Toolbox: How do you go about making a scientific comparison of cycling training over time? Every parent, at some time or another, probably has given the “when I was your age” speech to their kids. And within any sport, an ageless argument is always how the current generation of stars match up to the titans of the sport’s history.
Merckx and the new Merckx – Evenepoel
As an university professor, a perennial favourite topic of conversation in the faculty lounge are the state of today’s students compared to previous generations or, heaven forbid, how we behaved as undergraduates. We obviously live life through a very biased perspective, because the underlying theme is almost always that students today are nowhere near as hardworking or committed as our own generation.
That’s of course largely baloney, as our selective memory generally tend to forget the times that we spent partying during our undergrad days and, in my case, the lousy set of marks I had in second year because I thought I had university nailed and didn’t bother working hard at my studies!
The same is true for sports fans, in that the default bias is generally that today’s generation of athletes are too coddled, obsessed with money, and somehow not devoted to the sport. In contrast, the giants of years gone past are somehow better. Of course, much of sports marketing, such as various Halls of Fame, build on the nostalgia for this history.
In our world of cycling, the argument of the greatest cyclist of all time starts and ends at the legendary Eddy Merckx. However, healthy debate still exists at other levels. For example, the Hour Record has continuously improved, in part due to equipment and different positions such as Graham Obree’s super-tucked and superman positions, and in part due to improved training and sport science support. In an ultimately futile attempt to limit progress, the UCI reverted to an “Eddy Merckx” standard for the Hour Record, then backed off on this in 2014, leading to the recent spate of record attempts.
Such accomplishments lend further fuel to the fire: how would today’s generation of riders compare if they were dropped into the peloton with Fausto Coppi, Jacques Anquetil, Eddy Merckx, Bernard Hinault, or Miguel Indurain? Or how would those great cyclists fare in modern cycling?
How would Coppi fare today?
A Fair Comparison?
In some ways, such a comparison is a classic apples and oranges one, in that the sport of cycling has changed to the point of unrecognizability except for the general use of a diamond shaped bicycle frame. But surely there must be some way to scientifically begin to address this question? At the core of the question is this – has the athlete changed or is it the technology that has changed? How do you separate the two?
The closest analog comes from the world of speedskating and a 2010 study from my colleague Jos de Koning from the Vrije University Amsterdam in the Netherlands. Upon analysis, speedskating offers many of the same discussion points as comparing cycling, and especially with more “controlled” environments such as track events. In both sports, there have been a host of technological improvements introduced into the sport. In cycling, despite the diamond shape bicycle being a relative constant, there have been huge advances in aerodynamics.
In the case of speedskating, these include:
• Moving from all outdoor rinks to the domination of indoor ovals (e.g., since 1994, all Winter Olympics have been raced in new indoor ovals. This removes extreme weather and wind as a variable in elite performance.
• Improvements in icemaking technology and surfacing equipment, reducing friction.
• Revolutionary improvements in the skate, namely the introduction of the “Clap Skate” in 1998, leading to much higher power capacities (~12%). Indeed, nearly every world record was shattered in the first year upon their adoption by elite competitors.
• Finer improvements in skate technology, including blade metallurgy, shape, and grinding.
• From the 1950s era of wool touques and tight fitting wool clothing, speedskaters led the way in aerodynamic clothing. First came the introduction of lycra skinsuits in the 1970s, evolving into the highly refined and aerodynamically-shaped speedsuits from 2002 onwards. Overall, there has been about 10.5% improvement in efficiency due to clothing design.
What de Koning attempted to do was to try to factor out these technological advances and to see what was left in terms of being due to the athlete themselves. He took as his basis of comparison the 1,500 m event. At the 1956 Winter Olympics in Italy (1754 m altitude), the world record was jointly set by the Russians Yevgeniy Grishin and Yuriy Michaylov at 2:08.6. As of the analysis, the current world record was set by Shani Davis of the USA, with 1:41:04 skated at Salt Lake City Olympics in 2002 (1423 m altitude), a >20% improvement!
