What does the power profile of professional U23 riders look like? And how do they train to perform on UCI Continental level races?
A recent study by Leo and colleagues examined the training and power profile of professional U23 cyclists over three seasons (1). They then analysed how changes in training variables related to changes in performance throughout the racing season.
The study produce some interesting numbers that adds to our understanding of training for better race performance, whether you are an up-and-coming junior or a seasoned age grouper.
Who were the riders?
30 male U23 professional cyclists took part in this study. All riders were members of a UCI Continental U23 development team.
The physical and physiological characteristics of the athletes are summarised below (mean values given):
- Age: 20 years
- Height: 182.6 cm
- Weight: 68.9 kg
- VO2 max: 73.7 ml/kg/min
21 of the riders were characterised as allrounders and 9 as climbers.
What did this study look into?
According to the authors, the aim of this study was two-fold:
- To compare the power profiles derived from training and racing
- To assess the variation in training and examine the relationship between these changes and changes to the power profile
In order to investigate this, the scientists collected data from the power meters and heart rate monitors of the riders during 3 entire years of training and racing.
Each season was split into 4 periods: pre-season (Nov-Jan), early-season (Feb-Apr), mid-season (May-Jul) and late-season (Aug-Oct).
In each pre-season, the athletes performed a VO2 max test and a critical power (CP) test. * The CP test involved 2, 5 and 12 minute maximal efforts. Mean maximal power data (MMP) were also extracted for the 4 periods of each season.
* Critical power is a much used parameter by scientists and coaches. It represents the intensity at which further increases in intensity can no longer achieve a steady state of VO2 consumption. For the sake of simplicity, you may think of CP as a power value that is close to anaerobic threshold. Furthermore, calculating your CP also allows for estimating your anaerobic capacity, known as W′ (2).
The authors then analysed the training data with regards to volume, training load and intensity. Finally, changes in power profile were contrasted to changes in training characteristics.
How did they test?
The initial pre-season testing resulted in the following baseline results (mean values given):
- VO2 max: 73.7 ml/kg/min
- Ventilatory threshold 1: 256 W (3.8 W/kg) *
- Ventilatory threshold 2: 367 W (5.3 W/kg) *
- Critical power: 382 W (5.5 W/kg)
* VT1 and VT2 is typically considered to draw the line between low and moderate intensity (VT1) and between moderate and high intensity (VT2) work.
While the authors do not provide analyses of how the power profile developed throughout the season, the do provide this data in a plot comparing MMP results from training and racing.
They do however state that “the power profile was maintained throughout a competitive season, with some improvements in the relative power profile due to reduced body mass.
What can you learn from this study?
The first and very obvious finding from this study was that mean maximal power (MMP) recordings achieved through races were consistently higher than those achieved through training.
This is useful knowledge if you are actively using your MMP curve to monitor you training progress. The lesson being that if you are not in a period with regular racing, your MMP values are likely to underestimate your true power profile.
It has thus been recommended to verify field derived MMP values with a minimum of two maximum effort field tests per season (i.e. CP tests) for baseline comparisons.Leo et al, Sports 2020
Increase in volume throughout year
The volume of training increased throughout the training year before dropping off in lete-season. The total hours of training in each quarter of the season were 167 – 202 – 219 – 150 for pre-, early-, mid- and late-season, respectively.
The authors note an interesting observation between pre- and early season.
From pre-season to early-season, the riders increased their total training volume from 167 to 202 hours. Furthermore, total work performed and work per hour increased during this period.
Effectively, this means that early-season included more training hours, and that those hours may have involved a higher mean intensity. *
* The authors actually used several parameters for tracking intensity. These parameters showed somewhat diverging results. However, I would argue they more likely than not indicated an increase in hourly training load.
Interestingly, the changes in power profile (for 2 and 5 min MMP) of the U23 riders between pre- and early-season were inversely correlated to the increases in training workload. Furthermore, changes in 12 min MMP and CP in the same period was inversely correlated to the number of race days.
