The Seamless Experience

The Seamless Experience

I recently wrote a post concerning frame stiffness and a phenomenon I encountered in which some of the bikes I’d ridden on dirt road descents would begin to oscillate if the surface was particularly rough in turns. Based on the feedback I got from several smart people I know in the bike biz, the culprit appears to be a lack of sufficient stiffness in the frame.

I’ve had a few questions about why increased frame stiffness always seems to be everyone’s answer to a better performing bike or, alternately, what is wrong with a bike that lacks stiffness.

The first point I want to make here is that there is no one correct answer to frame stiffness. Provided the bike isn’t inherently unstable due to a poor design, an experienced cyclist can ride a bike with very little stiffness, and do so well. The simple proof to this is Sean Kelly. For a great chunk of his career, one of the finest riders of his generation rode a Vitus 979. I’m willing to bet a Ben Franklin that the Vitus 979 was the least stiff bike ever ridden in the Tour de France (caveat: in the modern era). And yet, despite the fact that the bike was extremely flexible, King Kelly won stages of the Tour de France, and great classics, such as Paris-Roubaix, where that softness did him favors on the cobbles, but conferred no benefit in the final sprint.

Kelly is proof that any rider can get used to most any bike. With experience, a rider can learn how a bike flexes in a sprint or on a descent and muscle memory simply incorporates that experience into their expectations. And that’s the key, right there, expectations. When a rider expects a bike to react to input in a certain way, and the bike doesn’t do that, that surprise can be unnerving.

Case in point: Some time in the late 1990s, I was on the Donut Ride in Palos Verdes and at the bottom of a ripping descent into San Pedro the road bends left. What I didn’t know prior to entering the turn was that my front tire had a slow leak and its pressure was somewhere south of 40 psi. I entered the turn and the sidewall of my lightweight clincher began to collapse (known in the mountain bike world as tire squirm) and in that instant I could feel the bike begin to oversteer as a prelude to low-siding.

On a scale of 1 to 10, my pucker factor was about 34. I can still feel the jolt of adrenalin as it shocked my body.

I straightened the bike out to get the tire back under me, hit the brakes and then, because I was headed for a stone curb, began to ease back into a turn even as I continued to brake. I fixed the flat and then chased the group (never caught them, natch) with a couple of friends.

It was on that same stretch of road perhaps a year before that I was riding a frame I knew with an especially light carbon fiber fork from Look. I expected the bike to enter that turn and I’d carve through and get back on the gas ahead of a thundering sprint effort. But the feedback I was getting at the handlebar seemed off, like I couldn’t quite tell exactly what sort of arc the bike was carving, and then the bike began to oscillate; I braked until the oscillation stopped and found myself at the back of the group. That Look fork was absolutely the most flexible fork I ever rode, though I’m aware of a titanium fork that I’m told was even softer. Ugh. Could I have become accustomed to that fork and the way my bike handled with it? Sure. But I’m not sure I’d ever have been able to attack descents in the same way because the way the fork handled conflicted with my need for a precise sense of what the front wheel is doing, where it is pointed.

A corollary to this is the high-speed wobble; I got to speak with Ben Serotta about this issue many years ago as I was working on a review of a Serotta. Other colleagues reported about this as well: Serotta did research into what causes a bike to enter a high-speed wobble. The problem is simple enough: at higher speeds the front wheel begins to oscillate left and right and as speed increases the rate of the wobble increases as well. The folks at Serotta never did pin down an absolute cause, but when their customers reported a speed wobble in one of their bikes, they would ask them to ship the frame back and they’d simply remove the top tube and replace it with one that was stiffer. A stiffer top tube eliminates a speed wobble.

Stiffness is, if I am to be fair, not the be-all, end-all of frame design. If I lived in a flat place and wasn’t racing criteriums, a flexible frameset wouldn’t be much of an issue because I wouldn’t be putting all that much force into the bike in turns. Never once, in all the years I lived in Memphis, did I break 40 mph on a bike, and the closest I got to that threshold came while I was going perfectly straight.

A listener to the Paceline brought up Jan Heine’s taste for more flexible frames (thin-walled tubing with a traditional 1-in. top tube and 1 1/8-in. down tube); when you add in his preference for 650B wheels with tires pumped to relatively modest pressure, it becomes apparent that the Bicycle Quarterly publisher does not find a lack of stiffness to be an issue. A rider who has lots of experience on more flexible frames can move from one to another without much issue. However, moving from a stiffer bike to a more flexible one, or vice versa, will cause many riders some confusion.

The phenomenon that occasioned my investigation—oscillation in a turn on a rough, unpaved road—is a simple fact of a more flexible frame. Could I learn to live with it? Certainly. So what are the consequences? I simply wouldn’t descend as aggressively. What’s the problem with that one wonders? Well, I like going downhill fast. That’s where the flow is.

So, lest a reader think I’m advocating that more stiffness = better, my larger point is that a bike needs to handle in a predictable way. A stiffer frame and fork can yield a bike that handles more consistently. But predictability comes in a few different forms. Consider my observation about the slow leak I had on the descent. The front tire behaved in a way that surprised me. The unexpected = no bueno. That returns me to another important point: pumping a bike’s tires up to the same pressure before each ride is key to developing critical muscle memory.

