Tires: How Wide is too Wide?

track_tire_test

How wide a tire is too wide for optimum performance? Our research shows that wider tires don’t give up anything on smooth roads, and gain a significant advantage on rough roads. This has been shown for tires up to 31 mm wide.

It’s now a well-established fact that wider tires roll faster than narrow ones. Professional racers now use 25 mm tires, which are 20% wider than the tires that most racers used just 20 years ago. Will this trend continue? Can we expect racers to be on 30 mm tires in the future? No matter what the pros do – they are influenced by many factors that have little to do with science – the real question is: Up to what point are wider tires faster?

5208508437_58d18286d0

It is obvious that the tires in the photo above will not roll very fast. Clearly, at some point, the performance benefits of wider tires (shorter contact patch and thus smaller hysteretic losses; reduced suspension losses) will be outweighed by the disadvantages of extra weight and increased wind resistance.

Test_Woodland

In our original tire tests (above), we tested the same tires in 21, 23 and 25 mm widths on a moderately rough “backroad” surface. The results were clear: The 21 mm tires were slowest, 23 mm was in the middle, and 25 mm tires were fastest. The speed difference between 21 and 25 mm tires amounted to about 2.5%. Over a typical 200 km brevet, I would gain about 11 minutes. It’s not huge, but significant. These results appear to have prompted the current trend of racers using wider tires.

What about tires that are wider than 25 mm?

RumbleStrip

Our testing on rumble strips showed that on very rough surfaces (the equivalent of cobblestones), 42 mm tires are faster than 25 mm tires. However, few of us ride all the time on cobblestones, and what we want to know is whether we give up anything on smooth roads when riding wider tires.

To determine this, we tested Grand Bois tires in 26, 29 and 31 mm widths on a super-smooth asphalt surface (see photo at the top of the post). The results were the same for all three tires. On the smoothest asphalt, you don’t gain anything by going to tires wider than 25 mm, but you also don’t give up anything.

Those tests were run at 25 km/h (16 mph). At higher speeds, the aerodynamic disadvantages of wider tires might be greater. Does that mean that 31 mm tires are a fine choice for riding at moderate speeds, but that you would be better off on 25 mm tires when you go faster?

drafting

We tested both 25 and 31 mm-wide tires in the wind tunnel. The result: The raw data showed a 1% increase in wind resistance for the wider tires, but the results weren’t statistically significant. Even if we accept them at face value, the added wind resistance is too small to make a noticeable difference. For example, at a very high speed of 40 km/h, decreasing your wind resistance by 1% only adds 0.4% (or 0.14 km/h) to your speed.

What about the heavier weight of wider tires blunting the acceleration of your bike? That doesn’t appear to be a major factor either, since wheel weight is less important than many riders believe. (See this recent post for a discussion of wheel weight on professional racers’ bikes.) If you use smaller 650B wheels, you make up some of the greater weight of a wider tire through a lighter rim.

All this data shows that 31 mm tires roll as fast as 25 mm tires, even on very smooth roads. And when the roads get rougher, the wider tires roll faster.

rando_ti

What about even wider tires? Our on-the-road experience suggests that even 42 mm-wide tires do not roll slower than 25 mm tires (above), but without rigorous testing under controlled conditions, we can not say for sure. We hope to test this soon.

Of course, there are other reasons beyond performance to ride wider tires. You gain comfort. You will incur fewer flats, since you run wider tires at lower pressures, so they roll over obstacles that would get hammered into narrower tires. You’ll be safer, since a wider tire will be less affected by small cracks and railroad tracks.

Most of all, you’ll be enticed to go on small roads that have great scenery and little traffic – roads you might have avoided with narrow tires because the pavement tends to be rough. With more comfortable tires, you can even enjoy roads with no pavement at all!

