Saddle Height – perhaps not the cause of back injuries!

There appears to be an awful lot of nonsense written regarding how to determine optimum saddle height. The advice of Canyon – the manufacturers of my bike state that the heel should be able to sit comfortably on the pedal at bottom dead centre (bdc) of the stroke. Of course no reference is made to shoe heel thickness or anything else. This allows both the feet to easily touch the ground when sitting in the saddle – but it doesn’t allow you to pedal the bike with anything even remotely resembling efficiency. 
For a while I believed in the magical formula of 109% of in-seam, measured from the pedal spindle centre to the top of the saddle. Like everything else out there this also appears to be founded more on fetish than fact. Perhaps most people do find this height ideal, but then most people unless properly trained, would drown if they fell into a deep pool of water.
Two things happen to me when I raise the saddle significantly – that is – over an inch above this 109% mark.  A. I go a lot faster – especially when climbing and B. My lower back disintegrates.
Sports Science Bulls**t
Lets assume for the moment that nobody really knows anything about this sport. The more I look into sports science and bio-mechanics it would appear to confirm that this really is the case. The scientists seem to get bogged down in scientific method and procedure and don’t question whether or not their tests are even relevant. Nobody for example is capable of comparing barefoot running and Nike style mega-cushioned running in a relevant manner through bio-mechanical measurement. Many scientists are truly “educated idiots” and they do not deserve the high level of respect that they are systematically accorded – or the funds that they often gain through crass misrepresentation (ie Anthropogenic Global Warming nonsense). Science is not there to be believed but to be ripped apart – that is how real science works.
Real Sports Science Rule 1 
When something is correct then it naturally crosses over into other activities
From my own athletic and coaching experience I have observed that when something is correct then it naturally crosses over into other activities – opening new doors to perception. When something is seriously wrong then it doesn’t.  Let’s consider this a basic scientific criterion to go by. A good example of this is how a bicycle is controlled. To turn you simply allow the centre of mass to fall very slightly to one side – the bike changes shape and cuts underneath your new trajectory to bring you back up. This interaction makes a turn. Skiing works exactly the same way – but most people don’t realise that. Sports coaches for generations have told skiers to avoid doing just that – so much for sports science! I even met one idiotic engineer who after announcing that he was “…the most famous engineer in the world”, then proceeded to state that everything is in equilibrium – the entire universe! So much for Newton’s second law – the mechanics of disequilibrium! Newton was right however and despite being difficult to perceive the principles of disequilibrium are correct and common to many physical activities.
Running to Cycling
Where is this leading?  Well lets see if any common points can be connected. In natural running the leg extends behind the body. The foot is lifted off the ground with the ankle extended.  The hamstring flexes the knee bringing the heel up and the iliopsoas and other core muscles pull the leg forward back underneath the body. The leg remains slightly flexed as the foot is posed directly downwards to the ground with the ankle still extended. Shock absorption is due to the foot progressively lowering to the ground from front to back and the calf muscle regulating this with all of the foot muscles and tendons being active. 
A recent photo of myself on the Col du Galibier showed my ankle so flexed that my heel was below the pedal  at the bdc of the stroke. Proprioception (more like propriDeception) did not make me aware of this – only photography did. 
There is a lot of debate in general about whether “heels down” or “toes down” is more efficient. For me the answer is crystal clear – “toes down” is much faster.  At this very moment I’m watching a breakaway group in the Tour de France and they are all pedalling with their toes constantly down. Look at the excellent video in the following link: they all have toes down. The author here “Todd” states that he has analysed thousands of successful elite cyclists and they all pedal “toes down”. 
When you look at the video think about the description I gave in the previous paragraph of “natural” running. Not much difference is there? There is no shock absorption required so the foot doesn’t need to flatten as in running – if it did it would be absorbing energy. Basically this implies that the toes should naturally be pointing downwards to some degree all the time – a bit like a sprinter (runner) who gets maximum impulse from his tendons and muscles. Basically here the pedal stroke simulates the running stride as closely as possible. 
The Power Sponge
The much higher saddle height permits this permanent pointing downwards of the toes. My personal tendency is to forget and flex the ankle – but I’m also aware that this is because it feels lazy – a sort of “giving in”. In reality it’s more like sinking the heel into deep sand – a form of power sponge – that slows you down and tires you out progressively. All of the muscles of the feet just become useless jelly.

