Why 700C wheels DON’T roll faster

These days, it’s widely accepted that 700C wheels roll faster than smaller hoops. It’s a bit like the old belief that narrower tires are more efficient – so self-evident that it doesn’t even need explaining. After all, simple physics tells you: A bigger wheel rolls faster, because it hits bumps at a shallower angle, and it bridges small holes better, too.

It’s just that we’ve been unable to confirm this supposed superiority in the real world. In Bicycle Quarterly’s tire tests, we’ve tested different wheel sizes in a number of tests. Our first tire tests included roll-downs of identical tires in 700C and 650B wheel sizes. We later ran 700C, 650B and 26″ tires on smooth pavement and a very rough surface (rumble strips of a highway) with a power meter. We also ran them on a track, again with a power meter. The results are always the same: There is no measurable difference in how fast the three common wheel sizes (26″, 650B, 700C) roll on real roads – paved or gravel, even cobblestones – as long as you keep the casings, tread and other factors the same. In other words, a 650B x 38 mm Rene Herse tire is as fast as a 700C x 38, and so on.

So what about the physics? Why doesn’t a big wheel roll faster? Let’s try to figure this out…

With a solid wheel, it’s easy to see why the big wheel on the left rolls more easily over the same-size bump than the small wheel on the right. By making the wheel larger, you make the bump smaller in relation to the wheel. Everybody knows that it’s easier to go over small bumps than over big bumps. The same applies to holes – a larger wheel makes the hole smaller.

That’s why horse-drawn carriages had very large wheels. It allowed them to travel faster and more comfortably over rough roads. It’s important to note that the carriages had solid wheels that were (almost) completely round, which made the physics very simple.

So why don’t cars and buses (and bikes) have huge wheels, too? It’s because their wheels aren’t solid. A wheel with air-filled tires behaves differently. It is not round, but flat at the bottom. The weight of rider and bike flattens out the tire where it touches the road surface.

Small obstacles are absorbed by flex in the tire, not by lifting the entire vehicle over them. How fast the wheel rolls is determined by how much energy the tire consumes as it goes from round to flat, and back to round again with each revolution of the wheel. Wheel size doesn’t really matter as long as the bumps are small enough so they can be absorbed by the tire.

If you want to absorb bigger bumps, you make the tire taller/wider, not the rim bigger. A good illustrations are off-road race cars. Looking at the winner of the famous Dakar Rally (above), the first thing we notice are the huge wheels. But when we look closely, we see that the rims are relatively small. In fact, they are as small as possible: The brakes have to fit inside the rims. Look at the front wheel, and you can see the (silver) brake rotor filling almost the entire rim, with just enough clearance for the brake caliper. The rims can’t get any smaller without giving up braking performance.

So we’ve got small rims with huge tires, designed to absorb bumps without the wheel having to go up and over each bump.

The same is true for bikes: Especially on rough terrain, a bigger tire on a smaller rim will roll faster than a narrow tire on a large rim. In other words, a 650B x 48 mm tire will roll faster than a 700C x 28, even though both have the same diameter. (Or in the photo above, a 26″ x 2.3″ compared to a 700C x 23 mm.)

What if we keep the tire width the same – say 48 mm – and just make the rim (and hence the entire wheel) larger? For example, what if we compare a 700C x 48 tire with a 650B x 48?

In that case, there is a factor that could make larger wheels faster: The shape of the contact patch is different. For a given tire construction, air pressure and load, the tire’s contact patch will be the same size. If the tires have the same width, the contact patch will also have the same length. On a larger wheel (left), this means that the tire doesn’t have to flex as much as on a smaller wheel (right).

Look at how close the rim comes to the road surface, which shows how deeply the tire deflects: The small wheel (right) flexes the tire until it’s about 50% as tall as it is unloaded. The large wheel (left) flexes much less. And less tire deformation means smaller hysteretic losses and less rolling resistance.

For the same reason, wider tires faster. (In that case, the contact patch is shorter, so the tire also deflects less.) So in theory, the wheel with the lowest rolling resistance is the tallest and widest you can fit into your frame. In practice, once your tires are tall enough and wide enough, the benefits of going wider/taller is too small to be of any importance – which is why we can’t measure it. It’s also possible that the benefits of larger wheels are canceled by their increased air resistance.

