This should have been a blind test, without them looking at the screen, possibly without even the riders knowing that their pedalling efficiency in particular is being analysed.
This is only five years late, but I couldn't agree more. The direct live feedback would have so contaminated the "experiment" that the results would be essentially meaningless.
I think the pros just pedaled normally as they have nothing to prove. Si and the rest attempted to smooth out there pedal stokes to improve their scores. Should have hidden the screen and told everyone completing the test they were investigating something else.
Yes, exactly this, 100%. Should've been completely blind. No one should've had any idea what was being researched. Putting a monitor in front of everyone introduced bias to the experiment. :(
Yeah they are comparable but you may be measuring whether pros are more willing or able to adapt to the feedback on the monitor which has nothing to do with the question they were trying to answer.
Yannis Zographopoulos I’ve run Osymetric, Rotor Q Rings, and Standard, and find my pedaling efficiency increases on the non round Rings (inefficiency in this case would mean wattage applied in any direction other than the direction of pedaling.) measured with a Pioneer power meter. I don’t know if this is due to the irregular shape being better suited to muscle groups used to pedal or simply because the changing mechanical advantage through the pedal-stroke isn’t interpreted correctly by the system.
Thank you Ryon Beachner! I believe shedding light in this kind of staff is really interesting. We always try to improve our wattage and fitness level but if there is a shortcut to free speed, why not take advantage of it ;)
Yannis Zographopoulos It’s certainly interesting! I only stumbled upon the Osymetric rings that I now use all the time because I had problems with my knees, and I found that they helped alleviate this. I was trying to pedal in a circle, and actually found myself with an overuse injury from improperly using certain muscles. (Chondromalacia patella) Now I can’t pedal in a circle to save my life, but I’m much more comfortable and going faster little by little with the Osymetric.
Despite the repeated claim - and occasional ludicrous attempts to draw valid conclusions - GCN *never* does Science, and you've made very clear some of the significant shortcomings in all these sort of videos.
Being a former pro Si you should know why pros pedal they way they do. However you touched on it right at the end. I believe the US team 7-11 did a study regarding pedalling back in the 80s. In basic terms they found that pedalling for more of the up stroke that they gained more power however this did not improve their performance as it affected their efficiency. Pros pedal in the way they do because it is most efficient. Allowing the leg to relax and rest for a moment on the up stroke is a good thing. People teaching cyclists to pedal through the up stroke while riding normally on flat terrain (not under sprinting load) is quite an ignorant thing to do.
Very interesting!!! Please revisit this subject and do more tests. Regarding "pulling", there appears to be some bad information floating around by some well-meaning bike fitters that you should not pull up on the stroke. They usually cite the fact that our hip flexors are not powerful enough for us to do this and it will lead to hip problems. This is true if you try to do it too much. But there is nothing wrong with "pulling" for short durations at strategic points in a race or ride. The most obvious example would be during a sprint when "pulling" is essential for winning! In my own experience, I've found that I can increase my power output (measured with a Stages power meter) as much as 40% by pulling and there is very little perceived increase in effort. I do this during steep climbs while remaining on the saddle and it enables me to maintain a higher cadence and power level. But it's not something I would try to sustain for a long time. I also "pull" in intervals on flat roads when I'm getting tired and need help keeping my cadence and power up (as when doing an FTP test). Pulling temporarily allows some relief for my primary muscles since I can relax my "pushing". I don't think we can look at our pedal stroke as a static thing where there is only "one" best or optimal pattern for us. What I think we need is a variety of pedal techniques that we've mastered for different situations so we can use the right "tool" out of our pedaling toolbox for the "job" at hand. Lastly, I do not think that a fixed trainer is a good device to use for these tests. The problem is that it doesn't allow you to lean the bike from side to side as you pedal and this may artificially increase pedaling asymmetry. I'm aware of only one trainer that may overcome this problem: the Rock and Roll Smart Trainer from Kinetic ( www.amazon.com/dp/B01J0BQM2Y/_encoding=UTF8?coliid=I1J31Q7QA9H4BC&colid=1VQDTWUQD9MXU&psc=0 ) but I haven't had a chance to try one yet. I believe your data will correlate better to the "real world" if the cyclist is able to move the bike laterally as they pedal. Is there a power meter you can use on the bike that will provide as good of data as the Elite Drivo? (It would be great it if could distinguish upstroke power from downstroke power.) If so, please do your next tests either on the road or on a Kinetic Rock and Road smart trainer. Thanks.
