[Closed] What size wheel? Hopefully this will end it.......

[img] [/img]

I'm fairly sure that nobody cares...

£100k of kit to prove 29ers are faster?

Where is part2? I'm interested to see the results

Part 2 out next Friday.

I'm looking forward to hearing the results. Fantastic to finally see some real science being applied to this question. Good on them.

Tick
as I assume someone will post part 2 here as well.

I am with you bantasanta, don't really care, as long as you out on a bike and enjoying it. BUT with "wheel size" treads appearing every other day & 26" is dead etc, thought it be interest to others, more so as STW massive like to see lots of data 😉

So what do you think the results be?
29" the fasted by just a few sec over the whole lap, but 5 times the energy used.

Oh also did they take in rider height? I think that is a big factor as well.
I know they weighed them.....

Results were on twitter now I don't know if I can mention them without it being a spoiler! To me the results tend to indicate a problem with the test but maybe they will discuss that in more detail.

Can you post a link to the results akak?

If this link works see the replies... https://twitter.com/Howard_Hurst/status/558694857783640065

I tend to agree with your analysis of the results akak (that their results tend to indicate a problem with the test)

Thanks akak - see what you mean about the results!

27.5 the slowest, test must be wrong.
Lmfao.

I would have told you 29ers are faster for a bargain £25k....

It'd be interesting to know how used the riders were to the different bikes. I know that switching between bikes takes me a while to get used to the quirks each bike has.

Great result.27.5 is the slowest.
All my bikes are 26 so my bikes are better than all your fashion conscious wheel geek type wheel things... 😉

What about fatbikes? Would also be interested to see the geometry changes and how santa cruz accounted for that.

I dunno about fatbikes, but I wanna see a result with fatter riders, them guys is a wee bit trimmer than me 😉

I have bikes in all 3 wheel sizes I m all of a muddle now !

Nice, I'll offer them £450 for that 26" Santa Cruz when they're done, being obsolete an all...

chip

27.5 the slowest, test must be wrong.

Or 27.5 is a sucky compromise that is neither one thing nor the other?

😆

[img] [/img]

[img] http://media.wtfunny.com/content/pictures/c6b12025e981920fd05241ddb86f34f7jpg [/img]

Interesting but let's face it, it's just about impossible to test this scientifically. There are so many variables and you can't blind people to the wheel size they're riding. Luckily, it's been proven that MTBers have no psychological bias or emotional attachment when it comes to wheel sizes. 😉

Cheezpleez - Member

Interesting but let's face it, it's just about impossible to test this scientifically. There are so many variables and you can't blind people to the wheel size they're riding. Luckily, it's been proven that MTBers have no psychological bias or emotional attachment when it comes to wheel sizes.

Did you not see the part where they have covered the cyclist with sensors? They can measure the exact effort used, and a massive amount of other variables. You saying "impossible to test this scientifically" doesn't make it so... they ARE testing it scientifically. The rider knowing which bike they are on doesn't matter in this test, they are not just timing a loop, they are measuring effort as well, and presumably using maths and science and stuff to reach a scientific conclusion.

Now, once the results are published in full, you would be welcome to read through them and questions the results scientifically.

Now, once the results are published in full, you would be welcome to read through them and questions the results scientifically.

A week,That's if the scientist in question doesn't find himself "sleeping with the fishes" first. 😀

Please can someone put a gif up of a dog chasing its own tail?

I think that is a suitable metaphor for wheel size threads on here, and I'm sick of the cats/kittens getting all the modeling work as a by product.

As it's a wheel size topic it has to be a couple of kittens...

[img] [/img]

[img] [/img]

Maybe this might prove that the bike companies have very good marketing departments and sell/force us to new wheel sizes. I have 26 and 29 bikes. I like the very different feel of both. Tried some 27.5 and to be honest they were nothing better than my 26 enduro could offer, marginal improvements at best. Certainly not worth the outlay in my opinion. all three sizes have been around for a while, why the big thing now? Yeah manufacturing improvements have helped but not to the point we need to bin a while size!

There is a place for all sizes of wheel as variety is a good thing but not at the expense of another IMHO.

So, we had bikes that were fast and flighty downhill, and also bikes that were super fast cross country race machines… and we're steadily replacing both with bikes that are somewhere inbetween. Very surprised that the inbetweeners did worse than the little wheels at Clayton Vale though, would expect them to much the same, it's not challenging or rocky terrain, and the downhill sections are pretty short, don't see how either would have an advantage over the other really… unless the little wheels they used were lighter.