It’s important to remember that the common wheel sizes are very close in diameter. The smallest 26″ wheels are just 10% smaller than the largest 29er/700C wheels. It’s not a big difference… And the drawings showing the tire deformation are exaggerated – in the real world, tires don’t deform that much.

The important question is not

How much does the tire deform?

but

How much energy does it take to deform the tire?

And that is where supple casings come in. The difference in energy absorption between a stiff and a supple casing is much greater than all other effects, for a number or reasons:

  • A supple tire needs less energy to go from round to flat and back to round again.
  • A supple casing makes the tire absorb bumps better without lifting the entire bike. A stiff casing behaves more like a solid wheel.
  • A supple tire reduces the suspension losses that are caused by vibrations of the bike.

At this point, you may think: “I understand that the effect of wheel size is small. And yet, once you’ve optimized your tire casings and everything else, why wouldn’t you run 700C wheels? It seems that they are faster, even if it’s too small to measure in the real world. What is the down side?”

If you’re lining up with racers like Ted King (No. 1),  Lachlan Morton (2361) and Payson McElveen (2361), you’ll need any advantage you can get, even if it’s mostly psychological. If you’re running tires in the 38-44 mm range, like most gravel racers, there’s little to lose by running 700C wheels. Since everybody else tends to run them, it makes sense to do so as well, just in case we’ve overlooked something in our analysis. If you don’t worry about making the wrong wheel choice, you can focus your energy on winning the race. It’s a bit like replacing your water bottle screws with titanium ones – it can give you the confidence that you’ve done all you can to optimize your equipment.

Then why would anybody run 650B wheels? There is an advantage to the smaller wheels: You can fit wider tires on your bike. And as we’ve seen above, wider tires can absorb larger bumps and roll faster over rough terrain. A second benefit is that the smaller wheels make the bike more nimble, which helps during fast, twisty descents. Huge 700C tires will make your bike more stable, which may be desirable for a 29er mountain bike, but can make an all-road bike tricky to handle during fast, twisty descents.

Ted King’s tire choices illustrate this. For last year’s big gravel races, he ran our 700C x 38, 42 and 44 mm tires with great success. This summer, he’s on 650B x 48 mm Rene Herse Switchback Hill tires. A few weeks ago, he rode more than 300 miles non-stop across the state of Vermont, almost entirely on dirt roads. He reports that he felt faster than ever, because extra air is more important than extra wheel diameter, and supple casings have greater benefits than both.

Once gravel races start up again, we’ll see what tires he’ll choose. I wouldn’t be surprised if he’s back on 700C, because he doesn’t need to worry about whether his wheels are faster or slower than his competitors’ on any given section of the course. Or perhaps he’ll continue to ride his 650B wheels, because he’ll want to keep the advantage that the wider tires give him on rough ground.

Conclusion: Dial in your wheels based on how you want your bike to feel. Run 700C wheels if you like, but don’t worry about giving up speed if you are on 650B. Just make sure your tires are supple.

Credits: Ansel Dickey (Image 1, 9, 10); Natsuko Hirose (drawings); Peugeot (Image 5)

38 Responses to Why 700C wheels DON’T roll faster

  1. John July 17, 2020 at 5:16 am #

    This is something I’ve always wondered about. If the difference between 700c and 26″ rolling resistance is to small to measure, then how small is to small? Why not test 24″ wheels or even 22″? It seems to me that all else being equal, you would want to run the smallest wheels you could for all the same reasons you cite for preferring 650b over 700c — lighter, stronger, and stiffer wheels. If wheels smaller than 26″ could be shown to be just as quick, and not have any other unforeseen disadvantages, then it would open up a huge range of frame options. The most obvious I can think of would be at the interface between the bb shell and chainstays. With a smaller rear wheel, the stays wouldn’t have to thread such a delicate line between the rear tire, crank arms, and chainrings. Any weight you carried over the wheels on a rack would also be centralized lower. I have a Brompton with 16″ wheels and I definitely prefer my touring bike with 26″ wheels when I have the option. I suspect the ideal wheel size is somewhere between 20″ and 24″.

