VeloClub is CyclingTips’ membership program which brings us closer to our members, and connects likeminded cycling enthusiasts.
by Matt Wikstrom
September 6, 2017
Photography by Matt Wikstrom
It’s almost inevitable that every road cyclist will start to wonder about the crank length they are using. It’s a topic that is surrounded by a lot of personal anecdotes and opinion, but how much formal testing has been carried out?
In this article, Australian tech editor Matt Wikstrom takes a look at research on the influence of crank length on the performance of road cyclists, and explains that the results are actually quite clear.
There’s no denying the fascination that surrounds cycling equipment. For some, it is a matter of style and/or function; for others, it represents an opportunity for improving their performance. Given that there is a strong competitive aspect to cycling, it’s not surprising that the latter has been responsible for an incredible amount of innovation.
The whole notion of free speed and improved efficiency is compelling in an endurance-oriented sport like cycling. “Marginal gains” has become a popular catch-cry for coaches and bike engineers alike but the compulsion for re-visiting the design of any part of the bike and every piece of kit with the hope of finding free speed is decades old.
Cranksets have received a fair share of this attention. Originally made from steel, they have evolved from largely utilitarian creations to become lightweight and elegant. Aluminium alloy remains the most common construction material, however the last decade or so has seen the successful introduction of composites. The same period has also seen immense proliferation in axle and bottom bracket designs while chainrings have been getting smaller and using fewer bolts.
One thing has remained constant throughout all of this refinement: the length of the crank arms. It has hovered around 170mm since the inception of the safety bike at the turn of the 20th century, and with good reason: it’s long enough to serve as an effective lever yet short enough to remain within the range of motion of the human leg.
At one point, during the dominance of the English bike industry prior to World War II, an attempt was made to standardise crank length (6.5in/165mm) but road racers started to challenge that notion in their quest for an edge. Campagnolo’s new cotterless cranks, introduced in 1958, were distinguished by a generous range of sizes, 165-180mm in 2.5mm increments, and remained so for decades to come.
Interestingly, this range of crank lengths pre-dated much of the formal research on the impact of crank length on a cyclist’s performance. One early study was published in 1953; otherwise it wasn’t until the ‘80s and beyond that the issue was examined with any rigour.
While research on this topic may have been slow to start, it has received a lot of attention over the last 10-15 years and efforts are ongoing. As with any field of research, there has been some contention, and some of the results may run counter to conventional wisdom. I will discuss this in more detail below, but for those hoping for a quick answer, here it is: there is no evidence that crank length has an effect on a road cyclist’s power or speed.
Any debate on the influence of crank length normally starts out by considering the problem in terms of simple physics. When viewed from this perspective, a bicycle crank is considered a lever, and hence, any increase in the length of the cranks has the potential to provide the rider with extra leverage.
While this approach does a lot to simplify the problem, it does not allow for the influence of biomechanics, which, as it turns out, is quite considerable. After all, there are three human-powered joints involved in driving each side of a crankset that require energy in extension and flexion, so there is more to the problem than simply calculating leverage.
Nevertheless, the influence of crank length on leverage for the drive train can be demonstrated under a set of very specific circumstances, namely from a standing start with a fixed gear over a short distance (100-200m). Then, longer cranks allow a rider to develop more speed than shorter cranks, even when the difference is as little as 2mm.
This kind of scenario is quite removed from road cycling, since riders spend most of their time seated and have the freedom to change gear ratios as they please. Under these circumstances, crank length has no effect on maximum power output, and indeed, near-identical results have been observed for a substantial range of crank lengths.
For example, Inbar et al. (1983) measured the mean and peak power output for 13 subjects during a seated 30s effort using crank lengths 125-225mm. While the authors identified an optimal crank length of ~165mm for this kind of effort, there was no significant change in power when cranks were as long as 200mm or as short as 150mm. Beyond that, there was evidence of a small decline in power for 125mm and 225mm cranks, however the losses were relatively small (2-5%).