As with almost everything in life what is optimal depends on a case by case basis among individual types as well as road circumstances and demands. Everyone always wants a universal and simple response when variables are the reality.
was this test blind? Did all your volunteers know you were testing for smoothness? if they did, they might be changing their pedalstroke, especially if they can see how smooth they're being on a live feed in front of them.
that higher wattage for them is not more effort - but that should be the reason I think. If you train and get fitter, those strong muscles get much stronger, but those muscles unable to do a significant percentage of the overall power will not benefit as much from training and will not get nearly as much stronger
This is a good experiment. Can you guys redo this with a modern oval (absolute black, wolf tooth, etc...) chainring to prove what they're claiming? #torqueback
Oh simon, so desperately clinging on to "you get power pulling up" Tbf the results are pretty logical, irregular downstrokes being the result, with increase of power only on the downstroke, because that is the most (only) efficient way to pedal. Power on the pedals = accelerating momentum.
So Elite's mind boggling array of data doesn't include the upstroke power, and therefore the pedalling uniformity calculation presented is useless, got it.
absolutely. crazy. BTW there are machines that sense force vector through 360 degrees, and a study of elite track cyclists (4000 meter specialists) showed they don't scrape nor pull up while in steady state.
i never knew what pulling up would do except make the whole movement feel weird. i prefer to power through the strongest spot to keep better momentum in the dead spot in combination of a 95rpm cadence. works hella good with TT. 85-90 rpm for MTB and normal spirited riding
Yeah. Of course if you are seeing the results real time you're going to try to be as uniform as possible. This is like testing "can pros hold 300W on instinct???" and then showing them their power output during the test..
As Si says, more science needed. I appreciate Si's willingness to try flat pedals. To do flat pedals correctly, need good mtn bike flat pedals and good stiff shoes. Most important is to move the pedal to mid-sole to take the calf muscle out of the stroke. (Ankles and small calf muscles prevent all power from large glut muscles from reaching the pedals. For example, would you prefer to be on your toes or flat footed when you attempt a vertical high jump?)
I wonder if you had everyone ride at relative ftp percentages if you would see a different trend then with the absolute wattages. I would guess the more fit people can just hammer it while the less fit have to give their best, smoothest performance.
I know this is an advertisement for the trainer intended to give useful information, but it just doesn't. These tests should have been done outdoors over a period of several rides with several pro and amateur riders. And definitely shouldn't have been done staring at the graph.
Maybe we are lucky and sponsors just sponsor, but don't influence the "science" ? That's how it should work in real life scientific studies - but doesn't always happen.
For sure lately my most favorite sessions on the trainer are those with variable cadence. Gradually or abruptly changing, I don’t care, just bring’em on!
I change my stroke often during a ride to rest muscle groups. For instance I'll alternate between 3 distinct styles: stomping, scraping, and kicking or use combinations of the three. That said, fit and comfort on the bike is more important to me than performance.
Track cyclists somehow believe they are superior because they are fixed. I was berated by an experienced track racer while learning track coming from road. He challenged me to two sprint laps, both of us on 88 inch gear which forced us to max out cadence. I won both races against him back to back. He’s been quiet ever since and has now shifted his criticism on me to ‘track craft’. My point is that most seasoned cyclists will have a pedal stroke that’s right for them. Being trained is much more important than pedalling smooth and trying to force a pedalling style that you’ve not grown into. Teaching a child to pedal smoothly is perhaps a good idea but not 30-odd year old blokes who’ve rode and raced bikes most of their lives.
The body knows. What's right for one isn't right for another but our bodies find an efficient way for us over time, don't overthink it...maybe...always worth a bit of experimentation to see if the unthinking instinct might've missed a trick.