    • Jan Heine July 17, 2020 at 6:52 am #

      We tested the common wheel sizes, because that is the question most riders ask themselves. The main reason for choosing wheels within a small range of diameters is because of the handling they offer. It’s the same in motorcycling, where scooters haven’t replaced large-wheeled bikes, and where the rim diameter has gone down as tires became larger, then up again as low-profile tires were introduced.

    • Jacob Musha July 17, 2020 at 8:05 am #

      I had a fork made for my Velo Orange Neutrino to obtain a 73 degree head angle and about 30mm of trail, essentially the same as my custom 26″x52mm tire bike. With 20″ (406) wheels and 50mm tires, the Neutrino rides and handles well enough. It can be ridden no-hands. But it’s twitchy and not as stable as I’d like.

      Based on my experience I would say 20″ is not an ideal wheel size unless the rider is too short to use anything larger. 26″ + wide tire, 650b + medium tire, or 700c + narrow tire seem to be ideal and were probably not chosen by random accident.

      • Mark P. July 18, 2020 at 10:15 am #

        I’m curious why you wanted to change the trail of the Neutrino. In my own trial-and-error experiments with 20″ wheel bikes, I found that trying to match the trail of larger-wheel bikes didn’t result in handling that I liked with 20″ wheels.

        • Jacob Musha July 19, 2020 at 4:51 am #

          Mark, the main reason for the new fork was to use a cantilever brake (I setup the bike fixed gear.) Of all the silly applications of disc brakes, a travel bike seems like the worst possible one. The new fork is also half a pound lighter…

          Anyway, the new fork gave me a chance to experiment with offset. I think Jan implied that Hirose’s mini-velos that handle well use similar geometry as a larger-wheel bike: steep head angle and low trail.

          I briefly setup the bike with the original fork (no brakes!) and didn’t notice a big difference in handling. What is your preferred geometry? Does it retain the straight-line stability of a large-wheel bike?

    • Alex July 17, 2020 at 10:19 am #

      I too have a Brompton and read this with interest. Sadly there are no Rene Herse tires available for 16″ wheels, so I can only guess at how it would perform with nice supple tires on it. But I suspect that even with equivalent tires, the difference in rolling resistance over bumpy terrain would be measurable/noticeable. The big advantage (other than folding) is in stop-start city riding, where it takes much less energy to get a small wheel up to speed when accelerating from a red light.

      • Jan Heine July 17, 2020 at 11:05 am #

        There’s likely a minimum wheel size for any terrain – smaller gets significantly slower, but bigger sees comparatively small gains…

      • Stuart Fogg July 17, 2020 at 1:46 pm #

        A smaller diameter wheel will have to spin faster at given speed. AFAIK if all of the mass is at the edge (a pretty good approximation for a tire and rim) its effect on acceleration will be exactly twice the effect of the same mass not rotating, independent of diameter.

        • Jan Heine July 17, 2020 at 1:50 pm #

          Good point on acceleration. The physics are pretty simple: Wheel diameter doesn’t affect acceleration at all. The larger wheel has a greater rotational inertia, but also spins slower. So the only difference is in the weight – the larger wheel inevitably will be heavier.

      • Citybiker.at July 19, 2020 at 11:37 am #

        Try the Schwalbe one Brompton tires and tubolito tubes, feels great and faster than any other Brompton tires

  2. Ron Blumenthal July 17, 2020 at 5:52 am #

    “,,,,,, you’ll need any advantage you can get, even if it’s mostly psychological”

    Funny – after. years of delightfully Jan Heine inspired 650b and 700c wheels and conversations on rides on rough New England roads, and all 5 of us going back and forth between bikes and wheel sets, and customs and etc and so on. That’s the same conclusion we all came to. It’s mostly in the head (with the small and reasonable exception of bigger = less flats / more traction on potholed dirt roads / trails.

    Same thing with low trail / high trail.
    And all the riders in our little group…big supporters of BQ & Compass.

    Yet….it’s mostly all in the head? Guess so!