Martin and Spirduso (2001) essentially repeated this study with 16 trained cyclists for a 3-4s effort and five crank lengths (120/145/170/195/220mm) with very similar results. The researchers noted a small decline (~4%) in power for the shortest and longest cranks, otherwise there was no difference between 145mm, 170mm and 195mm cranks.
If crank length has no effect on power output for a road cyclist, can a rider save energy by changing the length of the cranks? Research on this question goes back as far as 1953 when Astrand measured oxygen consumption by cyclists riding a bike on a treadmill. Changing the crank length from 160mm to 180mm and 200mm had no effect on oxygen consumption whereas a change of tyres did.
Morris and Londeree re-visited this topic in 1997 with a group of six trained cyclists and found that a small change in crank length (5-10mm) increased oxygen consumption by up to 11% during a lengthy submaximal effort. However, it’s worth noting that the subjects in this study were required to maintain the same cadence (90rpm) for each crank length tested, which may have influenced the effort required.
Indeed, a subsequent study by McDaniel et al published in 2002 clearly demonstrated that the metabolic cost of cycling was largely dependent upon power output, cadence, and pedal speed. A switch between three crank lengths (145/170/195mm) during the course of this study actually had no effect on metabolic cost per se.
Ferrer-Roca et al. (2017) subsequently confirmed these findings with a smaller range of crank lengths (±5mm preferred crank length) while considering the effect on biomechanics, noting that longer cranks increased flexion and the range of movement required at both the hip and knee. This wasn’t the case for shorter cranks, leading the authors to recommend that where there is indecision, cyclists should opt for a shorter crank to reduce the risk of injury.
Wondering what crank length you’re using at the moment? Look on the back of any crank arm to find the length.
Based on the evidence presented above and elsewhere in academic literature, there does not appear to be a strong argument for optimising crank length for an individual, at least in terms of pure performance.
But there is more to cycling than simply generating power. There is the demand of maintaining a highly repetitive activity for long periods in the context of fluctuating loads. The bicycle itself is a highly symmetrical machine while the human body is typically asymmetrical, so the potential for uneven loading is enormous and injuries are common.
In fact, a high proportion of cycling injuries relate to overuse for both recreational and professional cyclists. The legs are commonly affected, especially the knees, and while the causes are many and varied, the most common prevention strategy is to modify the rider’s position on the bike.
This is where the optimisation of crank length becomes important. While the position of the saddle can be adjusted to suit the overall reach of the legs, the length of the cranks largely dictates the range of motion. As a result, bike-fitters have come to view crank length as an important parameter that can be optimised for every individual, regardless of whether they own a factory-built bike or are selecting the parts for a custom road bike build. While this optimisation probably won’t improve the performance of the rider in terms of measurable power, it can add to comfort and prevent injuries.
The current market offers a pretty generous range of crank lengths, starting as short as 160mm and extending to 180mm, often in 2.5mm increments. In addition, there are a few manufacturers offering custom-built cranks outside this range, so it is possible to fit significantly shorter (e.g. 130mm) and longer (220mm) cranks to any given bike. Thus, there are plenty of products available for optimising crank length, but how does a rider to decide on a specific length in the first place?
Longer cranks can make a difference, but only for short sprints from a standing start with a fixed gear ratio.
While it is generally acknowledged that crank length should increase with the height and leg length of the individual, the exact association remains vague at best. One early study (1976) experimented with different proportions of crotch height and concluded that ~20% was the most suitable. Decades later, Martin and Spirduso (2001) arrived at much the same recommendation (20% of leg length).
A variety of other formulae have been proposed over the years ranging from simple equations to more complicated approaches. Each formula is an attempt to describe an association between measurable parameters (e.g. leg length) and a functional outcome based on a finite number of subjects, so a lack of consensus really isn’t surprising. Nevertheless, they have found favour because of the ease they offer, but in strict terms, they do little to settle the matter.
That’s because crank length is part of a system of hinges and levers that must operate in the larger context of an individual’s biomechanics. Most of these formulae fail to consider this at all, effectively isolating the issue from all other considerations, and for this reason, it is probably best to view any result as theoretical at best.
