starting with my presentation of the equations of motion of the runner on 16th of January, have elicited 372 comments (including my own responses to the comments of others).  I have been delighted by the vigour of the discussion, but am intrigued by the fact that of these posts, the one which elicited the least comment was my post on the increased efficiency associated with increased cadence, on 6th February, which elicited only 5 comments.   I suspect that this relative paucity of comments reflects a widespread acceptance that increasing cadence does improve efficiency.  The major issues in the other five posts were related to the question of the push that is required to get airborne.  This appears to be a far more controversial topic.

From my own perspective, the controversy regarding push rather than cadence is a peculiar inversion of the uncertainties of running mechanics.  The fact that a large push is required to get airborne can be demonstrated by simple application of the laws of physics, and is readily confirmed by examination of force plate data.  A large vertical push makes it possible to minimise braking.  However, elastic recoil can produce at most 50% of the required energy for the push, so the vertical push is not cost free.  I suspect that the controversy about push exists because many runners, especially those who have adopted the Pose style, have found that they suffer less injury when they do not focus consciously on pushing.  Of course avoiding thinking about the push does not stop it happening.   But the evidence does suggest that avoiding thinking about it does reduce the risk of some common types of injury.  My own view is that denying the occurrence of a large push creates a different set of risks, and therefore I think that the challenging goal of developing a safe efficient running style is creating a mental image that allows a runner to avoid a mis-timed push and other associated unnecessary muscle activity, without the need to deny the existence of strong push.

The question of how to develop the optimum mental image is a question I will certainly return to in future.   The question that currently intrigues me is the apparently widespread acceptance that high cadence is generally good (a view that I myself advocate, but with reservations).  This view does not account for the clear evidence that most runners employ a relatively low cadence at low speed and increase as they increase speed.   While it is commonly believed that a cadence of 180 steps per minute (or 90 gait cycles per minute) is the optimum cacence, it is noteworthy that the representative runner depicted in figure 2 of Weyand’s paper (J appl Physiol, 89, 1991-1999, 200) increases cadence for about 144 steps per minute at a speed of 3.5 m/sec to 234 steps per minute at a speed 9.5 m/sec.  In my experience, these values are typical.   While many runners have a cadence around 140 steps /min or less when jogging, elite sprinters usually exceed 250 steps/min at top speed.   Therefore, the view that there is a target cadence of180 steps/min only corresponds very loosely with what runners do.

There is an optimum cadence for a given speed and peak vGRF

To estimate the most efficient cadence for a particular speed it is necessary to compute all three of the major costs of running: elevating the body, overcoming braking, and re-positioning the limbs.  While the combined costs of elevating the body and overcoming braking generally decrease with increasing cadence, the costs of repositioning the limbs increases with increasing cadence and also with increasing running speed (see calculations page, in the side bar, where I demonstrate that a fairly accurate estimate of the repositioing costs per Km per Kg body mass is given by 1.32CV Newton-metres where C is cadence in steps / min and V is running speed in metre/sec  ).  Therefore, for a given speed and peak value of vertical Ground Reaction Force (vGRF), there will be certain cadence at which the total energy cost will be minimised.  In other words, total energy costs decrease as cadence increases up to a certain point, but after the point at which the increasing cost of accelerating the swing leg outweighs the saving in the sum of elevation and braking costs, further increase in cadence will lead to greater costs.

Optimum cadence depends on ability to push

However, there is no single optimum value for cadence. The optimum cadence depends on one’s ability to exert a well timed strong push.  Elevation and braking costs decrease with increasing peak vGRF at a particular velocity, so the cadence at which the repositioning cost outweigh the elevation and braking costs at that velocity, will occur at a lower cadence in a runner who can exert a stronger push.  As the cost of accelerating the swing leg is lower at a lower cadence, peak efficiency will be greater in a runner who is capable of exerting a greater peak vGRF thereby achieving peak efficiency at lower cadence.  In other words, we can increase efficiency by developing the ability to exert a stronger push, provided the push is delivered at the right time and without producing unnecessary contraction of other muscles.

A comparison of age with youth

The computations that I presented on 6th February, clearly demonstrated that at a speed of 4 m/sec, the combined cost of overcoming braking and getting airborne is less at a cadence of 200 steps per minute than at 180 steps per minute, when the peak vGRF is 3 times body weight.  In fact, I myself adopt a cadence a little over 200 steps per minute at a speed of 4 m/sec, but most runners do not adopt such a high cadence at this modest speed.  I do so because, being a 66 year old with failing muscle strength, I find it difficult to exert a push against the ground of more than 3 times my body weight without straining.   Many younger athletes can easily exceed a peak push of this magnitude without consciously trying.  I am currently trying to increase my ability to achieve a stronger, well coordinated peak push, both by means of increasing my muscle strength, and also by improving the coordination of the push.

Recently Ewen pointed out on his blog that in setting the British 3000m indoor record of 7:40.99  in Glasgow in 2009, Mo Farah exhibited a cadence of only 176 steps/min in mid-race when he was covering each Km in about 2:35 (almost 6.5 m/sec).  He did increase to a cadence of around 187 in final few laps.

It appears that Mo is able to achieve a high efficiency at a relatively low cadence.  This demonstrates that he is capable of exerting an exceptionally strong, well coordinated push.


In summary, while the combined cost of elevation and braking decrease with increasing cadence, the cost of accelerating the swing leg increases with increasing cadence.  The total cost of elevation, braking and accelerating the swinging leg will decrease as cadence increases up to a certain limit, but beyond the point where rate of increase in swing costs outweighs the saving in elevation and braking cost, increasing cadence results in increasing cost.   A runner who is able to deliver a well-timed large push without simultaneously contracting unnecessary muscles can achieve peak efficiency at a relatively low cadence.

74 Responses to “Is there a magic running cadence?”

  1. jhuff Says:


    My cadence is a very constant 198spm from 15min mile down to 4 min mile pace. Only when I go at sprint speed does it ultimately increase. Do you think that is good or bad? Personally I find it the most enjoyable cadence in my lifetime. Prior to using pose method my cadence was 175spm at slow speeds and 180s at faster paces. I really find the constant cadence to be more efficient in my experience.

    • jhuff Says:

      Had one of my faster interval sessions yesterday. I intended to check my cadence but forgot. To focused on running relaxed at 4:10mpm pace I guess. The workout did go as planned and I felt like I had my usual 198spm. I really enjoy the non-focus on pushing off. Infact I hardly feel vertical movement. The perception of releasing posture and recovering my leg/foot as I run is far superior to me than attempting to actively extend the hips and push off forcefully.

    • canute1 Says:

      A cadence of 198 at 4:10 mpm pace sounds good to me. As you know, I do not recommend a deliberate focus on pushing off except for elite sprinters, so what you describe makes reasonable sense for a relaxed but fast interval session. As I have been discussing with Hans and Simon in the comments section on the Natural Running post recently, when trying to go fairly fast (that is fast-ish for a deteriorating 66 year old), I personally focus consciously on a brief, quick descent of the contralateral arm, aiming to bring the hand rapidly from upper chest height towards the waist, with lightly opposed forefinger and thumb. I have spent some time trying to develop a non-conscious motor programme that coordinates the contralateral arm and leg. I also am aware of initiating an early hip flexion to initiate the swing. Maybe this is similar to what you describe as recovering your leg/foot. However I am afraid that these days I never get to 4:10 mpm pace – even when sprinting.

      • jhuff Says:

        Canute, personally I don’t see the arms as controlling the leg movement as you do. I see them as functioning independently of each other yet it is in our best interest as a whole that they function in harmony each doing their job

      • jhuff Says:

        Canute, personally I don’t see the arms as controlling the leg movement as you seem to think. I see them as functioning independently of each other. Though we have the ability to move them seperate and independent it doesn’t lessen the fact that we move best as a whole if they are working in harmony with each other and our body as a whole as we run or do any activity. We seem to disagree on what that harmonious relationship consists of though I am not exactly sure where we split ways at this point.

  2. canute1 Says:


    Thanks for your comment,

    I do not think it is ‘bad’ to have a constant cadence over a wide range of speeds. I also show less variation over a range of speeds that would be predicted from a simple consideration of the energy costs. I think there are several reasons for this.

    First, my calculations only give an estimation of the energy required to accelerate the swing leg, elevate the body and overcome braking but do not determine what proportion of this energy might be obtained via elastic recoil. In general we can expect that between 35% and 50% of the required energy will be obtained via elastic recoil. In the case of the swing, the extension of the hip at lift-off preloads the hip flexors and therefore makes the hip flexion that helps drive the swing more efficiency. At the end of swing, the decrease of hip flexion preloads the hip extensors and therefore makes the subsequent contraction of the extensors in early stance more efficient. Thus it might be that we can adjust the different aspects of the gait cycle in a way that maximises the proportion of the total energy cost that can be met by recovery of elastic energy. Calculation of recovery of elastic energy would be very complex, so I make the rather simplistic assumption that after practice, we acquire the ability to recover 50% of the cost via elastic recoil, but that is certainly over-simplistic. Maybe if you are very skilled at a particular cadence, you might have taught your body to acquire the maximum energy from recoil at that cadence.