    • Jan Heine July 17, 2020 at 6:53 am #

      In many areas, there are real differences in speed and handling that are easy to document. Wheel size isn’t one of them. High- vs. low-trail is also an interesting one, because it’s non-linear, and high-trail bikes handle more like low-trail bikes than mid-trail bikes…

  3. Alex July 17, 2020 at 6:46 am #

    You keep using off-road race vehicles in your examples when drafting these articles in favour of one wheel size or another, as applied to bicycles. Yet you completely overlook (willingly or not) one important fact: the massive, gargantuan, amount of suspension travel of the former, versus the zero amount of suspension of the latter. Yes, I know about that goofy Cannondale, but that is an edge case.

    Off-road vehicles also have twice as many wheels, all-wheel drive, 2000 fold the power output (800 HP vs. 300 W) and travel at five fold the speed. The resulting physics cannot be glossed over.

    I get the point you are trying to make, but really, apples to oranges.

    • Jan Heine July 17, 2020 at 6:56 am #

      The first draft of the post mentioned the car’s huge suspension travel, but the post is already complex without it. I agree that to go fast on rough surfaces, bicycles should have suspension. That’s well-documented, just check out Bicycle Quarterly’s suspension loss tests. So yes, rather than worry about wheel size, racers should think about how much flex their forks offer. But that is a topic for another post…

      You don’t need huge travel, because you don’t go through dips and over crests at the same speed as race cars, and you’ve also got a lot less weight. But the goal is the same: Keep the main masses of the cyclist/race car from moving up and down on every bump reduces the suspension losses and increases speed.

  4. Frank July 17, 2020 at 8:08 am #

    I understand that you have tried to keep the argument as simple as possible for this post. If I may gloss, your point is that a bicycle wheel does not climb over (small) obstacles, the tyre absorbs them by deforming.

    Reality is a little more complex. The wheel will always climb a bit over any obtacle, as the deformation of the tyre leads to an upward impulse that necessarily accelerates the wheel and rider upwards to some extent.

    The question then is how that acceleration changes with wheel diameter. Assuming constant tyre width, the upwards impulse will develop more sharply the smaller the wheel diameter is (in mathematical terms, with the second derivative) but the total duration of the impulse will also be shorter for a smaller diameter wheel.

    So which wheel diameter leads to the least vertical disturbance of the bicycle and rider? My intuition suggests that the answer is “it depends”. A short, sharp obstacle might actually feel smoother under a smaller diameter wheel due to the vertical impulse lasting less time.

    This whole discussing of course applies to obstacles that are safely less proeminent than the tyre section is tall. Riding over tree roots and boulders will always favour the larger diameter wheel.

    • Jan Heine July 17, 2020 at 8:25 am #

      Absolutely. A tire is never perfectly elastic. In fact, the stiffer the tire, the more the wheel behaves like a solid wheel. but I didn’t want to add one more thing about supple tires to the post…

  5. Kate July 17, 2020 at 8:13 am #

    What Switchback Hill casing did Ted King use on his Vermont ride?

    • Jan Heine July 17, 2020 at 8:23 am #

      He used the Standard casing.

  6. Conrad July 17, 2020 at 9:22 am #

    I suspect this post will trigger a long debate… Competition improves the breed. We just need competitions that make sense. The revival of the technical trials is a wonderful idea. One thing I love about these gravel races is that racers are forced to be self sufficient and ride on varied surfaces. As a result, I am interested in the what the winners bike and tire choice looks like. As opposed to mainstream road racing which promotes a disposable gimmicky bike. If we went back to the old rules where no neutral support was allowed and you had to finish on the same bike you started on, the racing and the bikes would be far more interesting.

  7. JaBig July 17, 2020 at 10:36 am #

    Thank you for this long-awaited article. I understand the theory better!

  8. Tommy Patterson July 17, 2020 at 10:50 am #

    “So in theory, the wheel with the lowest rolling resistance is the tallest and widest you can fit into your frame”

    Yes — this seems to be confirmed by the trends in downhill mountain bike racing over the last few years. A lot of the top UCI downhill mountain bike racers have gone to 29″ wheels and their entire goal is to go as fast as possible over seriously rough terrain. That said, maybe there are also changes to bike geo (ht angle/ BB drop) that are operating in tandem with larger wheels to make these bikes faster.