Skiing to Cycling

Perhaps the greatest fallacy in downhill skiing is the “bend the ankles” doctrine. Beginners are put into boots that permit the ankle to bend and the ski to be twisted into the direction of a turn. Racers are not expected do do any of this and so are put into rigid boots. The beginners are being taught complete nonsense which will later have to be unlearned and forgotten unless they want to be stuck on a learning plateau that will never go anywhere. The have been given a real “power sponge”  in an overwhelming sense. On skis there is a simple way to overcome this issue and that’s to get the skier to place the weight directly on the heels – not leaning against the back of the boot – but exactly the same as if standing waiting at a bus stop on in a queue.  Skiers with more awareness can then learn to raise themselves up slightly on the ball of the foot – engaging all the foot muscles and making the arches strong. The feet each have 26 bones (one quarter of the bones of the entire body are in the feet), 33 joints and over 100 muscles, tendons and ligaments. Most people just think of the feet as being uninteresting lumps of meat that only require attention when causing pain. Either supporting the body on the ball of the foot – through the active use of the foot muscles – or standing on the heel will cause the ankle to stiffen and strengthen – working in an “extension” rather than “flexion” manner. The strong ankle then stabilises the skier and allows control and feedback though the boots to the skis. Likewise in cycling the “toes down” position appears to activate the muscles in the feet – especially if you try to keep the feet more or less at the same angle to the pedal. You can feel the strength in the feet being used actively. The ankle in this case still bends – but as in skiing, though a more limited range and with extension. Much of the visible apparent change of angle of the foot to the ground is actually coming from the knee and hip bending – just like in skiing nearly all movement is in the knees and hips. The ankle seems to extend during the push – exactly as it does in skiing as the skier extends through the end of a turn (traditional instruction incorrectly teaches the skier to flex everything at this point). Clearly the listing of comparisons could go on here for a long time – but what counts is understanding the physical feelings involved and the common “door to perception” that we have found.

Since writing this article I have used the high saddle height in a hard race – the Etape du Tour de France – and no back problems were created. Perhaps the back problems were never created by the high saddle but by the flexing ankle instead – also the cause of most skiing injuries.


  1. Heel down is faster – on a recumbent!! Seriously, I've always ridden toe down on my DF bikes, to such an extent that I can't reach the floor from the saddle and have to move off it when I need to stop. I was shown this back in about 1978 when doing a time trial. If I try lowering the saddle (e.g. for my mtb in rough stuff) then my knees complain.

    Interestingly, on the recumbent I have the saddle to BB distance the equivalent of 50mm longer than on a DF, and it is even easier on the knees, and quicker. Not practical on a DF where you have to bend forward too much so run out of muscle/tendon length at the back of the leg/bum (we're optimised for standing upright – maybe another characteristic for you to train out?).

    Keep up the great blog

  2. I was teasing you – the recumbents are toe down relative to the rider (toe extended away from the head more than on a DF), but heel down relative to the road!!

    The interesting thing about recumbents is that the riding position is decoupled from the resting position, so can be optimised for pedal reach, back angle, etc., without worrying about the UCI regulations (your saddle IS flat, I hope, even though this is uncomfortable, damages your tackle, and no pro wants their saddle flat!!).

    MacCready with the Gossamer HPA's did lots of research on pedalling positions and ended up with a very straight body position (shoulder, hips, and BB all in a line). This is contrary to the racing fraternity who seem to favour 120 degrees between body and legs, but MacCready has the numbers to back up his position. Even my lowracer isn't that flat (I need to see where I'm going), but is more like 140 degrees.

  3. I'm glad! I did realise that of course the heels are lower (relative to the ground) because you are lying on your back – but I thought that your comment was serious. It's interesting to see another convergence on the same result though. Regarding the attitude of the upper-body I'm not sure. Likewise in skiing some advocate a very upright stance in racing and some quite a flexed one. In speed skating the flexed position is all you see – with the body tilted forwards (in slalom it is more or less upright but still flexed at the hips). I'd suggest that the increased relative angles of upper/lower body might be linked to the restriction of the feet from passing behind the body. Basically, if the feet can get behind the body then the body can be straighter. Straighter should be more power efficient given the option. You can definitely be more powerful on the pedals when you stand up straight on them out of the saddle – on a normal bike.

  4. Interesting analogy, but this may be down to the physics of how we can deliver power most effectively – basically downwards relative to the trunk. Thus on a bike we can be upright (or lying flat – effectively the same thing) and the gear train converts the downwards force into forward motion, but for a skater or skier they have to generate a backwards component of force to move forward, hence the trunk must tilt forward.

    Holding the body forwards requires something under it to support it, unless it is accelerating, which is why in skating or skiing all the big power is laid down in the turns, when the body is always accelerating even at constant speed, and thus both legs can be used purely for power, not for support.

    Skaters (and skiers on the flat) do this in a straight line too, by pushing outwards either side, effectively small turns in counter directions. Distance travelled is greater, but increased propulsive efficiency more than makes up for this.

    Upright cycling, skiing, and skating are also playing with the aerodynamic compromise, hence an exaggerated forward lean can be a net benefit even with a loss in propulsive efficiency, due to improved aerodynamics, but if we use a system that decouples these two effects, then the improved power position becomes evident. If you run the numbers on the top speed runs of DF's and recumbents on the flat, the difference is not fully explained just by the aerodynamic advantage minus the increased mechanical losses, the "inline" riders are able to put more power down for longer than the same rider can when bent forward (which is what MacCready concluded).

    Reading back through this post, and knowing your experience, makes me think of Grannies sucking eggs, but I'll post it anyway in the spirit of healthy discussion!

  5. I'll get back to this properly later – preparing for étape 2 with torrential rain promised. Speed skaters go fastest backwards- crosslegged – turning.

Leave a Reply

Your email address will not be published. Required fields are marked *