The rider knowing which bike they are on doesn't matter in this test,

The best experimental methodology involves the person not knowing what the experiment is about nor what group they are in.

This has neither of these controls and experiments consistently show this does matter.

Given this, and scientist do double blind experiments, could you explain your reasoning please?

I have 26 and 29 bikes. I like the very different feel of both.

Agreed, different enough to be well worth the hassle of 2 sizes of spares and stock etc for shops and riders.

There is a place for all sizes of wheel as variety is a good thing but not at the expense of another IMHO.

If only…

I'd say cheezpleez is right. Surely the only science in all this is the difference in rolling resistance and wheel weight and you don't need riders to test that. The rest is noise created by handling, rider preferences, rider input/feedback etc as well as terrain.

You just can't say 'this wheel size is faster'. You could say 'I prefer this wheel size' and that's cool, we all have opinions and go faster on bikes we like.

Junkyard

The best experimental methodology involves the person not knowing what the experiment is about nor what group they are in.

This has neither of these controls and experiments consistently show this does matter.

Given this, and scientist do double blind experiments, could you explain your reasoning please?

How do you expect to do a blind test of riding different bikes?

The measurements taken of the muscles, oxygen / CO2, power output at the wheels will all go to measuring if the subject put in more effort on a certain bike, and can therefore measure that and account for it.

Do it with 50 different riders over 50 different locations on 50 days over 12 months and then you'll have some data, even going to the lengths they've gone to isn't enough actual data to draw any conclusions.

My gut feel based on all the various bikes I've ridden is still that 29 is faster for general XC, and I say that as someone with an entirely 26 bike collection and no intention of changing in the near future, but gut feel isn't science.

Gah!

So really
Which Wheel size is faster over this course in a XC discipline for XC racing whippets.

Which is a bit more specific than which wheel size is better this will settle it.

The data pool is small, the range is small and the results are specific to a certain type of riding.

In a nutshell 29er was faster buy about 14 s over 4km. Strangely 27.5 was slowest.

On that course, there is no where that you'd expect any difference/advantage between 26/27.5, but if the 26 wheels/tyres were a good chunk lighter, then their results might make sense. I have to agree that their results are statistically irrelevant though. Then again, no worse than the “science” Giant threw at us last year…

[img] [/img]

I have to agree that their results are statistically irrelevant though.

Meh, handwaving. Only a proper critique of the results will establish an argument for relevancy or not.

Meh, handwaving. Only a proper critique of the results will establish an argument for relevancy or not.

Must be a very strange track. If 14s in 3.4km was representative of normal xc conditions it would mean no one on a 27.5 would ever win a world cup.

How do you expect to do a blind test of riding different bikes?

Would be interesting to see what they "normally" rode

Dont get me wrong its better than any debate but I don think it will end the debate

My take 29 er faster for XC racing
26 er more fun

I have ridden there and I dont think anyone climb does more than 30 - 50 metres [ height if that] and its undulating at best My kids cleaned it all at 7 on a 20 inch bike for example. I would like to see the breakdown though BUT larger trials across all ages and abilities are needed for anything conclusive. The best it will show which bike is best there for athletes.

Rode/ride there since day one as its around the corner and good for a quick.

Was faster around it on my 26er vs my current 29er. But on one of my loops I do I'm faster on the 29er. One test location is never going to be adequate IMO. I'll soon see whether my fashion conscious 27.5 is finished and see if that's the slowest of the bunch, not that it will matter as I'll still be grinning.

A few points:

First, the results haven't been published yet and we are reacting to one tweet. The first rule of any research study is "check who funded it". Santa Cruz may not have interfered with the study in any way, but if their lawyers are on the ball they'll have a clause allowing them to block publication of any results that could be detrimental to their business.

Second, they didn't test wheels (of course) they tested bikes. Getting the geometry just right for each wheelsize takes time. If the 650B is slowest then it could just mean that Santa Cruz haven't quite sorted the 650B geometry yet.

Of course a double blinded study is the ideal, but if you can't do one then you have to do what you can and just take the limitations of the study into account when interpreting the results. Sticking a needle in your eye isn't the best way to probe the function of the optic nerve, but it did give Newton a few useful clues.