    Furthermore, we appear to have the ability to adjust cadence voluntarily and we might over-ride the principle that efficiency should be maximised. As I discussed on Jan 16th, under many circumstances, efficiency increases if we increase vGRF, but under circumstances where fuel efficiency does not matter, we might choose to exert a lower force. For example, when cooling down after a race, avoiding large forces might matter more than saving a few grams of glucose. If you feel most comfortable at a particular cadence i see no reason to alter this except in circumstances where maximising fuel efficiency really matters As you yourself point out, you do increase cadence when you increase to very high speeds. Perhaps if you take up ultra marathon running, it might be best to see if a somewhat lower cadence is more efficient at low speeds.

    In my own case, my cadences at various speeds have tended increase as I have grown older and I am fairly sure this is due to loss of muscle power. In fact my efficiency (as crudely estimated from heart rate at different speeds) has deteriorated only slightly, so it is probable that I have compensated for loss of power by increasing cadence at moderate speeds. However, I have less scope for further increase in cadence at high speed. My top speed has decreased greatly. I hope that by increasing muscle power I can recover some of that speed.

    • jhuff Says:

      Canute, fairly confident I will not develop a Passion for ultra running so will have to go by what feels best for current slower running distances. Today ran 3mi at 12:14 mpm pace, at 101bpm heartrate, and cadence of 198spm. If I were to take up ultra running I can’t foresee and theoretical reason for me to purposefully alter my cadence and technique. Will just have to be satisfied with speculation for now…..

      • canute1 Says:


        If HR is a reliable guide, I think I might be a bit more efficient than you at 12:14 mpm pace My HR is well below 100 and my cadence is typically around 180-185 steps/min at that pace. Maybe I should challenge you to a 100Km race – but I certainly will not be challenging you to a 5K or even a marathon 🙂
        I accept that you are not aiming for an ultra-marathon in the foreseeable future. The reason I make fantasy suggestions about how you might run an ultra is to draw attention to the fact that one needs to consider that it might best to modify running style according to circumstances.

      • jhuff Says:


        If you and I were in a 100km race I would not be 12mpm pace nor worrying about my heartrate 🙂

      • canute1 Says:


        Although my response to your heart rate at low speed was framed as a joke challenge, there was an underlying serious point. After allowing for effects of stress sand emotion on the adrenergic nervous system, the main determinants of heart rate during running are 1) the cardiac stroke volume, 2) the ability of muscle to extract oxygen from the blood, 3) the ability use oxygen to generate ATP to fuel muscle contraction 4) the efficiency of converting metabolic energy to mechanical energy and 5) the efficiency of using mechanical energy to propel the body. I suspect you are exceptionally good at steps 1 to 4. I am not so sure about step 5. In particular, at low speed, I think you could produce a small improvement by decreasing your cadence. However I also accept that the effect would be small because swing cost is only a minor fraction of energy cost at low speed. However if you do ever develop an interest in running ultras, I suggest it would be worth experimenting with a lower cadence.

    • jhuff Says:


      “””””A runner who is able to deliver a well-timed large push without simultaneously contracting unnecessary muscles can achieve peak efficiency at a relatively low cadence.””””””

      This seems to imply that one would sustain a large push off even at slow speeds. Is this what you mean?

      • jhuff Says:

        Oh…and what range are you refering to when you say “relatively low”?

      • canute1 Says:

        I do not mean that one should sustain a large push off even at slow speeds. I mean that if you can achieve a well timed push you can maintain a relatively low cadence at high speed. For example, in his world record breaking run in the 100m World Championship Berlin in 2009, Bolt maintained a lower cadence at peak speed (264 steps/min) than Tyson Gay (288 steps/min) and in winning Gold in Beijing he had the lowest cadence of all the runners in the final.

        ‘Relativley low’ depends on the pace. At elite sprinting speed, I mean 264 rather than 288; in a 3000m I mean 180 rather than 200

      • jhuff Says:


        By “well below” you mean? They don’t give out awards to those with the lowest HR 🙂

      • jhuff Says:


        BTW, that is 50% of my max heartrate for a 3mi run at 12mpm. What is your percentage of maxhr for an equivalent run?

      • canute1 Says:


        I rarely record my HR at such low speeds. The last time I did it was 85 at 12 min mile pace. I am not as fit now, so I would guess perhaps 95 now. In fact HR is an unreliable measure of efficiency and should not be over-interpreted. You might have noticed the smiley face after my suggestion that I should challenge you to a 100Km race.

        However, it is possible that you might get your HR at low speed down a bit if you lowered your cadence..

      • jhuff Says:


        Considering 50% of maxhr is ridiculously low to begin with I can’t think of any practical reason to train to have it lower. Now if that starring giving out cash prizes I might entertain the thought 🙂

      • canute1 Says:


        Insofar as HR is an indication of energy consumed (which it is, despite being an erratic indicator because it is subject to other influences) then you might find that lower HR indicates lower energy consumption which would be an advantage in our 100Km race 🙂

  3. Ewen Thompson (@EwenThompson) Says:

    Canute, that’s an excellent point you make about there being an optimal cadence for a given speed (for each individual), beyond which, increasing cadence becomes less efficient (uses more oxygen).

    Cadence usually changes (perhaps Jeremy is an exception) at different running speeds, even if only slightly. Steve Magness wrote a good article on how different runners ‘kick’ at the end of a race – either by increasing cadence, or increasing stride length, or a combination of both –

    On the point of ‘how to attain a stronger well coordinated push’ for the aging runner, that’s something I’m thinking about too. It’s obvious that young runners have a naturally strong push and can cope with 3 x body weight on each stride. I’m wondering if more cushioned (yet light) shoes might be helpful? For example, when running barefoot on concrete runners don’t push ‘hard’ because their body senses the surface and they ‘run lightly’ for self preservation. If wearing a well cushioned shoe, I’m wondering if a runner could land from a higher height, generate better elastic recoil and more force for the next stride with carefree abandon, not worrying (or feeling) the ground contact force?

    • canute1 Says:

      Ewen, Thanks for your comment.

      The evidence regarding the effect of shoes is mixed. As you yourself imply, the weight of the shoes will increase energy costs. This is well established by experimental evidence. However some studies show that shod runners are more efficient, including this new study.
      Overall, the evidence suggests that we can exert the most effective foot strike when wearing a light shoe that absorbs some of the impact. Shoes such as the Newton explicitly attempt to store the impact energy and return it in late stance. This gets us into the controversial area of what should be regarded as an artificial aid. It appeals to my sense of fair play in competition that the rules should favour the person with the best greatest ability exert a strong, well coordinated foot-strike in an agreed standard shoe. If I were a fanatical purist, I suppose I would have argue that we should revert to the ancient tradition of nude barefoot competition. I think this problem of kit is even greater in swimming than running. In practice, I think the IAAF should continue with the practice of outlawing anything that is too far from the norm. If our interest is simply in improving own performance, we should focus on our own muscles rather than our shoes – while wearing shoes with the minimum protection adequate to minimise sustained tissue damage.

  4. Ewen Thompson (@EwenThompson) Says:

    Canute, I saw a link to that study recently:

    I think there’s an IAAF rule about the thickness of the midsole of shoes. I don’t know if the Newton shoes are ‘legal’ or if they help, how much. They’re popular with triathletes who are into technology (fast wheels for example) to improve performance.

    One of my good races in 2007 was a 5k at Stromlo where I ran barefoot on the grass. I think the cushioning provided by the grass was about perfect so that I was able to exert a strong push against the ground yet benefited from light weight (zero grams). I couldn’t have exerted the same push on a bitumen surface. So I think just the right about of cushioning in a very light shoe would be ideal for running on the roads. I’m not thinking of any possibly illegal ‘rebound’ from the cushioning, just the shock absorbtion so that a strong push can be exerted.

    • canute1 Says:


      Thanks for that link to Alex Hutchinson’s discussion of the Colorado study comparing barefoot with shod runners. His comment on the ‘cushioning cost’ hypothesis is interesting. This hypothesis is consistent with my own experience. I am aware of maintaining less tension in muscles of the hip and thigh when running barefoot. This allows a softer landing with greater flexion of the hip and knee, but almost certainly prolongs time on stance while diminishing vGRF. This is probably good for reducing risk of injury at the expense of efficiency. More like Pose and less like BK.

      I have always been somewhat amused by the incongruity of the bare-footed runner illustrating the BK video. He is wearing nothing apart from some rather minimal budgie-smugglers. The tension in his hamstrings and gluteus maximus just before foot fall is amazing –especially in the clip on M Erector Spinae.
      I would like to know what surface he is running on – it is out of the picture, but I suspect it is not like the grass at Stromlo. His feet must be extremely resilient. I have never seen another barefoot runner looking anything like him.