To answer the question in the headline: Even 42 mm does not yet appear to be “too wide.” Tires wider than that are hard to fit into the rear triangle of a bike without compromising performance (tread/Q factor, chainstay length), so perhaps frame design more than other factors limit the maximum tire width on a performance bike.

Wide tires are one of the few things with a lot of advantages, but very few disadvantages. (There are some downsides to wide tires, which we’ve mentioned here.)

For all our tests, we used tires that had the same casing material, tread pattern, etc., to isolate the effects of tire width. Of course, there are many other factors that influence tire performance, and width is only one important factor.

This post is just a summary of the research. The original data and much more detail were published in Bicycle Quarterly. Here are a few resources for further reading:

About Jan Heine, Editor, Bicycle Quarterly

I love cycling and bicycles, especially those that take us off the beaten path. I edit Bicycle Quarterly magazine, and occasionally write for other publications. One of our companies, Bicycle Quarterly Press publishes cycling books, while Compass Bicycles Ltd. makes and distributes high-quality bicycle components for real-world riders.
This entry was posted in Testing and Tech, Tires. Bookmark the permalink.

48 Responses to Tires: How Wide is too Wide?

  1. robertkerner says:

    This blog is the main reason I ride wider tires than I did 5 years ago, and it was the driving force behind me trying the Grand Bois tires. They are among the finest tires made, I’m sure. Imagine riding a couch cushion; they are that nice! Be prepared, though, for good natured heckling since may riders are still under the false assumption that wider tires make you go slower. Since trying the Grand Bois, I will not ride anything narrower than 25. 28 seems to be the sweet spot for me, with the 31mm Grand Bois being the widest I’d go for regular riding (I have a set of 37mm snow tires on my commuter).

    Since stepping up to wider tires I’ve noticed something that you mention toward the end of the post: the bike is noticeably more stable over bumps and cracks with the wider tires. That, in turn, allows me to ride with greater comfort and confidence that hitting a small hole will not send me flying or blow out the tire.

    If you’re going to ride wider tires, inflate them sensibly. You don’t need to pump that Grand Bois up to 110psi! Happy New Year.

  2. Andy says:

    Any tips for how to be comfortable on wider tires? I commuted, raced, and rode my new steel frame bike a few thousand miles with 28-32mm tires this year, but struggled with finding a tire pressure that worked for me. At the pressures BQ recommended, cornering at speed was uncomfortable unless I had them around 90psi. But at that point, I’m probably losing some of the benefits of having a wider tire that should be run much lower. For general commuting, I was fine down to about 35-40psi where it starts to feel more bouncy and with a higher chance of pinch flats on potholes.

    I’ve tried tires as wide as 35mm, but have never had the feeling that Jan and others describe as though a road has been smoothly repaved. At a wide range of tire pressures, I still feel every bump rather similarly unless I’m on my mountain bike with suspension and 2.2″ tires. It would be interesting to find a test that puts a measurable number to this, such as how much tire deflects or bounce (of something on the bike?) occurs from hitting the same bump or rough road on various tires. That would be of more interest to me than finding out how many tenths of a percent I could be faster on my 30 pound bike.

    • I’ve tried tires as wide as 35mm, but have never had the feeling that Jan and others describe as though a road has been smoothly repaved.

      Tire pressure and the feel of a tire depends very much on the tire casing. If you ride a Schwalbe Marathon, it always will feel buzzy, no matter how wide the tire. Back in the 1880s when they rode solid tires, tire width didn’t matter at all for comfort – if the tire doesn’t flex much, it doesn’t matter how wide it is.

      At the pressures BQ recommended, cornering at speed was uncomfortable unless I had them around 90psi.

      As to the cornering, you obviously don’t want the tire to collapse under the cornering forces. This, again, depends on the sidewalls, tires with more supple sidewalls need slightly higher pressures than stiff-walled sidewalls. Experiment and see what you find.

      Each tire is so different that it’s hard to generalize, and guidelines for tire pressure like these are only starting points for your own experimentation.