  9. Paruig July 17, 2020 at 12:49 pm #

    When will you develop a 20″ tyre?
    There are lots of fast recumbents with 406 and 451 wheels.
    I changed from narrow high pressure tyres to 44-406 Schwalbe Shreddas running low pressure and got a huge increase in comfort with no apparent loss of speed.

    • Jan Heine July 17, 2020 at 1:49 pm #

      The problem with 20″ is that there are two standards. Each makes sense… but we don’t have the resources to serve both!

      • Paruig July 19, 2020 at 2:18 pm #

        I think 406 is far more common than 451 based on my interactions with recumbents over the last 18 years. BMX often use 406 too.
        Lots more 406 tyres are available than 451.

        • Jan Heine July 20, 2020 at 10:03 pm #

          Yes, 406 is far more common, but it seems that the BMX racers use 451…

  10. john hawrylak July 17, 2020 at 2:44 pm #

    Good article with good explanations. the 2nd half mainly discusses the advantages of 650B vs 700C rim size, especially on gravel. The way i understood is it is the 650B is better since it can use a wider tire, you have more tire than rim vs 700C.

    Can you extend the analysis down to 26″ rims?? Would a 26″ (559) rim with a wider 26″ tire perform better than a 650B rim with a tire width to give the same OD as the 26″ wheel???

    I apologize if this was asked already. Did not have time to read all the replies

    • Jan Heine July 18, 2020 at 8:22 am #

      Our testing shows no difference in speed with 26″ wheels either, for tires the same width – and thus (slightly) smaller wheel diameter. In the future, we may try testing the same wheel diameter with different rim sizes. Something like a comparison of 700C x 23, 650B x 42 and 26″ x 54 mm. The only problem is that we already know that 23 mm tires are slower… and we don’t have tires wider than 54 mm at the top end (with the same casing).

  11. Tobias July 17, 2020 at 9:07 pm #

    I have done long-haul touring on both 16″ (old Moulton F) and 700c. Both work well.

    Two days ago I rode to work on my recently restored 1984 HON, with 305mm 16″ wheels, and Schwalbe green guard tires. I would not want to ride those tires on a 700c bike, but at the smaller diameter they roll fast, and my 8 mile commute was barely slower than on a “full sized” bike. I think there is a possibility that the tire quality is less important the smaller the diameter since the contact patch is natively smaller.

    MY QUESTION to the blog is whether rolling resistance is the same:
    559x40c @40psi = 622x40c @40psi
    Or
    559x40c with 2″ contact patch = 622x40c with 2″ contact patch.
    And if the latter, is
    559×2.4″ = 584x50c = 622x40c if contact patch is 2″ across the board, adjusted my pressure?

    • Mark P. July 18, 2020 at 10:39 am #

      Your comment about tire quality not being important with small wheels is interesting to me because I’ve suspected the opposite: that adoption of small-wheel bikes is hampered by the relative lack of high-performance/high volume tires. I could be wrong about that, however.

  12. Chris. July 17, 2020 at 10:32 pm #

    I’ve been an adoptee of 650 and just managed to get 48mm’s on my randonneur — a Juniper Ridge on the front and a Switchback on the back. It was a very tight squeeze, but the guys at my LBS were great. I’ve been testing the new set up and loving it — mixing in a couple of long paved stretches to get to the gravel, where the set up is a blast. And the steep run down on pavement is super fast! The bigger tires are so much more confidence inspiring.

    My question then becomes why the Fleecer Ridge? Why did Lael want a bigger wheel? The volume on a 29er x 55mm is almost identical to a 650b x 2.25in. My sense is that it really is about Frank’s comment and Tommy’sas well — about the size of the rocks we’re rolling over. I like my 29er x 2.25 XC over the 650b x 48mm by a lot when I’m running down the crappy fire trails above my house. There are lots of sections that really push the definition of “gravel”. On the bigger stuff, the angle of attack seems to matter more.

    The matter of supple is settled for me. This issue now seems to be about what you’re rolling over. Over gravel, I can’t see why you’d need a 29er/700c, with the 650b options we have now.

    Thanks.