While any of these formulae might provide a starting point for further investigation, it makes more sense to get some advice and direction from an experienced bike-fitter, if only because cranksets tend to be quite expensive.
Stewart Morton has over 10 years experience as a bike-fitter and he still considers crank length a can of worms. “For the most part, by understanding a rider’s cycling goals and their riding discipline, and assessing their body (flexibility, joint range, injury), I can figure which crank length might be most appropriate.”
“For those in the middle of the bell curve for height then 167.5-175mm cranks will work. The industry has done a pretty good job using anthropometric studies to create bike models with size-appropriate crank lengths,” said Morton.
Thus, in some circumstances, there is no need to change the cranks, and in others, it’s possible to accommodate the rider’s preference for a specific crank length. “Whether a rider runs longer or shorter cranks, I will still aim to get the knee extension and saddle placement neutral to allow injury-free and efficient pedalling.”
Morton also understands that a change in crank length can allow the rider to safely assume a more aggressive position on the bike without discomfort or the risk of injury. “Ironman athletes are running shorter cranks — down to 155mm, in some cases — to help maintain a healthy hip angle as they rotate their bodies around the bottom bracket,” said Morton. “They can move forward and lower the front end of the bike and still make a good transition off the bike for the run.”
Ryan Moody spends his days deciding the final fit for Baum’s custom-built bikes. “I try to put anyone on the longest crank possible within their range of movement to use the muscle mass they have,” he said. “This is especially important for strength-oriented riders. For those riders with greater cardiovascular strength, a shorter crank works better because they tend to ride at higher cadences.”
Deciding on the crank length is not a simple matter, though. The handlebar and saddle position can influence how well the legs are moving, as can the cleat position. Moody’s ultimate goal is to achieve a clean motor pattern for the entire pedal stroke and has found that even minor changes (1-2mm) to the position of a contact point can have a profound effect.
And because he is not working on modifying an existing bike, he is free to customise the geometry of the frame to accommodate each part of the bike, including the cranks. “The bottom bracket height should suit the length of the cranks, especially if they are longer than normal. Putting a long crank (180mm) on a bike with a low bottom bracket will cause problems with pedal strike when cornering.”
Interestingly, Moody also sees crank length as having an effect on the way that a rider can balance their weight on the bike. “A typical male is top-heavy because of their upper-body strength and a longer crank can help them keep it balanced over the handlebars. It’s the opposite for females, who don’t have the same kind of upper-body mass, so a shorter crank is often better.”
Both Morton and Moody are generally happy with the range of crank lengths available and consider 2.5mm increments adequate for their needs. Nevertheless, there may be a growing need for shorter cranks (155-160mm), and not just for Ironman athletes.
“The cycling culture is changing and it’s no longer all about reasonably tall men with lots of leg strength,” said Moody. “A lot of women need shorter cranks for a better fit on their bikes, shorter Asian populations too, but I don’t see crank lengths getting smaller on mainstream bikes yet. These are the kind of riders that are going to benefit most from a change in crank length.”
The biomechanics of cycling is complex and multi-faceted, so concentrating on a single aspect, like crank length, is bound to suffer from oversimplification and generalisation. Nevertheless, academic researchers have managed to examine this issue with growing sophistication over the last few decades to better understand the influence of crank length on the performance of cyclists.
On balance, the weight of the available research indicates that crank length does not influence the speed, power or efficiency of a road cyclist. What is more important is how the cranks are used, and this is where the training, experience and intrinsic capabilities of the rider make all the difference.
For those riders that have been tempted to try a different crank length based on the promise of extra leverage and perhaps increased efficiency, there are no such gains to be made. The cost of a new crankset will be better spent on formal training with a coach.
By contrast, those cyclists that have been suffering with a recurring overuse injury may find relief with a change in crank length, but this is not something that should be attempted through self-directed experimentation. A trained bike-fitter with extensive experience is likely to take less time and arrive at a more robust solution because they possess the understanding and objectivity to judge a cyclist’s position on the bike.