  5. Simon Says:


    I’ve come to this discussion a bit late but have a comment following on from Hans’s descriptions of running in the previous post comments and on the language used here in your first paragraphs. The question it provokes in me is; what is the right language to describe running both in a physically and physiologically accurate manner?

    The vGRF of a bouncing steel ball on a concrete surface would show very high vGRF. Would you describe it as a ‘push’? Physically it is a push, but the ball cannot produce force itself, all the forces are reactive and can be traced back in terms of energetics to the initial gravitational potential that the ball had before it was dropped. So the ball is not ‘doing’ anything, it is just reacting, which is all it can do.

    Whilst runners do not have the near prefect elasticity of a spring steel ball, the additional force production that tops up elasticity is largely physiologically reactive. It comes about from strecth-shortening and from the higher brain functions that want to maintain balance and the correct height above the ground. I would call the stretch-shortening purely reactive whilst the relationship between force production and balance/height is more deateable.

    So if you agree the force production is mainly both physically and physiologically reactive, what should the force production be referred to as?
    I think if you must use ‘push’ it should at least be qualified with ‘mainly reactive’ along with qualification for its magnitude under certain circumstances to stress its physical importance.
    Statements along the lines of, ‘running undoubtedly involves a large vertical push’ are true if viewed from a particular perspective but probably misleading when viewed from a more holistic standpoint.

    Apologies if your latest artical has discussed this, I haven’t caught up that far yet.

    • canute1 Says:


      I agree that the word push can create problems. Three things are clear: first, the force that is exerted against the ground is large; secondly, a substantial fraction of the energy (perhaps around 50% ) can be derived from recovery of impact energy via elastic recoil; thirdly, a conscious push can be counter-productive.

      I think that provided one makes all three of these issues clear, push is an acceptable word. However, I do not believe that it is advantageous to attempt to exert conscious control over the selection of specific muscles when executing a skilled rapid motor action. The non-conscious motor control system is more capable of producing the appropriate fine adjustments of timing, direction and forcefulness that are required to deal with the minor perturbation in the trajectory of movement that arise.

      So the question is: what is the best way to evoke the required non-conscious tensioning of muscles? Ultimately, each individual must discover that for him/herself, but it is useful to start with some images of what is required, and also to perform some conscious adjustment of attentional focus to facilitate the required effect.

      With regard to images, it might even be useful to have a mental picture of ourselves running. Our brain has so called ‘mirror neurons’ that generate a pattern of signals similar to those associated with performance of an action when we observe someone else doing the action, so it is plausible that imaging ourselves (or someone else, perhaps Haile Gebreselassie) running, prepares the appropriate neurons to become active.

      With regard to focus of attention, it is my belief that it is helpful to train the mind to perceive all the sensations of joint movement and contact between surfaces that occur during running. Although we do not consciously focus on all these sensations at all times when running, it is possible to have subliminal awareness of many of them. Our brains are very good at detecting a departure from what is subliminally anticipated.

      We might also benefit from drills that focus on the required coordinated neuromuscular action. Among the exercises that are helpful to sprinters and perhaps also helpful to middle distance runners, is the leg drive: it involves leaning against wall with one foot elevated via flexion of hip and knee; consciously driving the foot to the ground, and then rapidly lifting it back to the starting position. Lydiard advocated bounding uphill as a drill. Pirie felt that he was bounding as he ran; perhaps he was trying to evoke kangaroo-like elastic recoil.

      When I practice (my version of) Change of Stance, my goal is to help my non-conscious control system to learn to apply a well coordinated extension of hip and knee of the descending leg and downswing of the contralateral arm (i.e. I am fostering a well coordinated drive or push of the foot to the ground). Perhaps Pose practitioners focus on the flexion of the rising leg; I focus on this to only a minor extent.

      Overall, I think that push is an appropriate word provided one accepts that when running, it is not a consciously controlled push. Furthermore, trying too hard to push is probably even more injurious that ‘over-pulling’. In endurance running, a sense of lightness and relaxation is the holy grail.

  6. Simon Says:

    Hi Canute,

    I remember when learning to ski as a school child that our Italian instructor described the way to turn as applying more pressure to one ski. I followed that literally and produced incredibly tiring slow turns by pushing down on one ski as hard as I could. Then I noticed that better skiers didn’t do that at all, they took weight off the other ski and were characterised by nimble precise movements. Applying that shifting of balance and attempting finesse rather than brute force made an immediate difference to my skiing and was a lasting lesson in movement.

    Likewise in running, I don’t focus on downward push or downward driving and I don’t believe that would be productive. Instead I focus on the opposite hand rising and the trunk lengthening and extending upwards.
    I also know that by arranging my posture correctly, my body will be both stiff and springy which will automatically account for the majority of the push through elasticity and reactive stretch-shortening. If I am not getting the springyness and my running feels laboured, I know it is because I have started to push more and am missing out on more efficient mechanisms to generate ground force.

    I think the language of pushing will tend towards forcefulness rather than lightness if it is part of a runners conscious understanding of how running works in the absence of an appreciation of physiology and the largely reactive nature of force production.
    I don’t think it will be a problem for you personally, but it may be for some of your readers.

    I think the vertical displacement in running is probably of a similar order to skipping. Skipping does not feel like an exercise in pushing, it feels like an exercise in being springy through correct timing and cadence. Skipping on one leg is much harder, but still tends towards springyness.

    In conclusion, I think the language should focus on springyness rather than push as springyness encompasses both the pushing elements and the elastic return, plus it is immediately obvious that it is a physiologically reactive movement rather than a conscious one. Just saying springy push rather than push would probably suffice, though it is a bit of a clumsy term.

    • canute1 Says:

      We are in agreement that brute force is not good, though I think that when sprinting it is useful to aim for a large but exquisitely timed force.
      I consider the term ‘springy push’ is moderately satisfactory. I think that this creates an image similar to Pirie’s concept of bounding. If the image of a springy push evokes a sense of lightness, that is fine. However too much emphasis on a large ‘kangaroo hop’ would not be helpful.
      I hope that my frequent re-iteration of the recommendation to avoid conscious control of the push will minimise mis-interpretation. However I also think that failing to fully appreciate the magnitude of the ground reaction force can create its own problems.

      • Simon Says:


        Yes, language is tricky, especially on the internet where people will often just skim read. All aspects of running are inter-related so as you say ‘springy push’ could cause problems if taken without the proper context, though I still think it better than just ‘push’ which needs more qualification to avoid misinterpretation.

        When you mention sprinting, you often mention a large well timed force. Whilst this must be true physically, I wonder if it is really a holistic aim in itself? As Klas said, the speed of sprinting demands the high force – if you cannot apply the force you will simply not go any faster.
        That may lead you to think that a direct focus on more force application will create more speed. That is not my experience though. I find more range of motion and quicker repositioning speed creates more running speed – I cannot change force production directly whilst running and have to influence it indirectly instead. I think that is a key issue, at least for me.
        I also think sprinting is a special case of running and does not particularly relate to non-maximal distance running.

    • canute1 Says:

      I think we are largely in agreement. Each person has to find what works best for him/her.
      In seeking optimum efficiency, I have frequently emphasized the importance of short time on stance. That has been the main source of my fascination with Pose. Short time on stance is inevitably associated with large force. Pose emphasises short time on stance while directing attention away for the large force. Much of my recent discussion on my blog has addressed the conundrum of the push, not because I wish to draw too much attention to the forcefulness of push, but rather, to understand what conscious awareness might facilitate a short time on stance.

      As stated in several of my posts, I think that awareness of a precise, economical arm swing is helpful, but Pose theory has not traditionally placed emphasis on the link between arm and leg. I think Pose achieves the short on stance by emphasis on an early pull. However, I find the Pose concept of pull puzzling. I do not think a shortening of the hamstring at lift-off is helpful,. However, isometric contraction of the hams will result in knee flexion as the hip flexors contract.
      So in summary, I am looking for the best way to facilitate short time on stance. For me, focus on a precise, economical arm swing helps. However, if this is to be effective, one needs to consolidate the link between the movement of hand and foot (eg by drills) and at least for the older runner, it is necessary to devote effort to maintaining adequate leg strength.

      • jhuff Says:


        “” Short time on stance is inevitably associated with large force”””

        As stated before, I run at a high cadence from 15min per mile pace down to 4:30mpm. Almost exactly at 198spm. It would appear that you are saying the force is large at both ends of the speed spectrum. Are you saying that I run with the same force at each speed? could you elaborate on this issue?

      • Simon Says:


        Yes, I think we do largely agree and the difference in launguage is a fairly minor point.

        Regarding Pose, it does not empasise a short time on stance per se. It emphasises going as quickly and effciently as possible from one Pose to the next. So the impossible ideal is to always be in Pose either on one foot or the other, instantaneously changing between each. If that was achievable, the vertical force would be a mere 1xbodyweight and we would achieve the ultimate efficiency of the wheel.
        For most Pose learners, getting quickly from Pose to Pose is achieved by an increase in cadence so they will often not suffer any additional vertical force.
        I agree in some cases, the change in style will be accompanied by more vertical forces and this should be transparent and carefully managed.