Finally, the second rule of research is "check where the results are published". Have they appeared in a well respected journal having gone through a rigorous peer review process or are they in a YouTube video. I doubt the guy doing it really thinks this is science, but it's a bit of fun

As many of us have been saying all along, there are clear benefits for different types of riding in choosing 26" or 29" wheels, and they are different enough to happily co-exist, but choosing 650B as a middle wheelsize doesn't necessarily give you the best of both worlds, it may in fact give you the all the drawbacks of the other two wheelsizes (i.e. slower to accelerate due to the extra weight over 26" but without the extra rollover and stability of 29"). It sounds like this study might back-up that theory.

Caveat: I have never ridden a 650B but own 26", 29" and fat wheeled bikes so their may be a hint of bias. Other advice and points of view are available 😉

I'm going to suggest that the issue with all these studies, is that for the "average" case, the effect of wheelsize is insignificant (Statistically) compared to the other "noise" factors in the test.

In extremis, wheelsize clearly makes a difference, with a 29er being unsuitable to WC downhill racings, and a 26er being slower over 10 day adventure marathon course etc. But in the middle, the sort of riding most people do, the differences are tiny. Measurable, but tiny, and easily swamped by a million other factors.

For these tests, the only currently quoted figure is "14sec in 3.8km", which, assuming an average speed of say 20kph is 14sec in 684sec, which is 2% faster. To be 95% confident of a 2% improvement, we have to know the standard deviation of the results over enough runs to form that confidence. As yet, we don't know if that was done for these tests.

And, even if it was, and 2% is the answer, for You, or I, or Joe Blogs, if you're not actually RACING, who cares? 2% is nothing!

And, even if it was, and 2% is the answer, for You, or I, or Joe Blogs, if you're not actually RACING, who cares? 2% is nothing!

So we should all ride fat bikes and have more fun 😀

Listen guys if you want the definitive facts ignore the so called boffins. These are my findings.

650b felt the same as 26.
29 was well faster, and therefore more fun.

My study was self-funded, and hereby published in STW the only MTB publication of any import.

I suspect its about to be strenuously peer-reviewed.

a 29er being unsuitable to WC downhill racings

Nonsense. Wheel tech has moved on to the point where strong enough 29 wheels are possible, for sure, and that was probably the only “real” reason not to have DH 29ers. That and market segmentation/image. And tyre choice. The same reasons inbetweener DH bikes took a while to happen all hold true. Doesn't mean 29er DH bikes couldn't be made to work. But why bother? Adding extra wheelsizes adds nothing to the spectacle or competition really.

650b felt the same as 26.
29 was well faster, and therefore more fun.

Agreed on all but the fun… I'm sure you can make a bike overloaded with the ability to be fun with any wheelsize… and that's about more than going fast, for most if us, I'd hope.

So we should all ride fat bikes and have more fun

They're not laughing with you, they are laughing at you

Finally, the second rule of research is "check where the results are published". Have they appeared in a well respected journal having gone through a rigorous peer review process or are they in a YouTube video. I doubt the guy doing it really thinks this is science, but it's a bit of fun

From Dr Hurst
"We are hoping to get the findings published in the Journal of Sports Science and the manuscripts are currently in the final stage of review, so hopefully they will be out in the next few months."

This is actually quite farcical, embarrassingly so. The 'research' is a joke and the results are essentially meaningless.

The tweet about the the results seems to be missing/deleted. Anyone snapshot it before it disapeared?

I'm curious if they have evidence that 29 was faster [i]and[/i] less effort than 26. Having both (albeit on very different bikes to fairly compare) I feel fastest on the one I have recently been riding on the most.

Looking forward to reading the article in the journal, but I doubt it will bear much relation to what's been written so far.

Part 2:

What a joke. They could have arrived at a very similar conclusion with a stopwatch 😀

EDIT. Anyway, hasn't the proverbial horse well and truly bolted already on this topic?

It's not really cleared anything up! It would have been more useful to have had more riders (but still at a similar level) but going over more types of terrain, bigger ups and downs and longer circuits - same test on multiple routes. It's clear on one thing though (what everyone probably knows anyway), that rider and fitness is the biggest performance factor.