  3. Joe Ramey says:

    My take away from this article is, go as wide as you want, and never go skinnier than 25mm (27mm for me).

  4. AndrewGills says:

    Wow! So 23mm tyres are slower than 25s? I never knew that. I ride 23s because they are most readily available at a good price. 25s are almost always $5-10 more expensive. I don’t ride for speed but am fascinated by these findings. Once I finally get around to building my steel Audax bike it’s going to have wider tyres.

  5. bostonbybike says:

    To be honest, I think that in order to do this kind of testing right you should build a test robot where you eliminate any external factors – run a test in a wind tunnel, controlled environment, with no rider but an electric motor of known output powering the wheel, etc. Only then the results may be considered real. When you are seeing changes of ~1% – that’s ridiculously low to claim any bold statements that wider is better. It would be good to see raw data where human perception of riding comfort is not a factor, i.e. no rider in test the next time.

    • The rider is crucial in the whole setup. Without a rider, you don’t get suspension losses that are absorbed as heat in the rider’s body; you don’t get the legs churning the wind… Testing without a rider would be meaningless.

      To distinguish noise from true results, we always do a rigorous statistical analysis. That shows you whether the variability within the dataset is greater than the differences you are seeing between different setups (like 25 vs. 31 mm tires).

      The 1% difference was that wider tires are slower. I only mentioned it because it runs counter the argument I am making about wider tires being faster. Statistically, it wasn’t significant, which may indicate that it’s just noise in the data, or it may be too small to be measurable with our testing approach. The results that 25 mm tires roll faster than 21 mm tires and 23 mm tires was highly statistically significant.

      The raw data was published in the Bicycle Quarterly articles that are mentioned under “Further Reading.”

      • Scott Urban says:

        2.5% is also in the noise.

      • If you are trying to say that among all the factors, 2.5% doesn’t matter much, then that depends on your viewpoint. When I rode Paris-Brest-Paris in 50 hours, going 2.5% faster would have meant finishing in under 49 hours. To me, that would have been a very significant difference. Beyond that, small improvements add up.

        If you use “noise” in the sense its scientific definition, then you have to look at the statistical analysis of the data. And in the case of the wider tires, the results were statistically significant – and not “in the noise.”

  6. Vincent B says:

    Wat are your thoughts about the width of the rim?
    I found out that a 32mm tyre on a 13mm rim is quite ‘wobbly’ in the corners, and that 37mm tyres on a 17mm rim are fine (both tyres Vitoria voyager hyper). How wide are your rims for the 42mm tyres?

    • We haven’t found a huge correlation between rim width and cornering feel, but 13 mm (I assume inside dimension) is quite narrow for a 32 mm tires. Most 650B rims are 23 mm wide (outside), or about 18 mm inside.

      What does change significantly with rim width is the width of the tire. On a 23 mm rim (outside), a Grand Bois Cypres measures 32-33 mm wide, on a 20 mm rim, it’s only 31 mm wide.

  7. nelfilcon says:

    Any chance of a review for the Velocity Atlas rim? The arrival of 650B MTBs has created many strong v-section rims without a brake track (for disc-only frames). Velocity A23 recently received a favorable review in BQ, but there still seems to be a hole in the market for 650B rims suitable for touring, tandem, and clydesdale use with rim brakes.

    There are many fine rims catering to this market in the 700C size

    • Chad says:

      I’ve had the atlas rim since it came out, probably one of the first ones as it wasn’t even on their website yet and the race team I’m on is supported by velocity. The rims are wide and incredibly durable. They’re also heavy. I have them in the polished finish and they look stellar. Braking is solid even with my tektro canti’s on my space horse that I use for commuting/errands/whatever I want a durable bike for.

      I have the images of the wheels I have, built up by velocity with their hubs, on my website cyclingandsuffering.com. I’m not sure that I can post a direct weblink in the comments due to spam filters many sites have but you can run it through the search on my blog.