So stiffness isn’t the goal itself, but a characteristic that serves a goal: predictability. The more that a bike’s handling fits with the rider’s expectations the more assured a rider will be, and if a bike doesn’t respond as anticipated, then muscle memory can’t take over and without muscle memory guiding a rider’s actions, it’s impossible to feel confident on the bike.

Predictability contributes to a rider’s experience in what is one of the most important aspects of a rider’s relationship to the bike. That feeling of becoming one with a bike happens when feedback from the bike is instantaneous and conforms to a rider’s expectations. At a certain point that approaches a flow state, but may not be exactly correlated to flow, a rider’s parietal lobe begins to shut down. It’s a phenomenon known to happen in meditation, some religious experiences and mystical states. When the parietal lobe goes on break, our sense of where we end and the universe begins gets blurry. At its least extreme we experience things like being one with our bike, feeling the road surface while driving our car or, perhaps, typing with such fluidity that we simply see words appearing on the page.

Ultimately, it’s that seamless experience of merging with the bike that leads to our most rewarding rides.

, , ,


  1. Tom Milani

    Is there a correlation between stiffness and comfort? Our first tandem was aluminum. Definitely stiff, and I definitely felt it the next day after a 50-mile ride, despite the 2-inch tires. Our current one is titanium, running 35 mm tires, and it doesn’t beat me up. I don’t know how carbon compares or if there’s a difference between stiffness and shock absorption. I do like the idea that predictability is a precursor to flow.

    1. Author

      The short answer is yes, there is a definite correlation between stiffness and comfort, if you normalize other factors like wheels, tire size and pressure. However, you can gain more comfort in running a bigger tire at lower pressure than you can lose in comfort by riding a stiffer frame.

  2. 32x20

    It would be interesting to know if the bike you experienced oscillation on would behave differently with a different set of tires/pressures/wheels. Bicycles become such systems of components that it seems hard to pinpoint an exact cause. A stiffer frame may solve it, but one wonders if there is some other combination of features (geometry, tire characteristics, rider weight distribution, etc) that cause it.

    My interesting anecdotal experience: I’ve never experienced speed wobbles on a non-department store bike. I purchased an e-cargo bike from a shop mechanic last year. He warned me that at a certain speed it’d wobble. Despite having it loaded (and unloaded) in any number of ways and speeds from 0-34mph I’ve never experienced it. Weird.

    1. Author

      I’ve ridden that bike with three different sets of wheels and at least a half dozen different tires. On unpaved surfaces the bike would oscillate any time I braked in a turn or even if I wasn’t braking, but the turn was, um, lumpy.

      Speed wobbles are a really interesting phenomenon. I would be interesting to see what factors were different between you and the mechanic. Sometimes, eliminating a speed wobble can be as simple as shifting the rider’s weight forward a few centimeters. I can recall watching sprint finishes of big European races on TV and every now and then you’d see a rider sit up to throw his victory salute and the bike would handle fine until he was sitting perfectly upright with little weight on the front wheel. It was not uncommon in those incidents to see the rider clamp his knees against the top tube to stop the motion.

    2. Shawn

      @32X20 I was thinking the same. A lax wheel, spinning as it does, would seem to be a prime opportunity for an oscillation looking to spoil a descent.

  3. Steven Soto

    I have come to love my Cane Creek Viscoset on a bike that was susceptible to speed wobble no matter what I tried prior to installing this damping headset. Now, no matter what I do I cannot get the bike to speed wobble.

    Even the lightest damping configuration eliminated the wobble and I don’t notice it otherwise.

  4. Scottg

    It is wonder anyone is still alive who rode a bike pre 1990,
    what with 25.4 tt and 28.6 dt, and MAFAC brakes.
    The bottoms of hills were piled with the bones of cyclists. (smiley)
    Padraig, how fast is fast ?

    (Yes, there were oversize tubes pre 1990, but they weren’t in
    professional races)

    1. Author

      Mafac brakes. Yeah, that’s a brilliant shorthand for how much better stuff is today.

      The great thing about fast is how relative it is. Fourteen miles per hour on singletrack can feel as fast as 40 mph on the road.

    2. Bryin

      MAFAC brakes were no problem… (neither were the crap Nuevo Record or Super Record brakes)… brakes on a RACE bike are not really there to stop you (only people that don’t think ahead need to stop suddenly) they are there to slow you down and because it is a RACE, you don’t want to slow down too much. Shimano came along and made brakes that worked great… my second bike came with 105 6 speed and no one ever needed brakes better than those.

  5. Bryin

    Not only was that Vitus flexy but when you see a picture of Kelly on his Vitus and it is definitely NOT a “race fit” position. Not very “aero”. But that did not slow King Kelly… nor did friction shifting or clips and straps… both of which Sean used well beyond the point the rest of peloton had embraced them. When asked why, Kelly merely shrugged his shoulders and said he had not gotten around to trying them. Probably too much work to do on his farm when not riding. But I am sure of one thing, you put Kelly (at his prime) up against Sagan 10 times, they would probably split the matches even. Even with Kelly in clips and straps and on his Vitus.

Leave a Reply

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