Why? we already know the answer. They are oval...how could they ever ever ever produce a round pedal stroke? Oval rings are designed to produce a power down easy up pedal stroke.
totally agree. I personally switched to oval(not osymmetric) about half a year ago on my roadbike and 1 yr on my XC hardtail. whnever i switch to round rings it feels less pronounced and less smooth(at least a bit). My powermeter says that too that my pedal smoothness is slightly better with ovals(same cadence, wattage)
oval rings are the opposite of what they should be; they bump you on the top of your stroke instead of eliminating the dead part of the bottom of your stroke. FROOME rides this oval ring; obviously explaining his many losses :)!
t42 Press lets say, they feel smoother. Locigally they arent of course as ya mentioned because of the bump in power in the downstroke. But in the 3 o'clock position is the hardest part. The so called dead spot is slightly easier at the 6 o'clock position. If you were talking about shimanos biopace chainrings, forget them. They are making things 100% worse. Whereas oval ones don't make it any worse and potentially easier. I personally believe in osymmetric chainrings. They seem to work for me pretty well
Well, since smoothness is apparently a pretty irrelevant measure, it would be waaaay more interesting to measure efficiency* with round vs oval rings ;-)
Pushing through the dead spots may not add to efficiency, but it will allow you to use a higher cadence easier and more effectively. It will also allow you to put out more power on the up-stroke.
You should look up Control Entropy. It theorizes that elite runners have a les even stride because of their ability to adapt to differences in stride because of their huge amount of experience/amount of miles. In less experienced runners they see a much more even stride because they have less adaptability. When a person hits the wall, lets say after running a marathon and their body shuts down, they saw that the stride, even in elite runners, becomes very uniform, because the body has to preserve all energy it can, and thus doesn't have the energy to provide a big amount of adaptation. So elite cyclists could have a less uniform pedalstroke, just because they don't actually need one. Should be interesting to do some EMG-testing after complete exhaustion.
Now what about track riders? they don't have freewheels so there is no slop in the pedal stroke as long as they are moving at a constant speed, so would that be a better test? Also, what if the un-smooth pedal stroke actually helped carry momentum through the bike, almost like pumping your arms while running? Ive witnessed this in such athletes like a large track sprinter (Fosterman, Ceci, Boss, Staff, Northstine), what if the inertia of their leg, hips, and mid-line actually helped propel the bike forward thus not adding more "power" to the pedals but effectively adding more energy into the bike to accelerate it more quickly. Also would it matter if you were peddling smooth in a 125" gear (53,11) as long as you are adding force through the stroke you would be maintaining or accelerating?
What about one footed training where you remove one foot from the cleats, and trying to keep the freehub from clanking while spinning at a relatively high cadence? I've heard this cam be used to good effect not in my in increasing smoothness, but also increasing your ability to pull up on the pedals with more power.
It's good you did this test, and from your results it proves really nothing is important about the smoothness of your technique. Whatever gets you down the road works!
The entire theory behind oval rings suggest that smoothness does not matter when it comes to producing power. Rather, it’s more important to quickly move through the dead spot getting back to the downstroke. Also, diminishing returns suggest it would cost too much energy to pedal smoothly for the relatively low amounts of watts produced by the hamstrings through the bottom and over the top of the pedal stroke. No one is putting out 500 watts on the upstroke, but you can hurt your knees/hips/hamstrings if you try to. #PushdownRestup. Great video 👌
And of course by this metric 'uniformity' the value goes down with higher power outputs because the deadspot power remains 0 lets say, but the peak power increases spreading the gap between Av and peak. This should not be a surprise but highlights why this isn't a reliable metric.
As said below, this is only the Positive side of the stroke 0-180 degrees. There are negative and positive forces at work from 181-360 degrees as well. So this really is meaningless. I did a webinar on pedaling analytics on the TrainingPeaks>WKO4 channel. Explains it more completely.
I wonder if the variety in pedal efficiency could be due to training with power. If you train with power you might change the way you pedal in order to show the same power numbers but with a decreased effort, for example finding the place in the pedal stroke where your power meter best picks up power and using micro-efforts in each pedal stroke to display and record these higher number, which over time could cause an increased inequality in muscle strength etc. , hence explaining why pros have an increased inefficiency pros as they regularly train to power. This might not necessarily be done consciously, but may be subconsciously as a way of ‘increasing power readings’ for the same perceived effort levels when mid effort and when looking at power numbers. Just a thought.