  8. Bubba says:

    “wider tires blunting the acceleration of your bike? That doesn’t appear to be a major factor either, since wheel weight is less important than many riders believe.”

    Do you have any ideas to prove this assertion with testing? This one is still really common on the various forums. People will post “I concede that wide tires can roll really nicely, but climbing and sprinting, it’s no contest, lighter is better”. Just an article about the calculation of momentum would probably be a great start. It seems like it would be impossible to do an objective “sprint test” empirically.

    It seems that the thing that “many riders believe” is that light wheels “feel” faster to spin up. They might reply: “I don’t need your math, or your testing! My legs tell me that lighter wheels sprint faster”. Maybe it’s good enough for many to feel faster without actually being any faster. But people don’t like being told their senses are deceiving them.

    • A placebo effect is very powerful. I know that when I raced, I loved riding my bike with its superlight wheels and hand-glued tires. How much was the wheels, how much the tires, and how much the fact that I rode the race wheels in races, when I had rested before, was excited and generally in optimum shape?

      Jim Papadopoulos developed a spreadsheet that calculates bicycle acceleration, so it should be easy to get some realistic values for power output, wind and rolling resistance and figure out how much faster lighter wheels accelerate.

      In the mean time, the stories of professional racers trying and rejecting bikes with lighter wheels, like Saronni in the previous post, might be powerful anecdotal evidence. After all, the easiest way to make wheels lighter is to make them smaller. The current minimum wheel size is 550 mm, and a 700C wheel is more than 100 mm larger than the rules require. If it matters as much as most people believe, why hasn’t anybody taken advantage of this, and why have those who tried returned to standard wheels?

  9. Chris Bonner says:

    It seems you don’t allow comments on previous entries. I was wondering, in the third photograph in the post “Who are you calling fast and furious?,” (this one http://janheine.wordpress.com/2012/12/15/a-typical-ride/13_holmes_point_descent/ ) are you perchance riding on the road from Carbonado to Fairfax?

  10. Mark Eastman says:

    I run 700cx23 tires on a vintage steel bike I use as a lightweight commuter, but otherwise have favored wider tires in the last few years for most all of my other riding. In 700c the lower end/mid range tires may not accelerate as quickly as higher quality tires, but wider tires are generally more sure-footed, offer more comfort and versatility on a wider range of road surfaces. 650Bx42 tires in particular are amazing if you are used to 700c.

  11. Marc B. says:

    No 42s in the wind tunnel, really? If you ask the question, “how wide is too wide?” it seems to me you have a responsibility to test up until you reach that point. While there is plenty of good info in the post above, you certainly didn’t answer the question and that is REALLY FRUSTRATING.

    • In the wind tunnel, we had to use tires that had the same tread pattern and wheel size, since that could influence the wind resistance. Also, you want roughly the same relationship to the fork crown and blades, since a narrow tire in a very wide fork might have different air flow than the same tire in a narrow fork. Ideally, we’d have two bikes made that were identical in every way, except one had the tires and fork clearances scaled up for the wider tire. Wind tunnel testing already was a huge effort (wind tunnel time usually costs $ 500/hour, and we spent 20 hours in the wind tunnel), and this was not possible.

      The data we have already shows clear trends: If the difference in wind resistance between 25 and 31 mm is so small that it is not statistically significant, it’s unlikely that the difference between 31 and 42 mm is going to be huge.

      Our on-the-road tests, whether rolldown or on the track with a power meter, test all resistances, so they include the wind resistance (as well as rolling resistance, suspension losses, hysteretic resistance due to pedaling thrusts, etc.) These tests have shown that a 24 mm tire and a 35 mm have the same resistance on very smooth pavement (which is where wide tires have the smallest advantage). On rougher pavement, the advantages of wider tires are greater, and the 35 mm would be faster than the 24 mm tire.

