The cadence myth

Discussion in 'Cycling Archive' started by Nick Maclaren, Jul 29, 2015.

  1. FINALLY, I have bumped across a paper which has produced measurements
    debunking the myth that higher cadence is always more efficient. Yes,
    it is - at above 250 watts! The vast majority of ordinary cyclists
    will be in the 50-150 range (as in walking), where the optimal cadence
    for efficiency seems to be at most 70 (as in walking):

    Nick Maclaren.
    Nick Maclaren, Jul 29, 2015
    1. Advertisements

  2. Nick Maclaren

    Mike Causer Guest

    On Wed, 29 Jul 2015 12:00:31 +0100 (BST)
    8 subjects of which only two were cyclists, "Subjects had varied
    athletic backgrounds including: a track cyclist, a power lifter, and
    several who engaged in regular recreational activities, in- cluding
    swimming, cross-country skiing, running, and mountain biking." and they
    were measured after a 5 minute warm-up. I need about 20 minutes to get
    warmed up properly.

    Sorry, but I am not impressed.

    Mike Causer, Jul 29, 2015
    1. Advertisements

  3. Can't say I'm totally surprised; presumably the power consumed in waving
    legs about is about proportional to cadence but the proportion of power-in
    that represents will vary.

    Maybe next we can have a go at "more duty cycles at lower impact is always
    better for knees".
    David Damerell, Jul 30, 2015
  4. My guess is that the best cadence for knees is not far off the one
    for efficiency, in general. People will vary even more, of course.
    It's certainly a lot easier to ankle aggressively at lower cadence,
    and that means that one's knees are bent much less, which is the
    standard physiotherapy recommendation.

    Nick Maclaren.
    Nick Maclaren, Jul 30, 2015
  5. You can understand why I am so unimpressed by the claims that what is
    best for cyclists who have spent 10,000 hours adapting to a UCI
    position and a high cadence and can deliver 300-400 watts is also best
    for the majority of the population?

    And the majority of active people don't spend almost all of their time
    in one activity, or would benefit from 20 minutes warming up. That
    investigation is far more plausible for most people than the ones you
    imply that you favour.

    Nick Maclaren.
    Nick Maclaren, Jul 31, 2015
  6. Nick Maclaren

    Ian Jackson Guest

    I guess it depends what you think the target audience is. Lots of
    fairly ordinary people (`more ordinary' than me) are told by
    `authority' that high cadences are what they should be aiming for.

    As for warmup - 20 mins is longer than most of my bike journeys !
    Ian Jackson, Jul 31, 2015
  7. Not with eight subjects. Doesn't even matter who they are, it's too
    small a sample to get anything statistically significant out of.

    The only thing it brings to the table is that it agrees with your ideas.
    It agrees with mine too, but that's not important either.

    Richard Dawkins, who for all his faults is absolutely right on the
    mechanisms of scientific research, might repeat "Please stop telling us
    what you feel. Please stop telling us what your intuition is. Your
    intuitive feelings are of no interest whatsoever, and nor are mine. I
    don't give a bugger what you feel, or what I feel. I want to know what
    the evidence shows."

    And this is no evidence.

    Cheers - Jaimie
    Jaimie Vandenbergh, Aug 1, 2015
  8. Er, I am a statistician, you know. It is complete crap that there is
    a hard lower bound, and even a sample of one can provide information.
    If you want me to teach some basic statistics here, I will, but it's
    not necessary for my point.

    Your error is that you are misinterpreting the information I was
    drawing people's attention to.
    Bollocks. What it does is provide hard evidence against the dogmas
    that higher cadences are always better, and that the optimal cadence
    is independent of power output, and it is THAT which I was referring to.

    In terms of representativeness of the population as a whole, or that
    of cyclists as a whole, or that of 'serious' cyclists, it is pretty

    In terms of the actual numbers, it measured ONE factor under ONE set
    of conditions, just as do most of the papers I am deprecating. The
    interesting thing is that it does match what those populations
    actually do, which most of the papers I am deprecating don't.

    Over 40 years, I have observed what people actually do, sometimes
    actually collecting formal counts. You can also check up the gear
    ranges (and hence cadences) on typical 3-speed bicycles in the days
    when almost eveybody used them, and those were what most people
    used (if they could afford them). It is possible that 95+% of
    cyclists are completely wrong about what is best for them, and that
    those manufacturers produced bicycles that were completely wrong for
    their customers, but I would like to see some evidence for that.