        I think arm swing is a good focus and I use it as I noted earler, ‘releasing upwards’ from stance.

        The Pose pull on the other hand has not been very useful for me, although it is possible that my arm focus creates some kind of pull.
        I’ve often wondered if the pull is partly a mechanism to stiffen the leg prior to footfall to achieve shorter stance as we cannot pull a loaded foot to achieve shorter stance.
        The Pose pull also facilitates a limit on range of motion, so limbs can quickly be repositioned for the next Pose.

        For me the best way to achieve short stance time is to get the landing right. The Pose pull is supposed to lead in to the landing without needing to focus on the landing itself, but that has not worked for me.
        I use a few postural queues to get my landing right and don’t need to do anything else as it all seems to be reactive or learnt motor patterns from foot fall onwards. The exception is perhaps leg recovery, but I find keeping a quick cadence sorts that out without trying to impose micromanagement through pulling directly.
        My many attempts to influence gait during stance have been fruitless and usually make things worse, especially where they focus on a very specific movement rather than a more general one.

        I no longer perform any drills as I did not find movement from drills translated to movement in running. On the other hand, I plan to do restart plyometrics this year as I think they are a good blend of strengthening and tuning movements plus they easily lead into running type movements.

        This is all still a work in progress, but after years of experimenting I think I have many of the basics in place. I would still hesitate to pass out advice though and would instead encourage people to find what works for them with an open questioning mind and realistic goals.

    • canute1 Says:


      Constant cadence does not imply constant force.
      Perhaps to avoid all ambiguity, I should say, when time on stance is short compared with duration of the gait cycle, the ground reaction force must be large. If half of the gait cycle is spent on stance, the average vGRF/Kg while on stance is 2g. If one third of the gait cycle is spent on stance, the average vGRF/Kg is 3g. Most runners spend a smaller proportion of the gait cycle on stance at higher speed and therefore, exert a greater force at high speed.

    • canute1 Says:


      I realise that it is dangerous to make statements about Pose principles because it is difficult to get a precise definition of what the principles are. Several Pose coaches have emphasized to me that if Dr Romanov were publish a new edition of Pose Method of Running, there would be substantial changes, but it is very difficult to get a clear statement as to what those changes would be. For example, while many Pose coaches admit the illustrations of the foot with the heel high of the ground during stance are misleading, I found myself in a comical conversation on a Pose Tech forum a little over a year ago when I attempted to clarify this issue. A Pose coach who had very strongly emphasised the importance of allowing the heel to touch the ground, when talking to me in private, defended the book illustrations in his response on the Pose Tech forum to my comments which largely reflected what he said to me in private. Hence, when I refer to Pose principles, I appreciate that I am skating on thin ice. I mean one of four things: the theory presented in Pose Method of Running; the published journal articles by Dr Romanov and colleagues; articles on Pose Tech; and finally, what Dr Romanov said at the Pose clinic that I attended in Loughborough, though I acknowledge that these four sources are not mutually consistent in all respects.

      At the Loughborough clinic, one of the dominant themes was the concept of ‘Pose standard’, meaning the achievement of a certain minimum time on stance. When Dr Romanov commented on the video recording of the participants, he invariably commented on whether or not they achieved the Pose standard of no more than four video frames on stance. As you know, I advocate a relatively short time on stance, so I took special note of this. One of the things that I hoped I would learn at that clinic was how it might be achieved. I did not obtain a clear answer, but the best I could deduce from the confusing picture presented was that a rapid pull was considered important.

      I agree with your suggestion that focussing on a rapid pull might achieve a short time on stance by virtue of encouraging the appropriate tensioning of muscles prior to foot fall. I also agree that micro-managing what happens on stance tends to be counter-productive, which gets me back to why I focus conscious attention on the hand – a part of the body that most people can control with exquisite precision. We are also in agreement that many recreational runners can improve their efficiency and safety by increasing cadence, though this post attempted to address the question: is there an upper limit?

      • Simon Says:


        The writings and presentations of Romanov and people associated with Pose are sometimes differing and even contradictory. I have my own understanding of Pose, based mainly on the interpretations of the UK coaches and discussions I have had with them to clarify various issues.
        I accept it is hard to give definitive answers though.

        Regarding cadence, I agree with your analysis that limb recovery costs will pass costs of elevation and braking costs at some point. I’m not sure what that point is or what the many variables are that will decide it. I do not know if Mo Farah had selected the optimal cadence in terms of braking, elevation and limb recovery or whether some other variables such as muscle composition and muscle/elastic condition made the selection. I don’t think this is knowable, though it was interesting to see the relatively leisurely cadence.

        What we can take away is that most successful runners running at mid pace and above select a cadence of 180 (or thereabouts) and above. By experimenting with cadence we can find what works for us in terms of efficiency and injury resitance through decreasing peak vertical forces.

        I really should get on to your next article now 🙂

  7. Klas Says:


    I suspect that we are always running below what is the optimal cadence from a simplified mechanical point of view, at least when we are below top speed.

    Looking at running on the spot as an example, it seems to me that cadence is limited by the configuration of the joints and how quickly our muscles can react without inefficient tension.

    When speed increases, the limbs move more rapidly due to the range of motion and air time increases. Both contribute to achieving a more rapid bounce that enables an increase of cadence without inefficient tension.

    • canute1 Says:

      As outlined in my recent reply to your comment on the ‘Further reflections… ‘ post, I believe that for each speed there is a particular combination of cadence and stance time that gives the lowest total energy cost. However, for most recreational runners in the mid-range of speeds (3-4 m/sec) I think it is likely that they exhibit cadence that is to below optimal and a stance time that is longer than optimal. However because repositioning costs increase with cadence, there will be an upper limit to the desirable cadence.

    • Klas Says:

      I agree that there must be an optimal cadence wrt the trade-off with repositioning. What I’m suggesting is that other limitations make a lower cadence more optimal at sub-sprinting speeds.

      I don’t think we need to worry about the cost of repositioning in practice.

    • canute1 Says:


      I think we agree that at lower speed the optimal cadence is lower than the optimal at high speed. As you say some limitation other than repositioning cost must set the limit on cadence at low speed as repositioning cost is relatively small at low speed.

      I believe the explanation is that if we maintain high cadence at slow speed, step length becomes very short and we have to compete a large number of steps per Km. There will be a large number of swings per Km. We waste unnecessary energy getting airborne for long enough during each step to complete the swing (which must be complete in two airborne periods +one stance period).
      When I have completed the planned calculations we should have the information we need to answer this question confidently.

    • Klas Says:

      In practice, I agree that optimal cadence increases with speed. I can feel it, and overwelming data shows that.

      In theory, when just looking at the simple model of cost of braking, elevation, and repositioning, it is not clear to me how the optimal cadence varies with speed. I would appreciate if you could explain. I guess the combined cost of braking and elevation increases with speed at fixed cadence, but so does the cost of repositioning.

    • canute1 Says:


      This raises a complex three dimensional problem: energy cost varies with speed, cadence and stance time (or average vGRF etc). Furthermore energy cost arises from three main sources: elevation, braking and repositioning, that are each affected by the primary variables. Thus it is not easy to visualise all of the effects of variation of the three primary variables. I have not yet done calculation with speed as a variable. However, three of the main effects that will determine the optimum cadence at each speed are: 1) elevation costs per step (and per Km) decrease with increasing cadence because the cost of elevation is proportional to the square of airborne time; 2) repositioning costs increase with cadence and speed, such that at high speed, the increasing repositioning cost of higher cadence will dominate; 3) the number of steps per Km increases with increasing cadence at constant speed, and at very low speed high cadence will result in very short steps. However I am reluctant to draw many conclusions until I have time to do the calculations.

    • Klas Says:

      I don’t think step length is the issue. When running on the spot, step length is zero. The swing foot must be above the ground. If you look at the configuration of the leg, this means the heel must be at least 1dm or so above the ground at midstance in a relaxed posture. When changing support without inefficient force, it takes a certain amount of time for the heel to reach the ground. And then it takes a bit of time to reverse the movement and begin to change support again. This seems to impose a lower limit on efficient stride time, around 1/3 second in slow running. It is certainly possible to force a shorter stride time, basically drumming the feet onto the ground, but the muscular effort outweighs the lower cost of elevation.

      When we add forward momentum, the leg moves in the horizontal plane as well, but the principle in the vertical plane is very similar.

      At speeds above 3m/s or so, the range of motion causes stretch shortening reflexes that seem to make a shorter stance time possible without inefficient tension.

    • canute1 Says:


      I believe that the length of steps does matter. For a specified cadence (i.e. a given stride duration) there will be a lower limit to the sum of elevation and braking costs per step. During airborne time the body falls freely incurring subsequent elevation costs and while on the ground he/she incurs braking costs. However at very slow speed, the number of steps per Km will become very large. So small steps are expensive when the cost is determined per Km. However at slow speed the leg is never far from vertical so braking cost is relatively small. Therefore at slow speeds it will actually be more efficient to decrease cadence and spend longer on stance. This will result in only a small increase in braking cost, while elevation costs/Km will decrease because there are fewer steps/Km.