You forgot to ask one very important question and that is what muscle groups are they using to produce the power during the pedal stroke. I have a theory pros have a gastroc dominant pedal stroke while amateurs have a more quadriceps dominant pedal stroke.
Thanks for more "science". I have seen your comparisons of cadence efficiencies in the saddle. I notice a more narrow range of cadence rates in prolonged (not sprint short duration) standing pedalling cadence, I'd like to see you compare standing cadence efficiencies for 2-3 min intervals; if you can do it! Does that sound like a challenge? I think so! Cheers
I'm an engineer frame builder bike fitter, who has always been fascinated by the mechanics of cycling. I'm a firm believer in smooth pedaling from an energy conservation and injury prevention perspective. After running a watt-bike spin gym for a couple of years a number of observations became apparent these are by no means scientific. Pedaling smoothly can be learnt but has to be repeated enough with feedback so as to become a muscle memory effect to be useful. The fact that pro cyclists with big engines (ie capacity to carry large volumes of oxygenated blood to muscles for energy conversion) don't seem to need to pedal smoothly would point to the fact that muscles required for propulsion are not the limiting factor to speed. What we should be asking are athletes with the same size engine Vo2Max, Lactate Tolerance, etc. better off pedaling smoothly or not. Does pedaling smoother allow better energy conservation for the sprint at the end of the race? For myself as my pedalling became smoother my ability to time trial improved dramatically. It is my belief that pedalling smoothly is more efficient and is one way for the less genetically gifted among us to compete at a higher level, and may be clue to the GCN. Results!
I would include the smooth pedaling technique in the same class with Bermuda triangle, Big Foot and UFOs. In my opinion the best pedaling technique is the one that fits best your body geometry and body parameters. In my case I discovered that the left foot is more powerful and the right foot is more technical and this is why at higher cadence the right foot does 54-56% of the work while at low cadence the left foot takes over and does most of the work. And if I am to go even deeper in analysis I discovered that my right foot start pushing closer from 12 o'clock pedal arm position while my left foot pushes starting at a more forward point. This is why when higher power is required my right foot struggle while my left foot seem to be better. I would say my smoother leg is not good at outputting high power Another curious finding is the fact that I hold better cadences of 87 and 92 rpm but I have trouble holding 90 rpm. This might be resulting from the fact that only certain cadences are optimal for my body. Should I try and get better at holding the 90 rpm or instead I should be pro enough and use what works best for my body?
I too would like to see this. The conclusion I came to is that pedal analysis can only be reliably done with a power meter at the pedal, not in the rear wheel.
Some scientific studies have been made, at least on Osymetric, and i don't know about smoothness, but on power they can't say yes or no because it never reaches the point where it's considered statistically significant. So in short we still don't know. The problem with making a GCN test on oval rings is you have either to use people who are already using them, or doing the test for a long time because there is a time of adaptation, the pedale stroke is quite different, and some love it, some hate it.
I don't think smoothness has any effect on performance at all. In the end, the limiting factor is how much oxygen you can pump through your body. If that volume is the same, then the average power between a 100% smooth pedal stroke and a 50% smooth pedal stroke is identical. Rarely have I raced and thought "I need more power", the limiting factor is always fitness. (unless in a short sprint of course)
ever wonder why VO2 max is often higher in cycling than running? I think it may be related. When powerhouse work horse cycling muscles are working ( versus more auxilary running muscles) more oxygen is used and more work is done. what muscles have the power? the downstroke ones. which just make you feel like you are working hard? the auxillary ones that would make for a 'smooth pedal stroke'. I am equating running to smooth pedaling and non smooth pedalling more to typical cycling and the muscle efficiency of the big push downs. I notice when I try and go smooth, I tend to get a bit more winded given the same power output.
there was a training tool that was introduced maybe 20 years ago or so that allowed each arm to rotate independently, I forget what they were called now. But you could actually remain clipped in and do single leg drills at the same time. I think they were supposed to help smooth out your pedal stroke. I'm guessing they didn't catch on...
I started with clipless pedals thinking you should pull up as well as push down. Next I hear that creates negative torque and you should just relax on the up stroke. Now the test contradicts that and the pros technique isn't smooth after all and they may even pull up. The only thing I can be sure of is that clipless pedals require special shoes.