    Dawkins is the self appointed High Priest of Orthodox Evolution
    (which Darwin would have gagged at), and I have heard him claim that
    it is certain that there is no God. His evidence for that is
    exactly as strong as the evidence that there is a God ....

    Nick Maclaren.
    Nick Maclaren, Aug 1, 2015
  9. Nick Maclaren

    Peter Clinch Guest

    Alberto Contador's (amongst numerous others) track record of winning
    races up hills chugging along out of the saddle the whole way over the
    last several of years would have given you /that/...
    Peter Clinch, Aug 1, 2015
  10. Did anyone ever say "always" and "independent"? I've not met anyone
    saying so, so I wasn't considering that to be the point. The study
    certainly manages that.

    Cheers - Jaimie
    Jaimie Vandenbergh, Aug 1, 2015
  11. In uk.rec.cycling.moderated on 31 Jul 2015 17:18:20 +0100 (BST)
    When I first started riding the 'bent the received wisdom was to spin
    not to mash.

    So I spent a lot of time trying to keep at a fast cadence. I fitted a
    cadence monitor and tried to keep at 90.

    After a while I worked out for myself that for most long hills I
    needed to start in a lower gear early and try to avoid slow hard
    pushing because my speed would drop and my balance would suffer.

    On short hills I still mash...

    My general cadence now is about 70 or even less on the flat if I'm
    just cruising. It will go up a bit if I am trying to put on speed but
    I dunno I do 90 anymore.

    Did a 3 hour ride today, the only time the cadence got above 70 was on
    steep bits when the diamond frame riders were mostly out of the

    Zebee Johnstone, Aug 1, 2015
  12. God help me, YES, they DO :-( D. Tek refused to set up a tricycle
    for me to try because he regarded my cadence of 60 as too low, and
    told me to increase it. The same applies to saddle height, where
    I have had refusals to raise it even for a test in the shop! And
    that's just salesmen preferring to lose a sale rather than moderate
    their dogmas in the face of an experienced customer's request.
    I can assure you that many/most pundits without a financial interest
    are worse, and I have been subject to a lot of personal abuse.
    Why me and not most of the posters here? Because I physically
    CANNOT ride according to the modern dogmas!

    They also cause a lot of people to try cycling, and give up or cycle
    only as an alternative to walking, when they wouldn't have trouble
    with more appropriate advice, such as was normal in the heyday of
    cycling. And that's the serious aspect :-(

    Nick Maclaren.
    Nick Maclaren, Aug 1, 2015
  13. Nick Maclaren

    Jim Price Guest

    An engineer or a physiologist might have some understanding of the
    issues which someone just looking at the data from a statistical
    analysis point of view might not.
    An interesting pair of sentences...
    It's not the dogma that's relevant. Any dogma which says something is
    "best" is entirely subjective, and dependent on what is meant by "best".
    The science and engineering is that there is a power band for human
    cycling. Peak torque is most likely at 0rpm. Peak power you measure for
    each individual. The power curve is going to be measured against
    cadence, and it'll go up with rising cadence until other factors halt
    the rise and it'll start to fall away. Note that efficiency is also a
    rather loosely defined term unless you are specific about it. What the
    experiment is doing is trying to measure efficiency by a fairly specific
    method. Every subject participating in the experiment is already fairly
    fit, with a minimum peak power output of over 1200W. The results fit the
    power curve I described above. As to the matter of whether "the optimal
    cadence is independent of power output", that is because each cyclist
    has their own peak power, and their own specific power curve. There is
    going to be a range of cadences which will be so close to optimal that
    it makes no practical difference, and won't be measured by this
    experiment because the gaps in the data are far too big. The flat bit at
    the top of the power curve will be where the widest range of cadences
    will produce the most similar results in this experiment. Some people's
    power curves are going to be a lot flatter than others.
    The point of training, whether for racing or just general fitness, is
    that it is going to get you fitter. The ways in which it does that are:

    1. Improved technique.
    2. Extended power output, which is normally achieved by extending the
    power band, but doesn't have to be. The risk of doing it without
    extending the power band is that that implies improving torque delivery,
    and that is going to take you into knee damaging areas if you're not