I'm really quite annoyed by that. As a scientist you don't state that the results are not statistically significant then spend ten minutes talking as if they were. That's just misleading. This study wasn't sufficiently powered to produce a significant result. So go do another study that is. Anything else is just bad science.

original article seems to have gone as well ?
http://www.bikeradar.com/mtb/gear/article/how-does-mountain-bike-wheel-size-really-affect-performance-43481/

Can't help but think that 27.5 bike was slow because it looked shit. the 26er and 29er had nice paint jobs and stickers.

650B is the slowest up and down... TeeHee 😆

The latest thing isn't always better than what's out there already however much the trail comes alive 😉

On descriptive statistics i.e. average speed, 29er was faster on that course on that day.
For a proper statistical analysis they need more riders. The population was too small to have much statistical power, therefore the effect size needed to be really good to get a significant result. But there's an approach toward significance which may well be validated by a larger sample size.
Make of it what you will. I'll stick to racing and riding on a 29er.

So go do another study that is.

Yawns....

roverpig - Member
I'm really quite annoyed by that. As a scientist you don't state that the results are not statistically significant then spend ten minutes talking as if they were. That's just misleading. This study wasn't sufficiently powered to produce a significant result. So go do another study that is. Anything else is just bad science.

+1

you may as well just get the daily mail to run a feature on it

What we need is another test................really
29ers are faster them's the facts.

Surely it depends on the course. Fast and open = 29, tight and twisty = 26?

Yeah, surely no statistically significant difference is just that,

Dissapointed they didn't break it down further than time, they intimated 29 used less energy but not how many less calories or anything, clearly not enough to show a significant difference.

The best and fittest rider will always be the fastest... Fact.

But I do find it amusing that the "magic there can be only one" in-between wheel size was the slowest by a not insignificant margin. New and shiny isn't always best eh?

Dissapointed they didn't break it down further than time, they intimated 29 used less energy but not how many less calories or anything, clearly not enough to show a significant difference.

This I is what I was hoping for. How many calories does each bike need to get it around the course in say 10 minutes.

I'm getting a new bike soon, bottom line is the make I want don't make 26 anymore and I don't fancy a 29 so 650b it is - I'm totally unfussed, things change and that's that.

'm really quite annoyed by that. As a scientist you don't state that the results are not statistically significant then spend ten minutes talking as if they were. That's just misleading. This study wasn't sufficiently powered to produce a significant result. So go do another study that is. Anything else is just bad science.

Not really no.
Statistical significance is an artifice, and without knowing what threshold has been set you're not in a position to comment on the validity of the results. Let's say they've set a 5% level threshold and their dataset comes of with a 5.5% probability of being incorrect - whilst not then meeting the required significance, it's still pretty likely to hold true. And of course had they set a threshold of 6% then their results would count as statistically significant.
Sport science is one of those areas where the results are rarely statistically significant to typically used significance levels because the datasets are typically too small. That doesn't mean the results are invalid though.
The problem is not with the science, or scientists, but with people not understanding what statistically significant means.
Bit like when people say evolution is only a theory.

But they didn't try a 27plus that would obviously have beaten them all! 😈

I'll wait to read the paper and if his p value is 0.06 or something I'll let him off. Heck I'd accept a 90% confidence for this type of study. But to say that the results are not scientifically significant then try to claim some kind of "practical" significance and to talk about differences of 12sec as if that was absolute (with no mention of errors) bugs me.

Don't get me wrong, I'm no fan of the compromise wheel size and these results suit me just fine. But I don't think they do much to improve the public understanding of science.

Surely it depends on the course. Fast and open = 29, tight and twisty = 26?

I was really suprised that in a straight line it felt slower than my 26" and seemed to take more effort to get it going up to speed. Maybe it was just that particular 29'er but I was expecting to be blown away by it's straight line speed but it just felt like a tank.

I'll wait to read the paper and if his p value is 0.06 or something I'll let him off. Heck I'd accept a 90% confidence for this type of study. But to say that the results are not scientifically significant then try to claim some kind of "practical" significance and to talk about differences of 12sec as if that was absolute (with no mention of errors) bugs me.

Don't get me wrong, I'm no fan of the compromise wheel size and these results suit me just fine. But I don't think they do much to improve the public understanding of science.

Yup, I agree entirely.
BTW I recognised one of those researchers in that video. I think he might still be an STW member, so you there's an outside possibility that you might get an answer straight from the horse's mouth 🙂

Wanna know whats the fastest bike for the type of riding you do? Check out what the pros ride.

Listen, this is all well and good but which is the best for sweet jumps?

edit- schralp that gnar STW!