      At even lower speed it is probably also advantageous to decrease airborne time as cadence decreases, while allowing stance time to become very long. This is what we see in Daniel Lieberman’s video of Ken Bob (posted by Robert Osfield in his comment on my post of 16th Jan.). Ken’s cadence is 170 and he spends more than 90% of the gait cycle on stance.

  8. Klas Says:

    I don’t follow your reasoning. You have argued before that “Even though the number of steps per mile (or Km) increases linearly as cadence increases, because the energy saving is proportional to the change in the square of the duration, the energy saving more than offsets the increase cost due to an increase in number of steps.”

    I time the cadence of Ken Bob in that video to 175.

    • canute1 Says:


      It is the elevation cost/Km that is more than offset by increased cadence. This is why a relatively high cadence is more efficient at medium speeds (eg above approximately 3 m/sec). However decreasing cadence at low speed saves braking costs/Km. The proportion of time on stance is greater at low speed. Consequently, the saving in braking cost achieved by decreasing the number of steps per minute is the more important effect at low speed. This was why I emphasized that in order to improve efficiency at low speed, the increased time per step made available by decreasing cadence must be employed to increase stance time, not airborne time. I think Ken Bob illustrates this clearly.

    • Klas Says:

      Thanks. So how does braking cost vary with cadence? Intuitively it seems to me it should decrease at least linearly, in which case the increased number of steps would still not outweigh the reduced cost of braking.

      We both agree that the optimal cadence increases with speed. Whether the optimal cadence for Ken Bob is his 175 or 180 is not really significant. I don’t think he is trying to optimize it.

      There is no braking cost when running on the spot, and I don’t think Ken Bob would increase his cadence in that situation. So his cadence seems limited by something else, probably along the lines I suggested above. That same limitation should hold in slow running as well.

    • canute1 Says:


      The braking cost per step is approximately proportional to the square of velocity. Therefore the braking cost per Km is proportional to the product of velocity x cadence. Thus at a specified speed, braking cost per Km decreases with decreasing cadence.

      I think that Ken Bob’s cadence is between 170 and 175, but as you say, the precise value does not significantly affect the point that we are discussing. However the fact that he has a very small airborne time is relevant, as elevation cost per step should be minimised at low speed. He might not be maximally efficient for his speed, but I think he exhibits the main features of an efficient gait for low speed: modest cadence and relatively long time on stance.

      Running on the spot with a non-zero elevation has an infinite cost per Km, and is not directly relevant to the issue of the optimum cadence for minimising the combined costs of braking and elevation per Km. However, you should note that my calculations are for the actual costs incurred. Some of these costs can be recovered by elastic recoil. At very low speed, a point will be reached where the stance time that minimises the sum of braking and elevation cost is so long that there is a loss of elastic recoil. Once we reach such a low speed, it will not be advantageous to make further reduction in cadence. I suspect this point will be reached a speed near the natural walk-run transition. When running on the spot, I suspect that the most comfortable cadence is the one that maximises elastic recovery.

      In summary:

      At very high speed, the increase in repositioning costs dominate. As repositioning costs increase with cadence, there is an upper limit to optimum cadence
      At medium speeds (eg 4 m/sec), high cadence (eg typically 180 – 200 steps/min) is most efficient, as illustrated by my calculation on Feb 6th.

      At low speed (perhaps around 2.5 m/sec) it is efficient to decrease cadence and increase time on stance, while decreasing airborne time.

      At very low speed (in vicinity of walk-run transition speed) further increase in stance time will result in loss of elastic recoil, so there should be no further decrease in cadence.

    • Klas Says:

      Thanks. I’m sure that braking cost per stride is proportional to velocity, but are you saying that cadence is not also involved in that cost?

      There are two kinds of braking. One is due to push being directed backwards during the braking half of stance. It seems to me that this braking should be proportionally reduced with increasing cadence, since the duration of the braking half will be shorter and the angle less backwards. The other kind of braking is that the foot is standing still on the ground. This too is proportional to cadence.

    • canute1 Says:

      The braking cost per step is approximately proportional to square of velocity because the work done by the horizontal component of GRF in the first half of stance is the integral of hGRF multiplied by velocity over that time interval. (This is equal to the integral of force x distance during the first half of stance since distance travelled in a time interval is velocity x time.)

      The average value of hGRF is approximately proportional to velocity because the degree of obliquity of the leg at each time point increase in proportion to velocity. Thus the product of force by velocity integrated over time is approximately proportional to the square of velocity.

      Thus the work done to overcome braking per step is determined by velocity, GRF and time on stance and does not directly depend on cadence, though cadence plays an indirect role insofar as at higher cadence GRF is usually lower. To estimate braking cost per Km it is necessary to multiply cost per step by number of steps per Km. Since number of steps is proportional to cadence/velocity the braking cost per Km is proportional to the product of cadence by velocity. At a specified velocity, braking cost increases in proportion to cadence. At low speed, when the number of steps per Km is large, but repositioning cost is small, the most efficient strategy is to decrease vertical impulse thereby minimising elevation cost, and decrease cadence thereby decreasing braking cost.

      The braking produced by the foot on the ground is included in the braking cost due to obliquity of the leg. While the foot is on the ground ahead of the COG, the obliquity results in braking. At the instant when the COG passes over the point of support there is no horizontal force and at this instant there is no braking; once the COG is ahead of the point of support, the fact that the foot is pressing on the ground produces forward propulsion that compensates for the braking that had occurred in the first half of stance.

    • Klas Says:

      Thanks. By degree of obliquity, you mean the angle between the point of support and COG/hip?

      • canute1 Says:

        Klas, Yes. When I refer to greater obliquity, I mean that the line from point of support to cog is displaced further from vertical. This line is vertical at mid-stance. For a specified time on stance, the COG moves further in the period from foot strike to mid-stance when the velocity is higher. Therefore the line from support to COG is more oblique (ie further from vertical) at foot strike.

      • Klas Says:

        I see what you mean, thanks. As we have discussed elsewhere, the air time ratio typically increases with speed due to the flexling swing knee, but that effect is probably small at slow speeds.

        But similarly, obliquity should increase with reduced cadence, since stance time increases.

  9. EternalFury Says:

    This blog is like a candy shop for running nerds, isn’t it?
    Being a nerd myself, I am drooling at the many posts.

    This being said, I think theory needs grounding and empirical evidence to strengthen itself.

    In the case of cadence debates…how do you measure in practical terms the efficiency of running at a certain speed with a certain cadence?

    What practical (without requiring that you carry a lab on your back while running) formula do you use to compare the efficiency of (v1, c1) to the efficiency of (v1, c2), or (v2, c1) or (v2, c2), etc.

    Until we come up with *practical* means of validating theories, most arguments and debates are time wasted.

    I am not bashing the approach, I want to see its foundation.

    • canute1 Says:

      Thanks for your comment. I agree that to convert this ‘candy’ into something nourishing, a runner needs a practical way of determining the most efficient cadence for any given pace.

      The problem is that it is a three dimensional problem: the optimum cadence depends on both pace and stance time (which in turn depends on ability to exert a well timed push against the ground). When it comes to complex adjustments of several different variables, it is generally best to let the brain perform the adjustment non-consciously. We usually do more harm by consciously imposing some predetermined cadence (or stride length). The main point of my post was to challenge the widely the view that there is a magic cadence (considered by many distance runners to be around 180 steps/min). The theoretical calculations presented in my blog, and also the evidence obtained by observing elite distance runners as they increase pace in the final lap of a 5000m or 10000m, is that cadence should increase as pace increases. For most runners, the brain does this adjustment automatically, provided the runner does not interfere by consciously forcing an increase of stride length.

      However if you are concerned that your brain is not making the optimum adjustment and you do not have access to the equipment to measure expired gas composition when running on a treadmill, it might be helpful to measure total heart beats in the second Km of each of three 2 Km runs: at ‘natural’ cadence; at consciously greater cadence; and at consciously decreased cadence. All three runs should be at a similar pace over the same course. However if you experience appreciable HR drift after the first Km when running a constant pace and cadence for 6 km you need to allow for this.

      • EternalFury Says:

        Thank you for your reply!

        I think we are mostly on the same page, then. My current quest is to collect *practical* protocols to validate training interventions.

        Is my endurance improving? Is my running economy improving? Is my lactate threshold rising? Is my cardiac output changing? Is my body composition optimal? What is my level of chronic fatigue? Did I fully recover from my previous bout of training? Is my level of flexibility sufficient? Etc.

        Again, not to sounds like a broken record, but all of these questions and many more must be answered using *practical* means in order to track progress brought about by any training regimen.

        Otherwise, you are just following some arbitrary training plan, as thousands of runners out there, and you simply cross your fingers and hope that what worked for someone else is actually adapted to you. (It rarely is, although it can bring about 70% of the adaptations you generally seek.)

