Aerodynamic drag in cycling team time trials Blocken et al., 2018

Technology has always been an integral part of cycling. Obviously, bicycles are themselves technology, and also rely on technology (i.e. good quality road surfaces) to be a practical means of transportation. Cycling has also taken its fair share of the spoils obtained from the increase in the rate of technological development of recent years; innovations such as carbon fibre, power meters, and GPS computers (amongst many others) have transformed cycling at every level.

Technological innovations are also helping push forward our knowledge of the physics of cycling, and a great deal of this progress has come from the study of aerodynamics. Field tests and wind tunnels were a starting point, subsequently complemented by computational fluid dynamics (CFD), which utilizes modern information-processing capabilities to simulate highly complex physical situations in a much more comprehensive way than the relatively two-dimensional analysis possible in a wind tunnel.

The importance of aerodynamics is readily apparent to all road cyclists, especially those who have ridden at speed in a group. It's been known for decades that the energy saving obtained from riding immediately behind another rider (or other vehicle) is very large, and it has recently been shown that in a very large peloton the aerodynamic drag of a rider near the back centre of the group can be reduced to below 10% of that experienced by an individual cyclist.

The work outlined here utilized wind tunnel measurements and 3-dimensional CFD simulations to study drafting of up to 9 cyclists riding in single-file formation at various distances. Much of the work consisted of designing and validating CFD simulations. We won't get into the theoretical concerns (although aerodynamics is a fascinating subject, it gets extremely complicated very quickly); suffice it to say that the CFD simulations aligned very closely with wind tunnel measurements, and that reasonable assumptions regarding rider and bicycle geometry were made. All results used a velocity of 15 m/s (54 km/h, 33.5 mph), at which speed aerodynamic drag accounts for 90% of the total drag forces on a rider; for lower speeds this figure would be less, and the magnitude of the energy savings indicated would be smaller.

The main results are shown in the following figures (identical pacelines except for the inter-rider distances, which are 15 cm, 50 cm and 1 m respectively).

Distances of 5 cm and 5 m were also modelled, but aren't shown here (5 cm is unrealistically close, while 5 m is far larger than found in all but the most disorganized groups¹).

Interesting points to note include:

1) the lead rider in a group receives a small aerodynamic benefit compared to riding solo (due to upstream disturbance, i.e. following riders 'pushing' air forward);

2) upstream disturbance also means that the rider experiencing the greatest drag reduction is not the last rider for groups with six or more members;

3) riding second in a paceline reduces drag to 60-70%, and riding third to 50-60%, compared with an isolated rider. Beyond fourth place in a long paceline, however, additional benefits are much smaller. But remember that this analysis is for a single line of riders; as noted above a large, multi-column peloton reduces drag even more, since it produces a much larger wake from multiple lead riders.

The most illuminating point to me, though, was how little difference there is between the three distances shown in the charts: the middle rider of a 5-man group, for example, experiences 50%, 52% and 54% of the drag of an isolated rider at 15 cm, 50 cm or 1 m, respectively.

Although this is certainly significant for a team time trial or pursuit race, for amateur riders, even when racing, I'd suggest that the 2% reduction of drag when going from 50 cm to 15 cm isn't worth the additional risk of a crash: if the lead rider is doing 400 Watts this difference amounts to just 8 Watts², but 15 cm (6 inches) gives you almost zero reaction time, whilst 50 cm (over two-thirds of a wheel length) allows much more room to manoeuvre. In practice you might be somewhere in between these two distances, but the main message is not to worry about squeezing out every last centimetre, especially in a situation with unknown riders on a technical course.

To conclude with a personal anecdote: on my most recent group ride there was a 15-minute period where I rode in a 3-man line, during which I averaged 240 Watts and the group as a whole maintained a steady pace with evenly-spaced changes every 90 seconds or so. We know and trust each other quite well, so kept the gaps small. My average power during this period was as follows: on the front 320 W, on the back 180 W (56%) and in second 210 W (66%). These numbers are remarkably close to those shown in the charts. Each of us could probably keep up a 240 Watt average for two or three hours of riding, but wouldn't last more than a few minutes at 320 Watts.

That's the power of drafting.

1 although even gaps this large still produced substantial beneficial effects.

2 actually even less, because to keep things simple I'm crudely (and incorrectly) assuming that aerodynamic drag is the only force acting against cyclists.

