Why We Don’t Like Shoulder Knobs

When we developed our first knobby tires, we wanted tires that could handle all conditions, not just mud and snow. We wanted tires that rolled fast on the road, yet offered tenacious grip when conditions got really tough. We wanted a tire for adventures where we don’t know what lies ahead: pavement, gravel, mud, and maybe even snow.

There’s a standard way to make a fast-rolling gravel tire: Give it a smooth (or almost smooth) center tread and add corner knobs. The idea is that you roll on the smooth part most of the time, but you get added traction when you lean into a corner and need more grip. At first sight, it makes sense. And indeed, these tires roll fine on smooth surfaces, as long as you go straight. That’s no surprise, since you’re riding a tire with minimal tread.

It’s in the corners where these tires disappoint: The shoulder knobs don’t really help. When you’re on slippery surfaces, you simply can’t lean far enough to engage the shoulder knobs. You’ll start sliding while you’re still on the ‘semi-slick’ portion, and by the time the shoulder knobs engage, you’re already going down.

The opposite is the case on pavement. There you can lean much further, but once you get onto the shoulder knobs, they fold over and the bike suddenly breaks away. It’s not a good feeling… To add to the woes, when it gets slippery, you don’t get much traction, because the part of the tire that has the most pressure on the ground doesn’t have meaningful tread.

You don’t have to take my word for it. Gerard Vroomen from OPEN Cycles recently made the same assessment. He wrote in his newsletter about a semi-slick tire:

“The feature we don’t really like are the side knobs. In general we find side knobs superfluous on gravel tires. If you’re hanging at such an acute angle you’re about to hit the ground anyway. Even worse, the only terrain where you could achieve such an angle without falling is on asphalt, and the last thing you want is huge side knobs hitting the road instead of smoother rubber.”

Is there a way to make a better dual-purpose tire that works well on smooth and rough surfaces? What if you started with a slick tire and cut away the tread as much as possible? So instead of thinking of knobs, we thought about the rubber surface that we need on the road for speed and cornering grip.

You still need knobs, of course, to cut into the ground when it’s slippery. The big question is whether you can make those knobs big enough that they don’t squirm, yet small enough to dig into loose surfaces.

Our first calculations indicated this seemed possible. And since the large knobs don’t squirm, we could make them taller for even more traction without losing speed. That left the last question: Could we space the knobs far enough that the tread wouldn’t clog up with mud, but still offer enough support for fast cornering on pavement? Again, our calculations were promising.

Thinking about the tire in terms of rubber on the road also eliminated another problem with many knobbies: As you turn and lean the bike, you first roll on a line of knobs, and everything seems fine. When you lean just a little bit further, you get off the knobs and suddenly have much less rubber on the road. We placed the knobs so that you always have the same amount of rubber on the road, no matter your lean angle. On our most recent models, we slightly increase the rubber surface on the shoulder knobs to give you a little extra traction when cornering hard on pavement.

All calculations only take you so far – the proof lies on the road. Before you can get even a single tire for testing, you need to invest into a very expensive tire mold. It’s always exciting when the first prototypes of a new tire model arrive. After hours, when the production staff is gone, the engineers at the tire factory heat up the huge, brand-new tire mold to make just a very small batch of totally hand-crafted tires. As soon as they come out of the mold, they are airshipped to Seattle. They still smell of fresh rubber when we open the parcel. There’s no packaging, and they usually don’t have stickers yet. When we mount them on our wheels, we see the tire tread for the first time in real life, rather than just on a screen or on a sheet of paper.

And then it’s time for the first test ride. I’m happy to report that our dual-purpose knobbies exceeded expectations. They excel in mud, but that was to be expected. They hook up great in gravel – but so do most all-road tires. They work well on dirt and hardpack, too.

We wondered mostly about their on-pavement performance. And there the tires exceeded our – very high – expectations. It wasn’t just the cornering – although you wouldn’t want to lean most knobbies into a corner like this. (Sorry for the out-of-focus photo – the camera had a hard time catching up on the typical Seattle day when we first tested the tires.)

The biggest surprise was the speed. The tires felt fast. On familiar terrain, I was riding in the same gears that I use with our smooth all-road tires.

I took a bike with our first knobbies to an all-paved BQ team ride and asked Mark to switch bikes. We’re almost clones of each other with the same height, weight, build, power output and pedaling style, which is ideal for testing the relative performance of two bikes/tires/other components.