        I am not anti-theory, I do think understanding the physiology that underlies these more practical measures is essential. Yet, I think physiology and training are worlds apart when it comes to their practitioners.

        Finally, to return to cadence, running economy and your suggestion. What you said made me smile because I have used the following chart in an attempt to correlate my cardiac output with my cadence:

        This is from a field lactate threshold test I did last Wednesday. You can hover over various parts of the chart to see additional data.

        I have used this chart among others to validate various “facts”. When it comes to cadence, popular wisdom and evidence (in my case) does not necessarily match. To put it simply, my use of energy is not necessarily optimal when my cadence is exactly 180.

        You can see it in this chart, and I can see it in many other instances of this chart.

        My current assumption is that one’s optimal cadence may largely depend on the biomechanical features of the running body. Particularly, in terms of lean body mass, muscle strength and muscle elasticity.
        But this is more intuition than a finding I can easily verify.

    • canute1 Says:

      Thanks. We appear to be pursuing similar aims. Like you, I have been experimenting with various practical tests to assess aspects of running style; gains in fitness during training; and evidence of over-training.

      As you might have noted from various posts on my blog, I rely on heart rate as a measure the amount of energy expended. There are several problems with this. Heart rate drift, which is poorly understood, is the most serious confound, though is it is possible to correct for this by measuring beats/Km over 1 Km at similar pace and terrain at beginning (after warm up) and end of the session. Unless one follows a protocol in which effort is fairly constant over the session it is also important to allow for the fact that energy expenditure relative to heart rate is greater during periods when HR is increasing, and less when heart rate is decreasing. I have been experimenting with at a term that is proportion to rate of change of heart to the expression that relates heart rate to energy expenditure. I will do a post on this in the near future, along with a description of my recent attempts to develop a test o training effects (and over-training) based on a modification of the Lamberts and Lambert sub-maximal cycling test
      Thanks also for posting the link to your report showing beat per mile and cadence at various stage of your recent lactate threshold test session. I presume that your statement that this test indicates that 180 steps/min (ie cadence of 90 double-steps min) might not be the most efficient cadence refers to that you fewer beats per mile at 94 spm during the ramp up than at 93 spm during the test period. Alternatively, this might reflect HR drift during the session.

      • EternalFury Says:

        There is no doubt using HR as a indicator of: a) how much effort is expanded, b) how much energy is spent, c) how much fatigue has been accumulated…is not an accurate method. Yet, a strong correlation between these factors and HR exists and it’s pretty much the only practical way to measure these factors in the field. (while tests on a treadmill or indoors have some value, I find significant degrees of divergences between them and field tests)

        So, HR, cadence, pace/speed, HRV, RPE, TE, and, maybe, lactate levels are the only aspects that can be considered in the field short of carrying a lab on your back.

        In my experience, dealing with cardiac drift is not a big issue as long as you measure *submaximal efforts* that take place *within 1 hour of the start of your exercise bout*. (drift may still occur under these conditions, but it should not add much noise to the measurements)

        The previous link I posted may not have been the best example…here is another one to consider:

        Here you will notice that cardiac output during Cool-down turns out to be more efficient, even though my cadence dropped to 91.
        This was a recovery run on Sunday, all of it in Zone 1-2, under 1 hour.

        Finally, yes, I agree, if you want to measure effectively, you must fix as many parameters as possible.

        I also think that factors like “running economy” may be very difficult to measure in the field as punctual data points. You may be able to measure regressions or improvements in “running economy” over time by comparing the data from chronologic sessions, but that’s about it.

    • canute1 Says:

      I agree that when training in the low to mid-aerobic zone drift is not such a problem. In general the relationship between heart rate and effort expended is simpler in low to mid-aerobic zone. However, there is some evidence that heart rate at work loads in the upper aerobic zone provides a more reliable estimate of the effects of training on performance in events which are raced in upper aerobic zone (which includes 10K to marathon) so I think it is important to develop tests that examine heart rate responses in three zones (low, middle and upper aerobic) if one wants to assess training benefits and monitor for early signs of over-training.

      Thanks for posting the link to the report on your low aerobic training session. The fact that you find that the most efficient cadence during cool down is somewhat lower than what appeared to be your most efficient cadence during the LT session is in accord with the theoretical prediction (and much observational evidence) that the most efficient cadence should be lower at slower paces. However, it would be important to demonstrate repeatability of this observation across several sessions before drawing strong conclusions.

      • EternalFury Says:

        “there is some evidence that heart rate at work loads in the upper aerobic zone provides a more reliable estimate of the effects of training”

        It may be so. Although, too many runners expect their bodies to perform as machines. They aim to hold a constant cadence, a constant heart rate and a constant pace. Some studies have demonstrated that it is less efficient to run with such goals. In particular, letting your heart rate and pace fluctuate based on how you feel may allow a better overall performance.
        With this in mind, I would like to measure the effect of training using sub-LT efforts only.
        This is true particularly true for me since I am only interested in half marathon and marathon.

        Every run I do gets logged and I will soon have the ability to track long-term trends. So, I agree with you and I can’t wait to see what these observations look like over time.

    • canute1 Says:

      I agree that you should do a substantial portion of your training in the low aerobic zone. I believe it is worth testing how training improves your fitness at all paces from low to upper aerobic zones. I find that the effects of training are fairly uniform across the aerobic range. The interesting thing is that low aerobic training can lead to improved fitness at upper aerobic paces, as emphasized by Hadd and others. Nonetheless, evidence such as that provided by Lamberts and Lambert suggest that the HR in upper aerobic zone (when cycling) is the best predictor of 40 Km time trial performance. Therefore, like Lamberts and Lambert, I base my assessments on measurement at three different paces spanning the aerobic zones. In addition I think that heart rate drift is itself a quite important indicator of performance over HM and marathon, so I think that assessing HR drift is itself a useful indicator.

      In the near future, I plan to post a fairly detailed account of the assessment procedures I am currently developing. One of these is a formal test that I hope can be incorporated into the warm-up quite frequently (eg once per week) and the other is a measure that can be applied to any training session to provide an indicator of the amount of cardiac work done during that session. However I have a little more evaluation to do to convince myself that these procedures are reliable.

      Good luck with your training. I look forward to hearing the outcome of your assessments.

  10. Gazelles v gliders: Mirinda Carfrae v Chrissie Wellington | Canute's Efficient Running Site Says:

    […] higher for the Glider because repositioning cost increases with cadence (as discussed in my post in April 2012) and furthermore, the trajectory of the foot of a Gazelle results in a shorter lever arm of the […]

  11. The big debates of the past decade: 1) Running style | Canute's Efficient Running Site Says:

    […] a longer airborne phase, unless cadence is increased.  However, as described in my post of April 2012, increased cadence demands more energy expenditure to reposition the swinging leg, so we need to […]

  12. Steven Brewer Says:

    Hopefully this post is still being read.
    Very interesting topic, as I am a 51 year old male, attempting to become a competent (BQ) long distance runner, after really no running experience since I was 17 years old, and cannot assume that I was competent when I was 17!

    Q: With respect to optimum running styles and cadence. Faster cadence, also means less rest between strides. Is there any consideration given to “resting” your muscles between strides? The reason I ask, is that I have read that it is important for the heart to rest between beats, and since this is the “penultimate endurance muscle” of the body, I wonder if there is endurance value in our running muscles to get rest between contractions?

    For the first 18 months of training, I would consider myself a glider, but I have recently changed to gazelle.

    For the first 18 months as a glider, speed increases were pretty much a result of me increasing my average cadence, while improving my form. My last 5K, I ran in 23:20 with an SPM of 192. With training I believe that I could increase that another 5 percent over a 5K, but I am not convinced I could run an entire marathon with the increased cadence. Using this style, my calf muscles, required constant attention, and I regularly suffered from runners knee, ITBS and metatarsaglia. In fact I badly tore my calf muscle in Oct 2013. In hindsight, the tear was caused by lack of stretching the calf

    To challenge myself, I recently spent some time training to run a 6:00 mile, and found that it was difficult to achieve this goal as a glider. I believe that my core was much too weak to be able to achieve this goal.

    So I spent 2 months working on my core strength.

    While working on my core after a two months, I realized that I could now change my gait, and have switched to the gazelle running style. So far, I am able to run 1 mile using the new technique, and am continuing to improve my strength.

    What have I noticed from this change?
    Running is more fun.
    My hard earned upper body form seems to work “as-is” with the lower body changes
    My cadence has slowed, I haven’t measured it yet, but I expect it has fallen a little below 180.
    My runner’s knee issue has disappeared.
    I don’t have any issues with ITBS.
    Metatarsaglia still there, but not getting any worse, may be getting better. Time will tell.
    It is EASIER to run with LESS vertical oscillation, despite what I have read from some gait analysis.
    My calf muscles don’t seem to be as wiped out after a run, and require less attention.
    My glutes are screaming at me! This is good, as I am now putting them to work! Seeing as they are my biggest muscle group, I am more than happy to finally have them at the party!
    It feels easier to achieve the same speed with the gazelle gait.