  13. Warren July 17, 2020 at 11:11 pm #

    Have you tested any really wide tires? 26×4” or 5”?

    The ones I’ve ridden (and just handled in stores) are really stiff and heavy, but I was wondering if you had any thoughts or experience with something like your prototype Rat Trap Pass tires . . .

    An Extralight casing fat bike tire is something I’d love to ride.

  14. Kai S July 18, 2020 at 1:56 am #

    slightly off topic, but as for the by Alex mentioned Brompton 32/349 tyres (at 1019am july 17) their older yellow and green label original tyres were supple enough. and had really excellent rolling. they however had an abruptly ending rather narrow tread, which made them sketchy when leaning over in high speed turns. this effect was later relieved when widening the tread, but unfortunately rolling then clearly suffered.

    by privately sanding away the too abruptly ending lateral edge of the central diamond pattern of the earlier version, turning got safe! by doing a similar more pronounced reduction of the too much added shpulder rubber of updated version, its rolling improved to similar levels. being in a peleton with 23-25 mm 622 racers you then are on level as long as on tarmac, which is a quite impressive performance for wheels this size.

    since a few years, production of these tyres has stopped, and have been replaced by schwalbe marathon racer. it does not roll quite as well, but nevertheless acceptable, and can it also be improved by reducing shoulder thickness of tread:) sides are supple so there is good potential. its speedgrip rubber compound is better in rain and lighter snow, and i use them for now wintertime only; for summer use still have a fair amount of the original yellow label tyres for years to come.

    for heavier winter use i have michelin country 44/305s. these also have a sub 60 durometer grippy rubber quality and blocky tread pattern, but with quite supple casing they roll well. and much better than the sluggish 349 marathon (not racer), or marathon plus.

    which is on par with the subject of 622 vs 584 in this article:)

  15. gran July 18, 2020 at 5:51 am #

    in motorized vehicles, another thing larger overall diameter gives you is more tread life / resistance to the rubber overheating. hopefully life gets good enough that we can start using our time to hyper analyze this phenomenon on bicycles 🙂

    • Jan Heine July 18, 2020 at 8:26 am #

      That’s an important point. Conversely, you might make the wheel diameter smaller to get the tires warmer to improve their grip. Motorcycle and car racers heat their tires before races, but they also generate enough side forces to keep them warm. Our bicycles, our tires are pretty much at the temperature of the road/air, no matter how hard we pedal and lean into turns.😉

  16. Erik H July 18, 2020 at 10:41 am #

    The best choice of wheel size is usually influenced by the design of the bike that you have. Often, bikes that work with both 650b and 700c can accept a wider tire with 650b than with 700c due to the chainstay shape. I have such a bike and I have two wheelsets for it: one with Herse 650×48 knobby and one with Herse 700×42 knobby. Each of these wheels is as large as practical for my frame. I prefer the 650b wheels since the extra 6 mm of tire width translates to an extra 30% volume! That’s a lot of extra cushioning for the rough stuff. Plus the speed benefit of the extra width is significant, where as the penalty for the slightly smaller diameter is not.

  17. George July 18, 2020 at 12:31 pm #

    In Germany and Holland, the most used wheel/tyre size used for touring bicycles is 622-50mm. I would love to be able to buy that size tyre for my touring bike from Rene Herse but that is in between the 44mm (which is in reality 41mm) and the too wide 54mm.
    Is there any chance René Herse will “do” a 622-50mm or a 622-48mm?

  18. Stuart Fogg July 18, 2020 at 4:42 pm #

    I’m sure this is a topic for another post, a very long post, but I’d like to know the ratio of tire rolling resistance to total resistance. I’m sure this is a function of speed as well as rider position, surface roughness, and a thousand other variables.

    • Jan Heine July 18, 2020 at 6:48 pm #

      We wrote that article for Bicycle Quarterly many years ago… What’s interesting is the potential for speed improvement. Improving aerodynamics is the biggest factor, but it’s also very hard to improve once you are riding in a semi-inclined position and wear tight-fitting clothes. Rolling resistance is the second-largest resistance for most cyclists, but it has by far the largest potential for improvement, since you can reduce it by 30% or more.