On this morning, he looked at me, puzzled: “Why would I ride knobbies on a paved ride like this?” We never share our impressions before a test is complete (to prevent what scientists call it ‘confirmation bias’), so I just said: “I want to get a feel for how much speed we lose when riding knobbies.”

We switched bikes several times during the ride. Mark was running our Extralight all-road tires, which are known to be among the very fastest tires you’ll find anywhere. Between his bike and the one shod with the knobbies, there was no noticeable difference in performance. Whoever was on the knobbies had no trouble keeping up.

Sorry, I have no photos from that ride. We were going all-out, and there was no time to pull out a camera! Mark later commented that only the tire noise reminded him that he was on the bike with knobbies – and during a fast, twisty descent, when the wind noise drowned the hum from the knobs, he forgot that he was riding knobbies altogether.

From looking at our dual-purpose tires with their big knobs, you might think of them as clunky, heavy and slow on smooth ground. The tread looks almost like a mountain bike tire. But that’s a misunderstanding – these aren’t knobbies in the conventional sense, but slick tires with much of the tread cut away. You could ride them on a fast-paced road ride like we did, and have no trouble keeping up. They are faster than many ‘road’ tires.

So why don’t we ride knobbies all the time? The noise is an obvious drawback, although we are working on that. They have a little less directional stability, because the tread isn’t continuous –there is less of what experts call ‘pneumatic trail.’ They are a little heavier, too. And on a bike with fenders, you need more clearance, since the knobs can pick up stones and sticks that could get stuck in the fenders.

Those drawbacks are minor: For a ride where the conditions ahead are uncertain, even if it’s 90% paved and only 10% gravel and mud, I’d pick our dual-purpose knobbies every time.

Photo credits: Natsuko Hirose (Photo 1), Heidi Franz (Photo 5), Ryan Hamilton (Photo 6), Rugile Kaladyte (Photo 7).

23 Responses to Why We Don’t Like Shoulder Knobs

  1. Tim March May 12, 2020 at 10:54 am #

    Looks like a mitsibishi comp 2 1.75 tread from the 80s BMX scene.

    • Jan Heine May 12, 2020 at 2:15 pm #

      Mitsuboshi made some great tires, but these are quite different. The distribution of the knobs is key, not just their size and shape.

  2. Rick May 12, 2020 at 10:54 am #

    You are absolutely right about this. I tried the Specialized Fast Trak tires when they first came over 20 years ago. About the same tread as on the Moots photo above. I’ve been riding for 58 years now and rarely fall. Those tires put me down 4 times in a 1 week tour. Just a slow speed U turn on a gravel road and you are landing on your hip on gravel. I took those off my bike after just 1 tour. For real gravel roads you need mid to large widely spaced knobs, not that fine beaded pattern that shows up on many new “gravel’ tires. Here in Iowa, over 50% of the roads are gravel. Mostly crushed limestone about 1” or larger. Some counties use a mix of sand and river pebbles mixed with Loess hills dust for lubrication. You learn pretty quick that each mile is maintained on a different road work schedule by each of the 99 county road departments who each do it their own way.

  3. Tom Anhalt May 12, 2020 at 11:54 am #

    I would have to argue that the suppositions you and Gerard make about “file tread/side knob” type tires on dirt and pavement aren’t universal.

    For one, on dirt, side knobs are also useful for traversing off-camber surfaces, where the lack of a side knob can easily cause one to go down (e.g. I’ve painfully had that experience when riding Snoqualmies after being used to other tires with side knobs).

    Secondly, I’ve experienced certain tires which corner surprisingly smooth and secure on pavement. One of these is the Challenge Grave Grinder Pro, which has quite widely spaced knobs. Some of the common routes I take around here end up with a descent of the paved Gibraltar Road (featured as a finishing climb in the Tour of California a few years back), and the cornering performance of these tires is actually quite outstanding. My suspicion is that the handbuilt nature of these tires, with the 260tpi casing and glued on tread (allowing for a very soft, grippy compound) contribute much to the good cornering performance.

    I’d also like to see an objective “on pavement” test between your knobby models and the smooth equivalent models, measuring actual power differences.

    • Jan Heine May 12, 2020 at 10:40 pm #

      Totally agree that knobs are useful in many situations. That’s why we offer knobbies… Like you mention, it’s even possible to conjure up situations where knobs on the side of a tire that is slick in the middle could be useful. I still prefer tires that don’t trade extra grip on off-camber curves with unpredictable cornering (and a risk of falling) everywhere else.