    So far, my observations lead me to believe that I am on the right track, and will continue becoming more competent and stronger with the gazelle style. Key for me to getting better/faster/stronger is to continue to improve the height and co-ordination of the knee drive with the push off, as that feels like it will both increase my stride length and improve my efficiency.


    • canute1 Says:


      Thanks for your comment.

      The question of cadence remains a topic of debate. First, with regard to your specific question about the need for the muscles to relax for an adequate period between contractions, I suspect that the major requirement is ensuring adequate time to replenish intra-cellular ion concentrations in both nerves and muscle cells. Sodium, potassium and calcium ions all play a crucial role the conduction of signals in nerves and in the contraction of muscles. These ions are move in and out of cells in an alternating manner during the process of nerve signal transmission and muscle contraction. It is essential that the concentration of all three type of ion in the cells are replenished to initial levels after a burst of activity. Although I do not have all the relevant data to hand, I believe that the ability to restore intracellular potassium is likely to be a limiting factor during running. In the resting state, potassium has a high concentration within nerve and muscle cells, but the ratio of intra-cellular to extra-cellular potassium (i.e. the gradient across the cell membrane) falls appreciably during sustained exercise and this loss of gradient might therefore contribute to fatigue.

      However, I do not think that running with a lower cadence at a given pace will result in a lesser fall of the potassium ion gradient because the lower cadence will result in the need to exert appreciable greater force to overcome braking and /or get airborne, and the generation of greater force is likely to produce a greater efflux of potassium from cells. High intensity interval training (HIIT) results in quite marked depletion of intra-cellular potassium and therefore promotes the development of a strong mechanism for pumping potassium back into cells. Maybe this is one of the worthwhile advantages of HIIT.

      Since I wrote this blog post in 2012, the evidence in favour of relatively high cadence has grown stronger (i.e high relative to the self-selected cadence of many recreational runners). In particular, in a study of recreational runners, Heiderscheit reported that a 10% increase in step rate from a self-selected mean step rate of 172.6 ± 8.8 steps/min at a pace of 2.9 ± 0.5 m/s led to an almost 20% reduction in energy absorbed at hip, knee and ankle joints.

      There is an upper limit to the desirable cadence because limb repositioning costs increase with increasing cadence, but nonetheless, the optimum cadence increases with increasing speed. If cadence did not increase with speed, stride length would have to increase greatly resulting in prohibitive braking costs and/or the costs of getting airborne. This can be demonstrated employing the somewhat simplistic mathematical model that I presented in my posts earlier in 2012.

      So in summary, I doubt whether your moderately high cadence (around 190 spm) was in itself the primary cause of your problems. However if a ‘gliding’ style with low take-off angle resulted in excessive costs of braking, this might have contributed to both fatigue and injury. Because it is necessary to balance three costs: getting airborne, braking and limb repositioning, there is no simple answer, but as a rough rule of thumb, I think that at a pace of 4 min/Km (4.1 m/sec), cadence should be at least 180 spm and probably higher.

      • Steven Brewer Says:

        Thanks for the lightning turnaround time and thoughtful response.

        Interesting note on your opinion of intracellular potassium ion concentration. It would be interesting to know ion migration rate and equilibrium concentrations across the cell membrane at resting state. That said, I must remember to ensure that I provide enough raw material in my diet!

        Agreed, I don’t think that cadence was a primary cause of my lower leg issues, the reason that I raised the cadence variable, was that there is a limit and I felt I was entering a cadence range of diminishing returns (to your point: limb repositioning costs), and needed to look at other options for ways to accomplish the same running task.

        Interesting point about braking. I don’t think that my gait introduces excessive braking, but when you are fatigued this is more difficult to assess.

        Your observation about cadence evidence becoming stronger is interesting. I wonder how much of the self-selected cadence by recreational runners is generated by the demographics? How many of these people are like me and aging baby-boomers, that have few running skills? Might this be pushing the numbers up? On a personal note, when I recently started running I believe that I “chose” glider only because I did not know that gazelle was even a possibility!

        I believe the primary change in my gait from glider to gazelle, is one of load on the individual muscles in the kinetic chain. I am attempting to shift some work from the relatively small calf muscles to the hamstrings and much larger Gmax group. All other things being equal, if I can perform the same work with larger muscle(s) like Gmax, I may be able to delay the onset of fatigue and all of the issues that fatigue generates. It would be interesting to see the demand placed on each muscle in the kinetic chain for the two different running gaits’ vs the capacity for each muscle to supply adequate work before the onset of fatigue.

        I shall have to look at your mathematical model.


    • canute1 Says:


      You raise an interesting point about the possible over-representation of older runners in recent study samples. In the study by Heiderscheit, in which self-selected cadence was 172 steps/min, the mean age of the 45 runners was 32.7 ± 15.5 yrs; so neither young nor very old. In fact many older runners do run with a relatively fast cadence. I believe that it is better for older runners to run with a fast cadence because minimization of impact forces is crucial for the older runner.

      With regard to potassium ions, it is the gradient across the cell membrane that matters, so efficient ability to pump potassium ions back into cells is the major requirement. Mild reduction in total body potassium usually has only very minor adverse consequences though severe deficit results in muscle weakness.

      I certainly think it is advantageous to use the large muscles such a Glut max to push forward and upwards in late stance. I believe that one of the best cues to promote this is a sharp down and backwards sweep of the opposite arm. When I become tired, I focus on this arm movement.

      • Steven Brewer Says:


        Thanks for the tip on the arm movement cue to generate late stance GMax push. I will give this a try tomorrow on my Long Run.

        In my home town, where I am training, we don’t have any access to running coaches to improve technique, so I have been attempting to gain information from other sources, which is how I found your nice blog. Local running friends have been able to provide adequate advice on training programs/plans like Higdon’s and FIRST’s runLessRunFaster, but getting useful advice on how to improve running technique is very difficult. The more I learn, the more I am discovering that running is an extremely complicated process. (This makes me appreciate the simplicity of calculating the energy levels of an electron trapped in an infinitely deep well! Now, that was a nice simple model!) Trying to watch someone run, accurately, analyze their technique and then assist them to make positive changes in a safe manner is very, very difficult. Further, many people that offer advice, don’t know what they are talking about, or are afraid to offer advice in fear that their advice is inaccurate, or that it will be interpreted and/or applied incorrectly, making things.

        As a result my friends get frustrated, and end up just going out there to run, run run… When this works, great! Of course, it all falls down when an injury creeps up. I bet most can feel the discomfort before it stops their training, but feel powerless to change anything, because they simply don’t know what to change to make an improvement.

        With my recent change from glider to gazelle, I have been attempting to gather information to a) feel confident that I am making a positive change and b) gain a better understanding of my running technique, so I know what to improve next.

        Two things are puzzling me.
        1. I have read, (ie ttbikefit guy) for instance claims that gazelle motion increases vertical oscillation. So far, this has not been my experience during my transition. My assessment is that I have been able to reduce vertical oscillation. I am extremely tempted to purchase the new garmin 620, as it claims to have the ability to measure this. I haven’t yet, only because I believe that we have the ability to measure this ourselves while training on the treadmill, by focusing on a fixed point on the wall, and seeing the changes in vertical motion, when we change our technique.

        2. It’s not clear to me that I am experiencing more impact forces on my footfalls that might contribute to repetitive stress injuries. In fact my assessment is that I seem to have more time to reposition my lower limbs to get them ALL into the correct position for a controlled or “soft” landing.

        At some point, I need to feel confident that what technique I have is adequate to start training for my next race. My plan for the next 18 weeks is to put my current newly developed gait into my training program for a May half-marathon.

        Of course I can’t spend all of my time running, as recovery time is also critical, so I’ll be reading your “dance with the devil” and other posts on your site for a little light reading.

        There are no questions “per se” thus far in this communication, but I would be interested in your opinion on the importance of minimal vertical oscillation while running. If you find a technique that results in quiet footfalls, with minimal effort and oscillation, have you found a good running solution?


      • Laurent Therond Says:

        IMHO, running form is mostly innate, and what your body has come up with is likely to be what works best for you, short of spending years re-wiring your neuromuscular patterns to the point where whatever else you pick up becomes as efficient.

        Sure, in terms of physics, a certain form may be superior to another, but we are beings of blood and flesh, and what can sufficiently describe machines, may not be sufficient to describe us. (There is a lot of biochemistry in running.)

        When you observe a great runner with great form, the odds are that their form is the coincidental match of their innate form and what mechanics tells us should work best.

        Then, you can observe Haile Gebrselassie’s foot motion during stance and realize that innate form may not be a strong limiter of performance.

        In conclusion, I believe that if you run “a lot”, you will naturally gravitate towards what is the most efficient form for you.

      • Steven Brewer Says:


        Thanks for the response, and helping to give me comfort in the fact that there is no simple answer, other than keep running and the answers will come!

        I would agree that a runners form is found naturally from running a lot and your body discovers the best way. However, I might say it a bit differently. We use our innate talent to develop a form that works best for us.