  12. Rickonabike says:

    Im a middle-aged touring and casual rider. On my first cross-country tour in 2011, 5000 miles, Yorktown–SF–Portland, I rode an ’81 Fuji with 29s, then upgraded to 32s in Kansas. The rest of the ride was far more comfortable. In 2012 I rode a 9000 mile all over the US tour–Portland–Death Valley–Mt Rushmore–Gulf Coast–Atlantic Coast–then arced back down to New Orleans from PA, mostly on roads that were not typical tour routes from Adventure Cycling or USBR. I rode 35s to CO, then dropped back down to 32s. I ride heavy–a big steel KOGA with a usual load of 65-75#. 32s seem to give me the right balance of traction and road feel. Instead nof a long tour this year, my girlfriend and I rode all the Oregon Scenic Bikeways, plus the coast, taking weekends and short vacations for 2-3 day excursions with typical loads of 20# or less. For these and my everyday riding in the Portland area, I switched to 29s. I can ride them at a higher PSI and get better road feel. As for any differences in speed, I can’t say I notice any. I ride almost exclusively on paved road or the occasional crushed, packed granite rail-trail, but the only thing I ever ‘race’ is the sunset. I think I speak for a lot of casual and touring riders out there who are far more interested in feel and ride quality than speed and handling characteristics above 15mph.

  13. Tom ferris says:

    If I could only jam my Surly Bud & lou tires in my Colnago I’d be set!

  14. Garth says:

    I’ve been following “fat tire” bicycle evolution on the Gypsy By Trade blog. These tires are 3″ and up and are for super bumpy off road and snow. An interesting development in the mountain biking world has been the massive shift to 700c, which boggles me.

    • It’s hard to see how 3″-wide tires (75 mm) could be as fast on the road as moderately wide tires. That said, I once rode a fatbike for a few miles, and I was surprised how well it rolled. And of course, fatbikes allow you to ride in places where other bikes cannot be ridden at all, period.

  15. Jordan says:

    It seems like the “research” doesn’t take gravity into account, which is kind of important. There is a lot of flat road riding here but I think most of us agree that the difference between tires on the straight and flat is minimal. However, what about climbing? I would be shocked to hear that a 42m tire climbed as well as a 31m or that the 31m was as efficient as the 23c. Rotational weight is a much stronger factor than the stationary weight on a bike (saddle, handlebars). I believe that the watts required to spin a squishy 42c tire up a 7% grade would be significantly higher than if one was riding 23’s.
    I agree with the overall theme of your articles: Don’t worry about riding a wider tire, they are awesome. But at some point there has to be more acknowledgment of the varied terrain that we generally ride over.

    • Your point about gravity is well taken. We’ve thought about this quite a bit.

      The added weight is a factor, but a very small one. A Grand Bois Extra Leger 42 mm tire weighs just 357 g. The difference to a narrow tire isn’t huge – about the same as a couple of sips of water from your water bottle.

      Whether rotational weight is important when climbing at constant speed is debatable. Basic physics says it isn’t, but of course, our power strokes are nowhere near even…

      We’ve actually tested one of the very best racing bikes (with 25 mm) tires against one of the very best randonneur bikes (with 42 mm tires). For us, both climbed at exactly the same speed. The randonneur bike must have required a few more Watts, but the biomechanically more efficient frame must have made up for that. The randonneur bike allowed both testers to put out just a little more power on the randonneur bike than on the racing bike. A summary of that test is here.

      And just from personal experience, my fastest times in brevets (on the same courses) have been on my randonneur bike with 42 mm tires. Again, the wide tires may take a tiny fraction more power, but any rider’s power output isn’t a constant, and the bike is one of the many factors that influence how much power you can put out.