    Humans can get a feel for what's efficient, albeit they can make
    mistakes in judgement if they're not careful, so tend to get more
    efficient with practice. However, if they only practice at low cadences,
    they will achieve higher efficiencies at those cadences at the expense
    of improving their usable power output. Sports training is aimed at
    maximum power output over periods of time useful for the sort of event
    being entered. The same principles can be used for leisure/utility
    cycling, but if you want to improve your performance (as opposed to
    efficiency) and power is the way you are going to measure your
    performance, then you are going to have to look at improving your power
    band, and that means learning to cycle efficiently at higher cadences.
    That does not mean you have to forget how to cycle efficiently at low
    cadences if you have already done so. Absolute performance isn't
    critical to leisure/utility cycling, so it's not even a choice you have
    to make if your current performance is acceptable, only if you want to
    improve your performance. That's not dogma, that's engineering physiology.
    Where's the control group?
    Regular users of 3 speed gear systems will train themselves to most
    efficiently use the available gear range, with the caveats above, if
    they do their "training" entirely informally. If they intend to stick
    with a 3 speed, they could still improve their performance by extending
    their power band, but there might be little point depending on what sort
    of cycling they intend to do. I shall be interested to see where you
    think the "dogma" is in all that, but possibly not interested enough to
    carry on the argument.
    Jim Price, Aug 1, 2015
  14. Bleh. Religious fervor gets tedious, doesn't it? Looks like I've been
    lucky (or just good at ignoring that sort of thing!).

    Cheers - Jaimie
    Jaimie Vandenbergh, Aug 1, 2015
  15. Nick Maclaren

    kimble Guest

    My take on it is that it's usually easier to get on with recumbents if
    you're a natural stay-in-the-saddle spinner.

    However, they do lend themselves to small amounts of extremely (indeed
    sometimes excessively) effective mashing, simply by merit of having
    something solid to push against. I avoid doing it because of knees, but
    I can't deny its effectiveness on short sharp climbs and when caught out
    by traffic. Sometimes it's the only way to get moving.

    The main issue is low-speed balance, which tends to be easier without
    low-frequency pedalling exacerbating the wobble. But of course
    recumbents are far from equal in terms of low-speed balance. Trikes are
    completely immune to the problem, and higher recumbents are usually
    easier to balance. A less easily balanced bike can still be ridden at
    low speed by a sufficiently able and skilled rider.

    At the end of the day, if it works, it works. It's all good. I just
    reserve the right to wince :)

    kimble, Aug 2, 2015
  16. Been there - done that. As did that paper. The whole point of
    its research was to get experimental data on topics that had been
    investigated only from those viewpoints.
    That is ONE point. Many elderly people and some others train in
    order to reduce pain or reduce the rate of decay, not to get fitter,
    that that often helps by partially correcting old damage. There
    are other reasons and mechanisms, too.
    Please don't be silly. Control groups are relevant only in hypothesis
    testing, and not always then, and this was estimation.

    Nick Maclaren.
    Nick Maclaren, Aug 2, 2015
  17. Nick Maclaren

    Jim Price Guest

    Only for fit subjects.
    Maintaining fitness or getting fitter is what does those things for you.
    What is an experiment if not a hypothesis test?

    The problem with your approach is you've decided on the conclusion and
    are now engaging in an exercise of selective interpretation of other
    people's research. That is bad science, and I'm not going to waste any
    more of my time on this.
    Jim Price, Aug 2, 2015
  18. Nick Maclaren

    soup Guest

    Yeah, but not everyone can average 40 miles an hour. :O)
    soup, Aug 3, 2015
  19. Your dogma is noted. It is not shared by physiotherapists. Both my wife
    and I have had treatment and been given exercises for the reasons I said,
    which do not increase fitness.
    Your ignorance of science is noted, as is your difficulty in reading English.
    Estimation is the measuring of an unknown quantity.
    Good, because the above paragraph is false in all respects.

    Nick Maclaren.
    Nick Maclaren, Aug 3, 2015
  20. there is that in that unless you want to be a sporting cyclist clipless
    high Cadences are more of hindernces than a help.

    even as some one who is a fit powerful rider, pootling to work I'm often
    40 to 50 rpm.

    might be 100+ rpm at times on the sportier bikes, but riding to
    quite my commute/shops etc is just under 20mins away.

    Roger Merriman
    Roger Merriman, Aug 3, 2015
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.