        I do get your point that elite level success can be had despite less than optimal form, but I’m not exactly sure what your criticism of Haile’s form actually is. I viewed this link to get a sample of his running style.


      • canute1 Says:


        Many aspects of Haile’s form are excellent. The clip you posted shows only a side view. He gets well airborne with a strong drive off stance, and then drops his foot to the ground only a short distance in front of his body, thereby producing only modest braking. However if you find a clip showing him from behind, you will note that he lands with his foot splayed outwards and then his ankle collapses inwards. I do not have a good illustration to hand, but if you look at about 2:45 in the following clip of him running in the Hague in 2009 you will see the outwards splay of his foot on impact, and the very last right foot landing in the clip at 2:53 shows the subsequent inwards collapse, but it is not as clear on this film as in some others.

        Here is a clearer illustration. Look at his foot fall at around 2:18 in this clip, shortly after he runs though the Brandenburg Gate during his world record run in Berlin in 2008.

    • canute1 Says:

      I agree with Laurent that running is largely innate and therefore we should largely trust our body to run in the manner that comes naturally. However, many runners suffer injury, and when this happens repeatedly we need to ask why. While I greatly respect Haile Gebreselassie, it is noteworthy that he required Achilles surgery in 2000.

      In the past decade I think there has been an over-emphasis on faults of style as the major cause of injury. I think that subjecting the body to greater stress than it can cope with is the major causal factor in in many instances. Therefore one of the main strategies for avoiding injury is building up training load gradually, and at all times recognising when the body is near the limit of coping. This is not easy, though a variety of signs covering both overall well being (mood; heart rate variables) and focal signs such as accumulating aches in connective tissues or local muscle spasm provide useful clues.

      However, in some instances faulty technique plays a role. But there is a great deal of misleading information in circulation, due to over-simplistic analysis. The human body is very complex though it does obey several well established principles, based on both Newtonian mechanics, and the principles governing the behaviour of biological tissues.

      The first things to consider are the laws of Newtonian mechanics. We need to meet two goals: minimising energy expenditure and minimising stress on body tissues. Minimising energy expenditure requires the optimum adjustment of the three major energy costs: getting airborne; overcoming braking and the cyclical repositioning the limbs relative to the torso. Simplistic focus on just one of these leads to over-emphasis on a single requirement. For example, we can minimise the cost of getting airborne by reducing vertical oscillation, but that inevitably increases the cost of overcoming braking, at a given cadence. We can minimise both braking costs and cost of getting airborne by increasing cadence, but that increases limb positioning costs. Conversely, if cadence is too slow there is a strong temptation to increase stride length by reaching out with the swingling leg, thereby incurring braking costs that are inefficient and potentially damaging.

      The need to balance the three main costs leads to 2 important conclusions: first, simplistc rules such as avoid vertical oscillation without considering other costs are misleading and potentially dangerous. In general, it is best to let the non-conscious brain do the calculation required to optimise the three energy costs, because the brain is usually very good at finding the most efficient solution.

      However, as discussed above, there appears to be one exception: most recreational runners select a cadence which appears to be slower than optimal for fast running. Maybe this is because we are naturally adapted for slow running over rough terrain where maintaining balance is a high priority; alternatively it might be that a sedentary lifestyle encourages our brains to be too cautious. So if there is one simple issue we should address consciously, it is making sure that cadence is not too slow (eg at least 180 spm at 4 m/sec).

      The second general conclusion arising from the need to balance the three costs, is that large forces will necessarily be exerted. A quite strong push against the ground is inevitable. Running is a dance with the devil – gravity. We therefore must build up the strength to cope with this, by gradual increase in training volume. Specific exercises, both plyometrics and resistance exercises, can help.

      Associated with the large forces required to get airborne is the fact that the geometry of our hips knees and ankles results in quite large rotational effects around two, or sometimes all three axes, at these joints. Some of these effects are not easily envisaged, so there are certain injuries that do require a careful biomechanical analysis. Nonetheless I consider that gradual build-up of training load and trusting your non-conscious brain is the most effective way of minimising the risk of such problems. If if you do decide that you need to adjust style conscioulsy, it is safest to focus on a compact, brisk but relaxed arm action. This tends to promote a compact efficient leg action as a consequence of the way the brain codes movement.

      With regard to the physiological properties of body tissues there are several principles that are worth considering. The first is that running mobilises the catabolic hormones (adrenaline, cortisol) necessary to promote energy metabolism. Catabolic processes break down tissues that must subsequently be prepared by anabolic processes. The art of training is largely about balancing catabolism and anabolism – the most important requirement is ensure that recovery is adequate.

      The second valuable principle is that repair of damage tissues involves an inflammatory process that often involves laying down of collagen fibres in a randomly ordered manner. However, the relevant muscle or tendon functions best when the fibres are aligned in the direction of the usual forces, so once the initial inflammation has settled, gentle active recovery helps re-model the tissues in the optimum manner.

      Overall, the science of running is complex. Some problems are indeed more challenging than applying quantum mechanics to calculate the behaviour of a sub-atomic particle in a well, but a few simple principles provide most of the guidance a runner needs.

  13. Steven Brewer Says:


    I used your cue today on my long run for improving the force of my late stance push-off, with a stronger backwards arm swing. It did work for me. I must admit that I was surprised at how well it worked to improve push off. The cost of swinging the arms is fairly high, so I can see why it might be reserved for the later stages of a race where the legs are fatigued, but you may have spare arm and aerobic capacity. That said, there may be a happy medium that is worth me exploring in the future.

    I am in violent agreement with your comment on the over emphasis on style being criticized for the source of injuries. Blindly following over zealous training plans are likely the root of most of the injuries that runners suffer. We just need to gradually challenge ourselves, and then simply listen to our bodies, all while getting proper rest and nutrition to foster the adaptation process.

    I wanted to expand on the issue of innate ability, putting it into the context of my situation and an adhoc experiment I attempted lately.
    To add some context, I currently own PR’s of 5K=23:20,Half=1:58:30 using the glider technique. These #’s do not bode well for translating them into my 3:30 marathon goal.

    At the end of this year’s running season, in November and December, I decided to try something different and attempted to run a 6:00 mile. My “rationale” was as follows: Most race predictors, predict that 6:00 milers are able to run sub 3:30 marathons (If properly conditioned for the distance). My training plan called for running progressively faster 400m’s, starting at 7:00 pace and working my way down to the treadmill’s top speed of 5:15. What I discovered is that although I could glide at 5:15, the high SPM was stressful. I also started a program of increasing my core strength, as I felt this was limiting my performance. Yes, I know I am adding more variables to my experiment, but I was impatient. During this experiment, I also discovered the gazelle technique. It allowed me to keep my 5:15 pace, but at a more comfortable SPM.

    Lesson Learned? This helps me better understand why most running programs include “speed work”.

    This brings me back to innate running ability, and the potential to find what works best. I wonder how many other late-starting recreational runners who may have some innate talent but are handicapped by weak muscles, and little capacity to explore all potential running forms. And therefore they are gliders!!! It’s not clear to me that long distance running goals (like the marathon) are the best opportunities to explore your potential, however, today, they are the most popular.

    In summary, I wish I had tried this experiment soon after I started running, as it exposed my weaknesses, and opened up a running style possibility i had never previously considered.

    I suppose most successful marathoners likely were track and field (100m – 10,000m) stars in their youth, and were able to rapidly explore different running styles with strong, flexible bodies, and find the styles that worked best for them at the different goal distances. From there it is was likely much easier for them to extend their experience to the 26.2 miles.

    That all said, you and Laurent do give me comfort that I just need to keep running to keep improving and reach goals. Be sure to listen to my body to adapt at my rate, and I’ll figure it out!


    • canute1 Says:


      Thanks for the description of your own running. I agree that both speed work and strength are important for marathoners. As you might have noted from some of my more recent posts, I am an advocate of polarised training, in which a large volume of low intensity training is balanced with a small amount of high intensity training. Furthermore, in recent months I have been experimenting with Ed Whitlock’s approach to marathon training. As you might know he is famous for his multiple long slow runs each week, and phenomenal marathon performances. However it is also noteworthy that his first world masters title was actually a 1500m title and subsequently, in middle age, he did a lot of intense interval training before adopting his more recent program based on a large amount of low intensity running seasoned with a small amount of high intensity racing.

      Good luck in your quest for a sub 3:30 marathon.

      • Steven Says:

        Thanks Canute:

        Likewise, good luck with your training, and avoid those injuries! I do hope Ed is okay, as I see he did not run the Toronto marathon this year.


  14. More reminiscences and the science of running injury free | Canute's Efficient Running Site Says:

    […] very pleased two days ago when Steven Brewer posted a series of questions on my 2012 post ‘Is there a magic running cadence’. It was interesting to review the evidence that I had  presented in 2012, and to note that the […]

  15. Collagen Pure Says:

    Hi there everyone, it’s my first go to see at this website, and piece of writing is genuinely fruitful for me, keep up posting such posts.

Leave a Reply to Simon Cancel 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: