Safest bike manufactures?

After listening to a bunch of cycling podcasts (other than TR), In a new cycling tips podcast they interviewed Raoul Leuscher, one of the oldest carbon bike repairers. He talks about how new bikes are sacrificing structural integrity in the pursuit of aero and weight. In an earlier episode (when BMC had a recall issue with stems sheering off) in 2019 he says he had to scan around 20 forks before he found one that didn’t have any “considerable areas of voids in it” and that this wasn’t this case 2 years ago.

So my question is, is there a data pool that shows the safest bike manufactures? the least amount of recalls due to something that could potentially be fatal? the most thorough safety manufacturing checks? Reported critical failures on the road?

Trying to find any of these questions has proven near impossible. And i understand that there is a race between all manufactures to be the lightest bike on the market. But I’m surprised no company pushes how rigorous their safety manufacturing program is.

Given how difficult it is to monitor a bike’s life once it leaves the shop, I am not particularly surprised that there isn’t much of data in that regard.

The way I personally try to pick my bikes is based on two factors:

  1. Does the brand make their own components in house ?
  2. How many times has the brand changed hands ?

Once you’ve applied those two, it leaves you with roughly 4 or 5 mainstream carbon bike brands. Then, pick the bikes that are actually made in house, rather than outsourced to another manufacturer.
That leaves you with Look, Time, Giant and Hongfu.

Trek and Colnago make some of their frames in-house but as per my understanding, all of the other components, such as forks and seat posts are outsourced.

Leuscher is frustrating because he shows you voids, extra glue, and bits of bladder left behind inside frames but he never tells you if these things are actually unsafe or if they just aren’t ideal.

The fact is that we are never going to get aerospace level manufacturing of bicycles.

Is there a problem in the industry? We don’t see frames or forks exploding left and right. Just buy a quality brand name bike. Also, buy a bike that is a little heavier and not on the bleeding edge of low weight if you want some extra durability. For example, buy a regular Tarmac instead of an S-Works. Or, look at Giant. Leuscher sort of gave them the thumbs up and they even come with 2 years of no questions asked crash damage replacement warranty.


I agree. It is worth asking him however. He does tend to answer a lot of questions in the comments section.

The one time he said that the bike design was truly awful and unsafe was that Cervelo, where the fork would bang on the inside of the head tube to limit the rotation. That was some seriously scary shit right there. :frowning: :frowning: :frowning: :frowning:

Canyon CT scan all their carbon parts for their bikes. Each Canyon component on my bike has a QR code that can be traced back to its source of origin and the scans that were taken.


Yeah that’s a huge draw to Canyon for me. They’re the only one I know that does CT

1 Like

So that’s what those qr codes are! Thanks!

1 Like

IMHO catastrophic failures that cause injury and not due to a harsh impact are exceedingly rare. As far as I can tell most accidents due to equipment failure are the result of poor maintenance, e. g. if a bolt had rattled loose or had not been fastened to spec.

If you want to have a frame that has a larger margin, then you should look into burlier variants of a bike. Say, you are a road biker, you could get something like the Open UP, which is made for gravel and is therefore sturdier. Or you could get a trail hardtail instead of an XC hardtail. But even that is probably completely unnecessary. I’d just do the following:

(1) Stick to a manufacturer with a long, proven track record. (Or, in Open’s and 3T’s case, where people with decades of experience are at the helm.)
(2) Avoid weight weenie parts and light variants of frames. Sometimes these are rated for a lower rider weight, so if in doubt, check the specs.
(3) Maintain your bike regularly. (Or have someone do that for you.)

1 Like

Also, you should keep in mind that it is his job to deal with failed parts. If you talk to a fire fighter, don’t be surprised to hear a lot about fires.


RE Canyon, Hambini (who I do not like, for a number of factors, but who’s methods and expectations of engineering I have to respect) has not been very complimentary about Canyon’s build tolerances and quality. He does himself no favours by not showing the evidence (aside from the very obvious BB sizing demonstrations offering up and inserting the machined ally blocks).

The more info I read (notwithstanding I am careful about what I believe) and the more I experience the less confidence I have with some manufacturers. This is to the extent there are a few brands I would never buy based on what I have seen from others and a few I would never buy again from my experience.

I know this thread is about ‘safest’ so guess the focus is on what might break, but that for me should be a given and I am equally concerned about the quality of manufacturer and tolerances. For example, a frame with a poorly formed and misaligned BB wouldn’t be any less safe than a perfect frame from the same line but would give me a life of irritation and issues.

Personally I only know of one ‘frame’ that ‘failed’ through riding, a Canyon Spectral swingarm that cracked by the lower fixing point. That was fortunately found by the rider before catastrophic, the other pivot points and shock held the swingarm in place and to some extent disguised the issue (he said it felt floaty on the rear end but on initial visual inspection out on the trails there was nothing obvious!). In Canyon’s credit they eventually replaced the swingarm under warranty without him having to send the whole bike back, but it was far from a straightforward process.

Other issues I have seen from Focus (rim brake fork not symmetrical impacting wheel alignment), Cannondale (machining and tolerance issues), and Specialized (SWORKS Tarmac frame rear triangle twisted) is enough to make me very wary of each manufacturer but none of those were ‘safety’ issues per se.


I don’t believe Canyon CT scans every frame. In the article and their marketing (BS?) videos they mention only scanning the fork, headset and handlebars. The machine sits per my understanding in Germany and the frame manufacturing is out sourced to Asia anyway so how could they scan them…

I don’t think there is such a thing as safe manufacturer. The quality control generally in the industry is very poor, more so in the past especially. Everything is outsourced to the lowest bidder as the legal requirements are so low. If you look at how the frames are manufactured on some videos, there is so much jigsaw puzzling going on by hand that the quality is 100% depending on the person laying up the carbon. Up to 400 pieces of carbon in a jig by people who probably dont have the engineering understanding behind the design, most of the time it can only go wrong somehow imo. Sure you dont have to like Hambini’s personality or Leuscher’s approach and comments but it’s hard to disagree with them, if you know anything about composites, aerospace or motorsport quality standards. Bike industry is driven by marketing, not quality or engineering.

1 Like

Hambini’s beef are the shoddy tolerances and bogus aero claims, but I don’t remember him roasting any of the manufacturers for making unsafe products.

That’s a bit simplistic: Hambini and Leuscher make a living from fixing faulty frames. (I was surprised to hear that this is a business, in one of his videos he explained that some of the bike companies rather pay for one of his BBs than replace the whole frame.) So they are like fire fighters seeing fire all the time. I readily believe that the tolerances are not great, but we don’t know the extent of it.

Plus, crappy tolerances (which accelerate wear on components) need not make a bike unsafe.

Most — if not all — consumer-facing industries are driven by marketing to an extent. Nevertheless, I don’t think that level of cynicism is warranted, I think most people in the bike industry are in the bike industry, because they like bikes.

1 Like

The final assembly is carried out in Germany. From the article I linked:

[…] by checking every single frame and fork in their own industrial CT scanner.

Regarding the construction, surely it is not necessary for the person doing the assembly to understand the engineering behind then design, what they need to be is skilled in following the instructions given and be experts in their field which is carbon composite manufacture.

You’re correct though, all it takes is bad batch of resin, an autoclave with an element thats gone so it can’t heat correctly or whatever to spoil a batch somewhere in the mix, things can and do go wrong but I think the majority of recalls are for this sort of issue, not a flawed design.*

TBH if you want something as safe-as-it-could-be, then you are probably looking at something handmade in titanium from a small operation like Moots or Baum where you can probably find out who made the frame, on what day, and what they had for breakfast on that day!

*The only one I can think of is the BMC Impec which I don’t think ever made it to retail so you can’t really say it was a bad-design recall. Though @SomeCallMeTim mentions a Cervelo but I don’t know of that one.

Has anyone even heard of cases of accidents caused by faulty frames/forks? Even just anecdotally?
I think moving or wearing parts (Drive train, wheels, brakes, tires etc…) would fail much, much sooner than the static frame or fork ever would.

Yeah, it has happened:

That said… What industry doesn’t have recalls? At least eight children are known to have died from falling, improperly secured IKEA dressers.

That’s tragic, of course. I do think it would be ideal to have some sort of independent design review, at the least, let along production. That said, no, it’s not a concern I have when actually buying a bike. I suspect I’m much more at risk from riding on roads and at high speeds than from a failing part. We all make our own judgements, though, of course.

1 Like

Damn. Seems like staying in my apartment is riskier than cycling :scream:

Yeah. Would be cool if there would be independent bike crash tests. That basically counts for other cycling equipment as well. I’m always a bit baffled by helmet reviews in bike magazines in particular. Pages of weight, design, comfort… sometimes they even take them to a wind tunnel… But in terms of safety they often just regurgitate the manufacturers claims.

Hambini is an aerospace engineer for Airbus as far as I know. Bottom brackets are just a small side business. Leuscher for sure is more biased towards his business of course and only sees the problem cases. But still its hard to argue with some of crap they show, e.g. the latest Cervelo headset design is such a basic design brain fart its hard to excuse in 2020.

Being in the bike industry because people like it is not really and excuse for poor quality is it? I would argue that is exactly the problem. I would also argue the top engineering graduates have less desire to work in the bike industry than aerospace or motorsports for example even if you’re into bikes. I know I never considered it as an option. People end up there because they like it, often without the proper academic background or experience. How else would you see such designs like the Cervelo one going out? Any design engineer with some knowledge on composites would say its bs.

I don’t mean to be cynical, just realistic having some understanding on the topic. Most of the time the engineering & manufacturing is just adequate for the average cyclist. They will never notice if the carbon layup is bad or if the tolerances are off, they are just happy its made of carbon. It’s probably even safe under the average user. Problems appear when you have a crash or put the bike under some other extreme use case / racing etc.


Seems to be mixed messages out there regarding what is it exactly they scan and how. Or what does it mean its CT scanned? They scan it, put a QR code in it and save the scan in their database in case of warranty issues later on. Actively checking every single component with one machine in Germany before final assembly doesn’t make much sense to me otherwise.

Actually, a lot of Leuscher’s business is checking bikes after crashes and, if necessary and possible, fixing them – not so much just fixing faulty frames.
I took my Roubaix to him after a bad crash, to see if there had been any structural damage. No damage from the crash, but a ‘couple of small bubbles’ in the top tube that ‘won’t pose any problems at the stresses I put the bike under’.


IMO, the cynicism is most definitely warrantied as getting manufacturer to provide any real data outside of the marketing blob is nearly impossible.

Hell, it’s gotten to a point, where some coughCannondalecough, even start calling their marketing brochures, a White Paper.

1 Like