Login to VeloClub|Not a member?  Sign up now.
December 16, 2017
December 15, 2017
December 14, 2017
December 13, 2017
  • jules

    good article. in reality, the wind is rarely blowing along the same axis as the rider is traveling. i.e. there’s some x-wind. it would be interesting to study x-wind effects, how much drafting benefit can be attained and where to position yourself.

    • Liam Taylor

      Assume the ideal distance wouldn’t change for crosswinds, but the position on the wheel becomes essential, and often draft must be sacrificed in the interests of safety

      • jules

        when you’re chewing bar tape and desperately searching for maximum draft, you sacrifice nothing :)

    • I too found it a great article – minor point, cyclists also overcome a fourth force – mechanical resistance (but that’s perhaps nitpicking).

      On a related topic to crosswinds, the people over at FLO cycling studied wind yaw angles (for triathletes) and found that 80% of time was spent below 10 degrees YAW. This doesn’t answer your question, but really suggests that if these data are representative of cycling in general, cyclists don’t really spend a heck of a lot of time in the crosswinds to warrant wheelsets that are tested at crazy YAW angles. However, like I suspect you know from experience and as an avid cycling viewer, learning how to sit on the correct side of an echelon is valuable in reducing drag.

      • jules

        I’d qualify that a bit as while most riding is not done in strong x-winds, road races seem to be often located in non-metro areas with exposed, wind-swept roads. I’m unsure how that translates to > 10 deg. yaw, but even if it’s a minority of the time, that’s the point in a race that often counts the most.

        • DaveRides

          I have some doubts about the FLO Cycling data collection method (specifically with regards to having the windspeed measured so closely to the rider’s body, and why they didn’t just use a normal pitot tube) but they certainly were doing their data collection in well exposed locations on a number of major half-Ironman courses in the US which would be at least as windy as cycling courses.

          I would take the approach that it would be better to bank the energy saved by having a wheel optimised for the 80% of time, leaving me fresher for the minority of time. In the context of mass-start cycling races, this would also be helpful for final sprints in towns.

          • Karl

            Ah but triathletes don’t draft. Wouldn’t that change “effective yaw”?

  • Sufferfest

    I’ve always been curious as to effects of drafting with a strong tailwind. Does the presence of the back rider limit the “push” of the tailwind. Or if cross tailwind where should the back rider position themselves.

    • jules

      the appropriate measure is airspeed, as used in aviation. if you have a 40 km/h tailwind and you are riding at 50 km/h groundspeed, then you are riding into a 10 km/h headwind.

    • AlTilleythebum

      Also, is there a certain speed of tailwind at which it becomes more beneficial to sit up and catch more air (make your body a sail) instead of stay in an aero position? As someone who lives in one of the windiest states in the U.S., this is relevant to my interests.

      • jules

        see my post above. it’s not the wind speed. it’s your airspeed. if you are travelling slower than the tailwind speed, then it’s beneficial to make your body a sail. but you would go faster if you started pedalling.

        • unless like me you’re on a low-geared cargo bike much of the time, with a kid seat that acts as a sail (in both directions…) and above about 30km/hr you just spin out.

  • Lyrebird_Cycles

    Quote: Footnotes 1. The power required to overcome wind resistance is actually proportional to the rider’s speed cubed (assuming there is no wind). So
    the drag doesn’t just increase as speed goes up, it increases exponentially.

    Sorry that’s not correct.

    As stated Power increases as k.v^3 where v is the velocity. That’s a linear relationship (of the third order), for it to be an exponential relationship power would increase as k^v (k to the power of v).

    • jules

      don’t think so?

      • Lyrebird_Cycles
        • jules

          ok. I think the term ‘exponentially’ can also be used to refer to compounding rates of growth in a more generic sense, which fits here. but possibly that’s a bastardisation of the mathematically precise definition you’ve quoted.

      • DaveRides

        LC is correct.

        In certain contexts I’d accept lesser media outlets like Cycling Weekly using the term loosely, but CT should be above that – and should be extra cautious to use it correctly (or not at all) in an article aimed at providing a somewhat scientific analysis of a subject like this one.

        • jules

          I think it’s OK.

          “An exponential rate of increase becomes quicker and quicker as the thing that increases becomes larger”

          in maths, an exponential function is e^x where e is a constant of ~2.72 and x the variable, which is what LC is describing. but the term has been co-opted to mean anything that grows at an increasing rate (as LC and Nestor says below).

          this discussion is a bit eery as it brings me back to maths studies and I distinctly recall the same dilemma about how to describe these functions. I always felt that the common use of the term and precise mathematical one (involving the e constant) didn’t quite align. but there are bigger issues in life :)

    • DaveRides


      In an article aiming to provide a somewhat scientific analysis of a topic, this sort of error should be addressed to avoid it undermining the rest of the article. I hope the footnote gets corrected.

    • Nestor

      Well, not exactly. It is not a linear relationship – definition here: https://en.wikipedia.org/wiki/Linear_equation, and it is not an exponential relationship as you correctly point out, since the exponent is not the input variable – https://en.wikipedia.org/wiki/Exponential_function. It is actually a power function https://en.wikipedia.org/wiki/Power_function of the third order. But positively nonlinear.

    • I’ve updated the footnote slightly.

    • Alex Simmons


      The power – speed equation is cubic equation (which is as noted a power function, and not exponential one – a common mistake) however the equation has both linear components (i.e. gravitational potential, rolling resistance and wheel bearing friction) and cubic component (air resistance). Drive train friction in general is simply proportional to overall power demand. There is also a component for changes in kinetic energy, which in these discussions is often left out due to only considering steady state cycling.

      So from a practical POV, what that means is the power-speed relationship for steady state cycling is dominated by the cubic air resistance relationship on flatter terrain but becomes dominated by the linear gravitational component on steep positive gradients.

      So while a 10% increase in power might only provide a ~3.5% increase in speed on flat road, on a steep 10% gradient incline that 10% lift in power will get you a ~ 9.5% increase in speed.

  • Andrew

    Wheel sucking……. my favourite pastime

  • Altimis Nuel

    You also get more aero if your front wheel overlapped rear wheel of the front rider

    I seen GCN do the test, its reduced watts more than just directly behind rider

    Here video if anyone interest https://www.youtube.com/watch?v=meHOxRePOk8

    However, wheel overlapping is a bit extreme drafting, don’t do this to anyone who not your friend . . . its obvious

  • Ssanchez

    What about riding uphill into a headwind? It’s the worst

  • Alex Simmons

    I wrote about this 2 years ago after doing the aerodynamics experiment of comparing a rider solo, drafting another, being in the lead position, and also the impacts of riding side by side:

December 16, 2017
December 15, 2017
December 14, 2017
December 13, 2017