Be interesting to compare this to a track rider. Logically, Riding fixed would lead to greater smoothness as the wheels are helping turn the cranks over through the dead spot
I desperately want to see real science on GCN given the presumed income generated by the channel (or maybe I'm far off course). I really do appreciate the cycling content and the amount of work it takes to produce, so thank you.
Thanks, Si, such a fascinating topic, how we apply power to the pedals. And I recall Eddie B advocating, in a seminar in '83, that racers should rest a leg after every so many pedal strokes and then do the same with other leg. Do pros do this sort of thing these days? And for me, a sport commuter on a mountain bike, my pedal strokes are all over the road and happily so. I run a triple and I have options and I use 'em--spinning and standing and mashing and coasting and smoking.
My understanding of smoothness is different than this... Even power around the pedal stroke? Put everybody on rollers and calculate the range of motion in their tires and percentage of time within the smallest range. Losing power to movement elsewhere on the bike has to be part of the equation when it comes to the perceived smoothness we experience riding behind pros and amateurs respectively.
Is your metric for smoothness really the best? Should we be considering others? Possibly direction of force (pioneer does this). Do pro cyclists (or better cyclists) improve the downstroke more than the others (kick, drag, pull) hence skewing your metric for smoothness? I absolutely love this topic but know very little about it.
Rocking side to side? Its called peddeling in squares, plus raise your seatpost-greg lemond ref. Power can be smooth-except a, guess?- sprint! My thought.
Could it be better to have a less uniform stroke, as the muscles aren't under such near constant load? If you put proportionally more power into the down stroke, the muscles get a recovery on the rest of the rotation. Perhaps that could reduce fatigue and stave off lactic acid build up?
I agree with all those who said the test should have been blinded to the riders. Knowing the anticipated outcome creates the opportunity for bias and self adjusted performance. The theory that the upstroke improves power generation seems like a reasonable way to explain the variation in energy generated throughout the 360 deg rotation of the stroke as measured by the power meter. As riders we can feel that effort as our muscles contract. So to assess that you should add EMG monitoring of the hamstring muscles to the test to see the difference in how pro riders recruit their muscles vs the rest of us. On top of that, put all the riders on an isokinetic testing device to measure the torque they generate throughout their range of motion. The pros would likely put us all to shame.
Results don't surprise me - it is clear that maximum force is exerted where the outcome is more energy efficient - energy scraping through the bottom will just not yield the impact if the same energy is applied throughout the downward stroke just the way the muscles work.
So working on varied cadence is benefical? Sounds like fixie on a hilly/windy ride. I just wonder if there's much difference when you spin 160rpm because you want (with a freewheel) or because you have to (the fixie crank is turning your legs and you eighter go with it or you get catapulted out of the bike). For me, up to some 140rpm there doesn't seem to be any difference fixie or geared.
Oh, FFS. You can't have the riders aware of the nature of the test or have them staring at that polar graph during the session! If that is actually what happened, it surely biased and mostly invalidated the test. Also, why wouldn't use the Garmin Vector pedals for this type of analysis?
Would this have been different if you used a crankset based power meter like Rotor’s 2Inpower? Might Rotor’s analysis have told you what was going on in the up stroke?
I don't feel that having a perfect pedaling stroke is all that important however, I feel that it is necessary to have a smooth enough stoke that when you are really putting the power down you don't start bouncing out of the seat at high cadence, as well as not getting your upper body bobbing up and down as you grind only pushing down on the pedals. I would say that the down stroke of the pedaling circle is where you can get the most power but that should not be the only place you put power through. When you only push down and you start bobbing your upper body it is just wasting unnecessary energy. In addition, being able to do a perfect pedaling stroke can come in handy. For example, when doing an FTP test obviously your quads will start to burn at some point. being able to negate the down stroke of your pedaling circle can be useful, using only you hip flexers, hamstrings, and calfs allows your strongest muscles, your quads, to rest. Doing that for about 30sec to a minute every time your quads start burning and then going back to your regular pedal stroke seriously benefited me during that test. just what i have learned in my cycling experience. loved the video!!