#aerodynamics #drafting

#cycling #research #science

The 4000-m team pursuit cycling world record: theoretical and practical aspects Schumacher and Mueller, 2002

This is an extremely interesting research article, not least because it gives a detailed description of the training practices of world-class athletes. Such information is generally hard to come by, so when I find some I give it my close attention!

The paper actually begins in quite unremarkable fashion, detailing what were at the time the necessary power requirements (an average 480 Watts for the four team members over 4 minutes¹) to secure a world record in the track cycling Team Pursuit event, modeling the workload and physiological demands of the event, and giving an equation to estimate winning times in future competitions based on regression analysis of previous winning times.

However, for our purposes the most interesting material is the included discussion of the team's training program. Given that the event in question takes place over a distance of just 4 kilometres and thus has a high anaerobic component (with a need to maintain 100% VO₂max power – well above lactate threshold – for the full duration of the race), you might imagine that the riders would focus heavily on anaerobic training, such as high-intensity intervals. In fact, there was surprisingly little of this until the final days immediately prior to the event, at the 2000 Olympic Games.

Overall, the training in the months leading up to the event was dominated by high mileage, low intensity rides (3-8 hours/day; 29,000-35,000 km/year), referred to in the article as 'basic training'. This is exactly the same as the old-fashioned base training rides employed by road racers in the off-season, done at 30-50 bpm below lactate threshold heart rate (Z2 in a 5-zone system).

This training led into periodic workload peaks in the form of one or more road stage races, which increased in frequency and difficulty as the Olympics date grew nearer, and culminated in a short period of track training to finish each of 3 macrocycles (see Figure 1; note that the darkest boxes are the track training, and that blank sections involved unstructured light training at home).

The final track training period covered the 8 days immediately prior to the Olympic Games, and the two earlier ones were just 4 days each. That's 16 days of event-specific training in 6 months! This track training was divided into 'evolution' training (6 minute blocks within 5bpm either side of lactate threshold heart rate), and event-intensity 'peak' training (1- or 2-minute blocks at maximal 4-minute power). Both evolution and peak training were performed at around 130 rpm with 20 minute breaks between repeats. As shown in Table 2, a maximum of only 7 evolution and 2 peak blocks were carried out in a given day.

The training distribution outside of the stage races and track training (i.e. 70% of the training days) was 94% below anaerobic threshold, 4% around threshold, and 2% above threshold. These numbers would change with the inclusion of the stage races, although even the tougher races would also consist predominantly of low-intensity riding, punctuated with periods where increased effort was demanded.

The purpose of the track training was to boost anaerobic capacity, as well as allowing focus on discipline-specific technical and motor skills. As we have seen, however, this didn't require a great deal of time to accomplish (about 8% of total training days in the six months leading up to the Olympics).

The remainder of the paper discusses in further detail specifics related to the determination and track-order of team members.

However, for us the lesson is clear (and there's a reason that base training is so named): aerobic adaptations are the major physiological determinants of performance, even in an event consisting of just 4 minutes of maximal effort. Anaerobic capacity must be added on top of this.

Because of the much more rapid recovery following low-intensity riding compared with training at threshold or higher intensities, this type of training can be repeated consecutively for many days, allowing a training volume (and associated adaptations) to be amassed that is much larger than would be possible when training at higher intensities. This is an ideal way to build a potentially huge fitness base.

Furthermore, in addition to being built upon and benefiting from aerobic adaptations, many anaerobic adaptations can develop over a much shorter time frame than aerobic ones (partly because they actually have less development potential). Understanding why this is true necessarily involves examining the specific physiological changes that we're talking about in some detail, which is a large topic that we'll look at in future, across multiple articles.

The big caveat for amateur riders when applying this to ourselves, of course, is that most of us don't have the time and resources necessary to emulate this kind of training. So we must make compromises. This also is a huge topic that I'll repeatedly address in future articles. A clue for now: if we can't match the volume, we must modify the other term of the training equation i.e., intensity. But exactly how we do this is of great importance; some ways are much better than others.

1 Note also that, prior to the Team Pursuit, two of the four riders on the team finished first and second in the Individual Pursuit event.