      As an aside, the 260 tpi casing you mention is just marketing speak. You can’t really make tires with threads that thin – they wouldn’t be strong enough. The thinnest threads you’ll find on any tire are those on our Extralight and Endurance casings. The resulting tpi is 120 in a dense weave (Endurance) and somewhat less in a looser, more supple weave. Some companies count each casing layer – usually there are 3 – and a 120 tpi casing becomes 360 tpi. It’s nonsense, of course, otherwise, doubling up the casing with six layers would result in a 720 tpi casing, even though the tire would be far less supple.

      Regarding the tire tests of knobby versus smooth, we are working on that. We’ve got preliminary results, but we want to fine-tune our testing methods to get better results before we publish them, plus test other factors such as knob height, since BQ articles are always about basic research and not just a ranking of tires.

      • Tom Anhalt May 14, 2020 at 9:46 am #

        Understood…the preferences for knobs vs. texture in the center can be a personal one. I actually think that side knobs don’t necessarily mean unpredictable cornering on pavement, which is something Gerard actually mentions later on in that article you linked to. I also value relative speed on pavement, something that even the best knob designs give up something measurable. Much like gravel/all-road bikes, sometimes the best compromise isn’t optimized for the extremes of the use envelope, right? 😉

        On the tpi issue…are you speaking of fully vulcanized tires? I know that Continental in particular is notorious for “adding up the layers” (i.e. they claim 330tpi on models that have 3 layers of 110tpi material), but most other manufacturers don’t list anything beyond 120-127tpi for their fully vulcanized models, while the tpi listings for their “hand built”, or “glued-on tread” models list the TPIs in the 200s. As I understand it, those casing TPIs ARE the true TPI, not a Conti-style addition of overlapping layers. If that wasn’t the case, then why wouldn’t they do the same on the fully vulcanized models?

        Good to hear on the testing. If you’re open to suggestions, just hit me up 🙂

        • Jan Heine May 18, 2020 at 8:49 am #

          I talked to François Marie from FMB, and he confirmed that all the ultra-high TPI numbers are counting multiple layers, even on hand-glued tires. He also mentioned that TPI by itself doesn’t tell you that much, as the quality of the threads is as important as how thick they are. And then you get the issues of cramming your threads into a dense weave to increase TPI, because that’s what many cyclists look for, even though that creates a stiffer tire. As a result FMB doesn’t list TPI numbers, even though nobody doubts that their tubulars are the best in the world. (That’s why pros buy them for the really important races, even though they get ‘standard’ tubulars from their sponsors for free.)

    • Mark May 13, 2020 at 3:34 pm #

      Q’s:
      1. Performance on wet pavement?
      2. Wear rate vs your slick tyres?
      3. Given your claims, feasible/advisable/preferable to ride these all the time (esp. if c.10% of riding is on gravel or lightly loaded touring on unknown roads)?
      4. (Related to the above) do you ride these in Japan?
      4. Mudguards—how much clearance do you recommend? Or to put it another way, what size mudguard for each size of your knobby tyres?

      • Jan Heine May 13, 2020 at 6:19 pm #

        Good questions.
        1. They work fine, but I prefer our fine chevron all-road tread. It interlocks better with the pavement – see this post.
        2. They wear a bit faster, since there is less rubber on the road. Still, you can expect 3,000-6,000 miles out of a set, depending on the tire width, your weight, and your power output.
        3. It’s feasible, if you don’t mind the bit of extra noise. I’ve been on rides with more than 100 miles of pavement and didn’t mind taking these.
        4. Sometimes. I took them on a test bike, and really enjoyed riding on road, trail and snow on these.

        • Jan Heine May 13, 2020 at 11:34 pm #

          I overlooked 4.2 (mudguards/fenders): You really want a lot of clearance with knobbies. I’d design my bike with 30+ mm clearance between tires and fenders if I planned to run knobbies.

  4. Sean May 12, 2020 at 1:36 pm #

    Love my Rene Herse 650b knobbies! Now I just need some for my 26” bikes. Please bring on the knobby RTP!

    • Matt Flanagan May 13, 2020 at 1:52 am #

      I agree on a knobby RTP. I’ve just blown out the side wall on my rear Rat Trap Pass extra light and now I’ve got the dilemma of buying another one or trying something like the Panaracer Gravel King SK. Its now it’s getting on to winter here in Australia the slicks don’t cut it on the greasy sections of dirt roads. I‘m now running Specialized Fast Tracks but they will wear down way too fast on the long sections of pavement. I have loved the RTPs including on corrugations where they are amazing

  5. craigdemp@gmail.com May 12, 2020 at 2:56 pm #

    My rear Switchback Hill is near the end of its life. Thinking of switching the front to the rear and putting a Juniper Ridge on the front. Is that the way to roll? The only time I have had traction issues with the Switchback Hill is when there is a bit of greasy slop on top of Forest Service roads and it feels as if it is the front that wants to slide out from under me.