In developing the model, de Koning had to incorporate some basic physics. Namely, while the relationship between power and velocity to overcome ice resistance is generally linear, the effect of ice and air resistance at increasing speed is non-linear. In other words, the greater the speed, there is an exponentially greater impact of air and ice resistance, and therefore the wattage required to go at higher speeds become exponentially greater also.
The first step, therefore, is to develop a model to incorporate all the factors we listed above into an overall coefficient of ice friction. Just like when you read about wind tunnel testing and aerodynamic drag, the lower the better. Overall, de Koning calculated a 1956 coefficient of 0.006. Incorporating the changes, the coefficient was modeled to have dropped down to 0.0025.
So now that we have calculated the effect of these purely “technological” improvements on the coefficient of friction, the next step is to back-calculate what the actual power output of the skaters must have been to achieve their records. This is much easier for cycling due to the ready availability of power data presently, but of course it’s much messier for old cycling records and speedskating in general.
Overall, de Koning’s model calculated that approximately half of the overall improvement in world records were attributable to technological advances, and the remaining half due to the skater themselves. However, interestingly, the available historical data on VO2max and power outputs suggests that the real athletic improvement of skaters has not changed over that time. Therefore, the improvement must have come about in overall economy of movement. In other words, the net skating power output has increased despite no change in biological power output.
How might this have come about? de Koning pointed out other changes in the sport of speedskating, many of which are also relevant to cycling:
• The increased professionalism, and the money involved, meant that athletes could make a full-time career of skating or cycling, thus increasing their training time.
• The availability of indoor ice ovals has greatly increased the amount of days that top skaters could spend on the ice, from approximately 60 days per year in the 1950s, some of which may be cancelled or shortened due to extreme weather, to 200+ days currently. The increased skating time increases the specificity of training, and important component of overall skill and motor coordination. Similarly, the increased availability of overall racing, along with access to track time, increases the amount of overall training possible for cyclists.
• The higher speeds resulting from the above changes also has a synergistic effect on the skaters. The shorter overall time required to complete the 1,500 m effectively changed the race to a “long sprint,” such that the adoption of an all-out pacing strategy becomes more beneficial.
Correcting for all of the above factors, de Koning calculated that 1,500 m records from the 1980s and early 1990s would indeed be in the range of modern world record performances. This would suggest that much of the “modern” nature of the sport highlighted in this section happened prior to the 1980s.
Concerning the introduction of a “revolutionary” technology like the clap skate, de Koning calculated that, when the direct benefits of the skate were removed, the initial world record performances were not that extraordinary. So this was a case where the “early adopters” clearly won out compared to skaters who were more hesitant to try or adapt to new technology.
In the case of cycling, Greg LeMond’s early adoption of aero handlebars and helmet at the 1989 Tour de France clearly was revolutionary, and gave him a clear advantage over the pony-tailed Laurent Fignon in the final time trial in Paris.
The de Koning study raises some interesting questions and points of debate. It is clear that technology can have a revolutionary effect on competition results. Therefore, it definitely pays to be on the lookout for such new advantages. I have never had a history of great time trial performances, but in 1989, I got my dad to help me weld together a self-designed aero clip-on bars, and it resulted in my best-ever 5th place in the TT stage of a 2-day stage race, resulting in a 3rd overall GC placing. However, it eventually does become a level playing field once the technology becomes adopted en masse.
Therefore, it is very likely that modern racers would fare well in the older peloton, and vice versa, in that the basic human capacity has remained similar – human evolution doesn’t work that quickly! For example, the fastest Paris-Roubaix on record was set by the late Peter Post in 1964! This clearly demonstrates the highly variable nature of weather and pack dynamics in road racing. However, it’s the support system around the modern racer that has likely been equally important.
In the end, what this means for us to optimize our performance is that, in addition to smart training and keeping up with technology, the support system (coaching, sport psychology, overall athletic progression, recovery tools, etc.) need to be carefully examined too.
Ride safe and have fun!