#training #intensity #olympics #worldrecord

#cycling #research #science

Effects of saddle height on economy and anaerobic power in well-trained cyclists Peveler and Green, 2011

Bike fit is a crucial aspect of cycling, both for maximizing performance and minimizing the risks of injury. And probably the most important single component of bike fit is saddle height. There are various methods used for determining optimum height, including the heel, LeMond, and 109% inseam methods. However, the most direct method is to measure a precise knee flexion angle at the maximum extent of the pedal stroke. The current experiment complemented previous ones in determining the optimal knee angle for performance.

This study took it as well-established that the safe range of knee angles is 25°-35°, but investigated whether there were any differences in performance between the lower and upper end of this range. It had been shown earlier that the 109% inseam method leads to wildly variable knee angles (from 19° to 44°, and outside the recommended 25°-35° range more than half the time), presumably due to inter-individual variability in femur, tibia and foot lengths.

There were indeed differences, albeit fairly small. Two tests were performed for each knee angle, measuring economy and power. The economy test consisted of 15 minutes of pedaling at fixed resistance and cadence, measuring oxygen consumption as an indication of economy (lower oxygen consumption indicating greater economy). The power test consisted of a 30 second maximal effort. The subjects were well-trained males.

There were small but significant differences favouring 25° over 35° in oxygen consumption (1.0%), perceived exertion (3.5%) and mean power (2.7%). Like I said, these differences are quite small — likely unnoticeable for the recreational cyclist — but potentially significant during an actual race. Certainly, people do all kinds of crazy (and expensive) things in an effort to gain a couple of watts, so potentially getting some for free seems like a good deal.

The study excluded a high number of subjects, and so ended up being quite small, so I'd definitely like to see a larger one looking at multiple knee angles around 25° in future, and another thing that would have been very interesting would be a follow-up of athletes a few weeks later. There was no indication in the article about what the prior knee angle of each athlete was and therefore no way of knowing how different a particular angle was from their usual angle. If 25° was a significant departure from normal, having the athletes continue to train and race using this 'optimum' knee angle might actually increase the observed gains as they got used to the new setup.

#bikefit #saddleheight

#cycling #research #science

Anthropometric comparison of cyclists from different events Foley, Bird and White, 1989

Whether you're a serious cyclist, a novice or simply prefer watching others do the suffering, you're probably aware that, although clearly an endurance sport, different types of cycling suit different types of rider. Even the casual fan knows that there are climbers, sprinters and time trialists.

Obviously, there are certain essential characteristics that are shared by all strong riders and there is a large overlap between the categories, but nevertheless there are specific physical traits that predispose a rider to excel in one discipline relative to the others. Some of these traits are based on unchangeable skeletal features.

This study looked at this, based on rider specialization. Riders were placed into one of four categories based on their strengths: sprint (track and road), pursuit, time trial (including ultra distance) and all-rounders. The most significant differences between the groups were found in height, femur, lower leg and total leg lengths, foot length and somatotype (the ratio of endomorphy, mesomorphy and ectomorphy).

In all cases, the sprinters had the lowest values, i.e. they were the shortest in stature, leg and foot measurements and were the most mesomorphic (muscular). The time trial specialists were the tallest with the greatest leg and foot measurements and highest ectomorphy (slenderness). The pursuit and all-round riders were intermediate.

These results were consistent with earlier studies. They were explained in light of the crucial importance of strength and cadence in sprinting, respectively facilitated by a relatively muscular physique and short legs. In contrast, for time trial riding a more effective approach is to push a large gear at a relatively low cadence, so therefore there could be mechanical advantage to having longer legs and a leaner body frame.

This is quite an old study, and I failed to find any more recent follow-ups, which is a shame because I'd love to see a larger, more comprehensive analysis.

For what it's worth, I followed the authors' methods on myself which emphatically placed me in the sprinter category. This didn't come as a surprise, as (although I've never done a proper power profile test) I'm clearly stronger at short efforts relative to longer ones. Does this mean I should use this as an excuse to cut out the long rides, and just focus on intervals and pumping iron? Of course not; cycling is an endurance sport, and even short criterium races demand a lot of endurance; there's no point having a devastating sprint if you're too tired to produce it at the end of an actual race.

What it might do is give me more confidence that the specialization phases of my training should indeed be focused on power and top end speed and that my best chance for good results will be in criteriums as opposed to long road races or hill climbs. But my overall training plan should be quite similar to any other amateur cyclist, just with a different emphasis.

#anthropometrics #bodytype

#cycling #research #science