    • Jan Heine May 12, 2020 at 3:41 pm #

      You always feel the grip bleeding away at the front first, but the rear is just as likely to slide. Generally, at moderate and high speeds, you turn the bars much, and you need the same grip front and back. At lower speeds, and especially in technical terrain, the front tire does take extra side loads when you turn the bars suddenly. There, a knobbier front could be useful. That’s probably why gravel tires are the same front/rear, while mountain bikes often have different tires front and rear. Maybe… Nothing wrong with running a Juniper on the front and a Switchback on the rear – or the other way around if you’re likely to encounter during an uphill snow and want to optimize traction.

  6. Eric May 12, 2020 at 5:50 pm #

    Have you thought about making a 700c tire with the knobs of the Hurricane Ridge in a wider tire, like 45-47mm? More and more all-road bikes can take a bigger tire now, e.g the new Specialized Diverge just announced.

    • Jan Heine May 12, 2020 at 6:11 pm #

      We’re about to launch the 700C x 55 mm Fleecer Ridge…

      • Owen May 12, 2020 at 9:48 pm #

        Looking forward to these. Is 55mm the approximate casing width or the width knob to knob?

        • Jan Heine May 12, 2020 at 10:01 pm #

          Both are the same on these tires, since the knobs don’t stick out further than the casing. So it’s got more volume than a 2.2″ mountain bike tire…

  7. chris dresden May 12, 2020 at 8:21 pm #

    I got a set of Juniper Ridge (650Bx48) for a Masi Giramondo. I air them down to 35 psi for gravel and they have worked wonderfully. I can even stand on climbs and get minor slipping. Here in Idaho we have lots of deep sand, washboard, and gravel so the tires have been thoroughly tested. On the pavement I inflate to 50 and ride like a conventional road tire.

  8. marmotte27 May 13, 2020 at 2:45 am #

    Was thinking about getting some for my winter bike (for when there’s snow and, what’s worse, salt on the roads). But now it hardly snows any longfer…

  9. Keith Benefiel May 13, 2020 at 10:09 am #

    Last year our Team Medicare , Abuse Tours, took an 80/20 gravel/paved rail trail over four days along the west border of Yellowstone Park. I usually rode Babyshoes on similar routes, but used Pumpkin Ridges instead. They handled everything with aplomb; big rocks, little gravel, washboard, ruts, pothole/ puddle slalom and a detour stretch covered with Bentonite gumbo that stuck to everything EXCEPT my tires. On a loaded bike, they were undetectable on steep road sections, any noise too subtle to detect and handling the same as usual rubber. Babyshoes still roll better on asphalt, but the Punkins are unexcelled on rough stuff and the choice for any mixed ride.

  10. Johan Larsson May 15, 2020 at 12:07 am #

    Other than a well made knob design and pattern, an interesting option should be to go the way that Jobst Brandt did when he designed tires for Avocet. They had an inverted wide single groove 90 degree zig-zag pattern. Looks like nothing else.

    Apperently they were the fastest of all tires when tested by a variety of riders in a controlled test by Avocet, but likely due to potential problems with advertising losses from the large tire manufacturers, no magazine wrote about it and the test results were quiet down.

    I believe this pattern should have all the qualities you’re looking for in an all-round pavement/dirt tire, relatively silent and with a predictable grip when leaning, both on pavement and soft surfaces. I hope the original writings and info about these tires are still available in some Jobst Brandt archive somewhere, as it was many many years ago I read it. A quick Google search revealed nothing. I actually once found a pair of these tires on a bike in a repair shop I was working in, but they were rotten and completely broken, so I never got a chance to riding them.

    • Jan Heine May 15, 2020 at 7:41 am #

      We had a lot of Avocet fans on our team, and we tested the Avocet Cross in our original BQ tire tests in 2006. Even back then, they were already out of production. They rolled surprisingly fast, but they were heavy and the tread didn’t do anything positive off-road. It just clogged with mud immediately. As Jobst pointed out numerous times, bikes don’t hydroplane, so the grooves that you find in car tires to evacuate water aren’t necessary on the road, either. Still, the Cross showed that you could make a tire with thick tread roll fast, if you cut away the tread to allow it to flex. Knowing that made us think about knobbies that took the idea of negative tread to the extreme…