    • The quotation marks around “research” make it look like you are also questioning the credentials and expertise of the people behind our research. As you can read here, Bicycle Quarterly contributors have a lot more Ph.D.’s than the average bike magazine. If there is one thing we know to do, it’s design a scientific study…

  16. shastatour says:

    When buying my last bicycle having the capability to run a minimum of 28mm but really wanted to have 32’s eliminated about 95% of the bicycles currently being sold. Most bikes in the > $1K price range where you can get at least a 105 gruppo or equivalent will accept at most a 25mm and that is a stretch on many of them. I was really frustrated and ended up getting a custom made IF Club Racer which has been a wonderful bike though about double the cost of what I wanted to spend and is a heavier bike. I don’t understand why the large manufacturers of bike frames are so behind the curve on this.

  17. Don Genovese says:

    I agree that 31mm tires are more comfortable than 22mm tires but that’s not to say that 22’s are uncomfortable. A couple of my bikes have 28mm, one has 35mm, two have 22mm tubulars and two have 25mm tubulars. All of my riding is on smooth roads and I enjoy 22mm as much as 35mm. Further, if tubulars are glued properly they are safer than clinchers. Tubulars don’t blow off the rim at a puncture as do clinchers. I’ve ridden a flatted tubular for over a mile. Don’t think that you can do that with a clincher. My two cents worth.

    • I agree that tire construction is just as important for comfort (and more important for speed) than width. A high-end 24 mm tire will be more comfortable than a sturdy 35 mm tire with puncture-proof belts and a stiff casing. And as you say, tubulars offer even more comfort. I used to ride tubulars, until I had three flats in a cross-state race, and only two spare tubulars. I did finish, but that is a story for another day…

  18. David T. says:

    If you go from a narrow rim to a wider rim, for example from an aerohead to an A-23, and use the same tire, the measured width of the tire gets wider. What effect does this have on the performance and feel of the tire? Can you use a lower pressure in the second case, or would you keep the same pressure?

    • We haven’t tested that, since we tried to keep all variables except tire width the same. However, I don’t see any reason why a tire that gets wider because of the wider rim would not offer the performance benefits of one that is wider from the get-go.

  19. David T. says:

    I am going to do this. I am using 700c Cypres tires on aerohead rims. I ride up to about 20% of the time on gravel roads. Some of it is fine, where there is a tire track worn down to the hard dirt. Other sections with newly-laid stones are not comfortable, so I would like a little wider tire. The aerohead rims are obviously designed for a narrower tire.

    I think we should acknowledge also that riding on rough roads that are uncomfortable, with the handlebars jarring and the constant effort to keep the wheels going straight, is inherently inefficient and slow for a human rider, and also mentally exhausting.

    • riding on rough roads that are uncomfortable, with the handlebars jarring, is inherently inefficient and slow

      You are right about that. The vibrations that make it uncomfortable also cause the suspension losses. A study by the U.S. Army on vibrating tank seats found that the energy absorption in human bodies was directly correlated with discomfort. So the discomfort not only wear you out, but it also is a sign of direct energy loss. Even if you are super-tough, you’ll be slower on a bike that vibrates a lot.

  20. Martin Deschamps says:

    Love the article! My question would be how much the weight of the rider would influence the air pressure of the tires? I range in the 215-220lbs rider and always struggle with tire pressure for maximum performance, in both cyclocross and time trial.

  21. alliwant says:

    Thanks for advancing some themes that favor more comfortable biking, such as wider tires and efficient geometry. I also have some brevets under my belt, and know how important it is to be comfortable; feeling beat up results in many a DNF, in my rather brief experience. One area I would suggest for a little more emphasis is crankarm length. It turns out that the standard lengths appear to have been arbitrarily chosen. Might be worth a little inquiry.

  22. Ron says:

    Jan,
    I’m currently training on some “hard” 23mm wide tires, and after reading your articles about tires have considered buying some of the Grand Bois 26mm tires. But I’m seeing something in two of your articles that have me a little confused.