Psuedoscience. Pulling up must be matched by an equal and opposite force. Therefore more downforce on the other pedal. Therefore torque increased by 2 X upfoce X crank length. And consequent increase in power. I suspect that mountain bikers and cyclo-cross have smoother power distribution relative to roadies to avoid loss of traction at higher power. Similar may apply in extremely steep climbs, where higher power in the top of revolution reduces speed drop off and risk of stalling.
Derek Job , Fair point. But with dual pedal use, the vast majority of counter forces will come through the legs. The seat and bars are used primarily to counter the torque.
Even though they didn’t have a smoother pedal stroke why is it that pros are greater cyclists. Could you not do the test on the road rather than a turbo trainer, I think that they may have a more smoother pedal stoke on the road so that the bike can lean left and right especially when smashing out 450watts. As always love the video.
It would be interesting to see if there is a diffrences between endursnce track cyclist v road cyclist v sprint track and weather someone that uses fixed gears pedles smother then those who dont
wouldn't 2 x pedal power meter measure power up and power down and the bottom heel scrape force? with out all these metrics i do not see much point in this.
I'm still a believer that the best cadence is your natural cadence, will always be the most efficient for you, (you can work on it and change it with practice - but overall it's your natural cadence that is most efficient for your body)
However unscientific the test is deemed to be, the results seem to prove my thoughts on the topic. Average power achieved in a pedal stroke is what matters. How that power is distributed will have no effect on performance. What isn't measured here is effort. Smoother pedalling could require less effort over a longer period. But that sort of test would have to be conducted blindly over a large cohort.
So here is my take on it that I don't think was covered. Humans are adaptive and the pros are really good at their particular pedal style. When you have a pro up there you are looking at Thousands of hours of practice and adaptation of that particular pedal style. I think you should take some non cyclist split them into 2 groups and train one group to pedal in circles and the other to mash. But I don't think that is entirely practical.
Not that surprising. It is so tiring controlling the feet the whole way round. Maybe compare riding position comparisons like road bike on the hood, drops and aero bars and time trial bike position, see how the power delivery is altered.
I don't have a power meter (yet), but when I make a conscious effort to pedal smoothly during hard efforts, my speed goes way up. I think it has something to do with efficiency. Any movement not helping the bike move forward is wasted calories. Good video guys, you should do some live Zwift sessions that us GCN fans can ride along with. Its currently 8 degrees here in Colorado and I cant be bothered to freeze my bits off outside.
I'm not sure the experiment is valid if you had them constantly looking at that graph. You should have just given them their power meter and tell them to push X number of watts. I personally finds that if I be mindful not to slam my foot down on the down stoke, but try to keep the pressure under and on top of my feet as equal as possible (i.e. pull up as much as push down), it helps to keep my cadence up and my hip steady, which in turn helps me push that little bit harder.
It brings to light a few things about current power meters - of which I don't own one yet - where I have to question just what they're measuring. If pedal power-meters are only able to measure down-force, then that seems caveman-like. If, however, they can only measure force, but are not able to separate the forces from the position they are in, that seems less archaic, but solve-able. One way or another, it would seem that we still have a good ways to go before power-meters are telling us everything that they could, and should, be about our pedal strokes.
current pedal power meters measure forces tangential to the circle. and perhaps even non productive, non tangential forces as well. Not sure why the study's pedals could not do upstroke. well, they could measure upstroke but could not pull that data out from the down on the other foot. sort of silly. something that irritates me about the wattbike actually
So why didn't you try pedaling with just the left leg then just the right leg? It would give you the smoothness fairly accurately but as you said, does this really matter?
I think that having the GCN staff ride at lower wattages than the pros was a mistake. I think what happened was that the pros AND staff produce exactly the same amount of power in the dead spots - very little. However, the pros produce way more power on the power stroke (power produced on the upstroke of the right leg can be interpreted as power produced on the downstroke of the left and vice versa, there is fundamentally no difference). Thus, to produce the average power of 450 while taking into account the "weight" of the (say) 20 watt dead spots, the pros would have to spike comparatively higher on the power strokes than the staff would to reach (say) 350 watts. I think changing the wattage target was a huuuge mistake on your part, Si. I'd be interested in seeing a round 2, where you keep the wattage targets more attainable.
Watched this video then and now, I am as lost then as I am now, should I be more relaxed / loose with my legs, should I practice to pedal at higher cadence on a easy gear without bouncing.