    In this article, you mention that better tires would shave 11 minutes off a 200km time. That is not really a lot. However, in this article:
    http://janheine.wordpress.com/2013/03/04/getting-your-bike-ready-for-randonneuring/
    You make this statement in regards to tires: “During a 300 km brevet, this would take more than 2 hours off your time”. Now that is a HUGE difference, doing the math it looks like you’d shave well over an hour during a 200km brevet which is more significant.

    I’m not trying to point out a “you’re wrong” type of discrepancy, just trying to figure out roughly the value of changing tires. What am I missing?

    (If it helps, I’m on an older Trek 5200 riding Continental Gatorskins…. a tire which I seem to recall you saying as one of the slowest you tested. A FOLLOW ON QUESTION: I know little about rim differences…. would I have a problem mounting the 26mm Grand Bois tires on the stock Rolf Vector Comp 700c rims?)

    • The 11 minutes refers to using the same tires, but in a wider version. The 2 hours are the difference between using different tires, one sturdy and slow, the other supple and fast. As you can see, tire construction plays a much bigger role than width by itself. But once you have chosen your tires, using wider ones will get you further optimization.

      On a different note, 11 minutes in a 200 km brevet is a large difference for me. It’s the difference between having a good day and a bad one…

      As to the 2-hour difference between a fast and a slow tire during an 11-hour brevet, that is huge.

    • Rims can handle a variety of tire widths. A 26 mm Grand Bois tire will fit on almost every 700C clincher rim that is available today.

      Once you start mounting 42 mm tires on ultra-narrow rims, it may becomes problematic, but even there, mountain bikers did that for a while with few ill effects.

  23. Justin says:

    This is silly. There is no way you measured a statistically significant 2.5% change with a human rider in outdoor conditions. Human performance factors and wind would probably account for a 5+% performance difference even if the the tires were’t changed between test.

    This is the very definition of pseudo-science.

    • I understand your concern, since introducing a rider and testing outdoors increases the noise in the testing. We worked carefully to control that, by testing only on windless days, with constant temperature, and with a rider wearing the same clothing and riding in the same position. We tested each tire in at least three runs. We also tested the same tires at the beginning and the end of each test session to see how reproducible our results were. The difference between runs of the same tires were much smaller than the 5% you suspect.

      Then we took the results and did two statistical tests (T test and Tukey’s HSD) to see whether the differences between tires were statistically significant. Some differences between tires with similar performance were not significant, but the noise in the tire pressure tests was small enough that the results were statistically significant.

      When we later tested a bike and rider in the wind tunnel, the results confirmed that an experienced rider can assume the same position. We tested the same setup in subsequent runs and found that the wind resistance was within 0.37% for each run. Clearly, a 2.5% difference in tire resistance will show up against that measurement error, since it’s an order of magnitude bigger than the noise.

      Of course, you could suspect that we made up the data. Fortunately, for the tire tests, we invited a number of witnesses. For the wind tunnel tests, they were performed at the University of Washington wind tunnel, so we didn’t even generate the measurements. Since the UW is a public institution, the results are part of the public record. If you like, you can go there and check.

      So in conclusion, it may be premature to call the results “silly” and “pseudo-science.” The two lead authors of the study have Ph.D.’s, one of them in statistics, so before you call their scientific credibility into question, you should at the very least look at the study. (It’s in Bicycle Quarterly Vol. 5, No. 1, with the statistical analysis published in Vol. 5, No. 3. The wind tunnel testing was published in Vol. 6, No. 1.)

    • I think a healthy dose of skepticism is a good thing. After all, if we hadn’t done the multiple runs of the same tires, nor the statistical analysis, we still could have published the results.

      There was a wind tunnel test in Bicycling magazine many years ago that showed all kinds of weird things, but I later learned that the Texas A&M wind tunnel where they tested had huge problems with changing temperatures, for which they didn’t correct. They did not do multiple runs of the same setup, and it appears that most of the results were just noise in the data.

      That is why we made sure that we were seeing real differences in tire performance, and not just noise in the data.

Comments are closed.