What this means is that riders with bigger increases in training workload during early-season achieved poorer 2 and 5 min MMP results in that same period.
Similarly, riders who rode more races produced lower 12 min MMP and critical power results in early-season.
…if the riders increased training load or race days too much, a decrease in the power profile occured.Leo et al. Sports, 2020
As a consequence, Leo and colleagues propose the following practical recommendation:
…as racing is introduced in early-season, total work should not be further increased; to achieve this, a reduction in the intensity of the overall volume may be beneficial.Leo et al. Sports, 2020
Threshold distribution of training intensity
Over the course the 3 seasons, the riders totaled an average of 738 hours of training and racing per season (cycling work only, additional training methods not included).
This volume is within range of that which have been previously reported from successful professional road cyclists. By comparison, Thibault Pinot trained yearly volumes in the range of 840 – 876 hours at the age of 20-22 (3).
What I found more surprising was the distribution of training intensity achieved by these U23 riders.
The quarterly distribution of training across low, moderate and high intensity (separated by VT1 and VT2) ranged from:
- Low intensity: 17-19%
- Moderate intensity: 51-63%
- High intensity: 8-9%
Most strikingly is the considerable proportion of time spent on moderate intensity work, and the subsequent scarce use of low intensity training.
The authors characterise this as a threshold training distribution of intensity.
Professional riders seem to opt for pyramidal distribution
From my perspective, these numbers are in stark contrast to the intensity distribution often reported in elite athletes, namely the polarised and pyramidal training models (4-8).
In my preparations for a post I am writing, I recently performed a (highly non-systematic) review of the literature on training intensity among professional and successful amateur cyclists.
The very consistent picture seems to be that they adhere to a pyramidal intensity distribution, with far greater low intensity volume, and a more conservative use of moderate intensity training.
To put the U23 results of Leo et al in perspective:
The very lowest proportion of low intensity training reported in all the studies I examined was 63% (mean result across 4 riders). Across all studies, proportions of low intensity training ranged from 63 to 91%.
Furthermore, among these data the most highly performing professional cyclists consistently report between 80 and 90% low intensity training (by either power data, HR time-in-zone or by modified session goal methods of reporting).
In other words, the training of the U23 riders in the recent Leo et al study diverge considerably from that reported by more accomplished professional cyclists.
The obvious next question is – why is this?
The authors themselves propose several possible explanations:
- They initially hypothesised that the high amount of moderate training may have been due to the large number of race days. As intensity is arguably more difficult to control in-race in the midst of a competitive peloton.
However, the authors analysed for this and found no correlation between number of race days and time at moderate intensity.
- The high percentage of moderate work may have been the coaches’ preferred approach.
- The athletes may have trained with insufficient intensity discipline and ended up executing low intensity sessions too hard – which is a common mistake (9-10).
- A final possibility is that the training zones applied were inaccurately anchored to the riders’ physiological thresholds – which from experience is also a common pitfall.
There simply is no way of knowing for certain why these athletes reported so little low intensity and so much moderate intensity training.
However, I would suggest that there is currently sufficient evidence in favor of discouraging up-and-coming cyclists and avid age groupers from deliberately pursuing such an aggressive distribution of threshold training.
Enhanced power outputs with more polarised training
The authors note that between early- and mid-season, the riders moved towards spending slightly more time at high and low intensity.
Interestingly, these changes were correlated with an improvement in power output (1).
In other words, moving away from the threshold model in favor of greater polarisation of intensity had a positive impact on race performance. Although perhaps a bit speculative, this could be viewed as a subtle hint that riders would have possibly benefitted from a lesser degree of threshold training.
Firstly, this study demonstrate that mean maximal power results achieved during training may underestimate your true performance potential.
Secondly, higher training loads and amount of races in early-season was associated with a decline in power profile in these U23 riders.
As a consequence, the authors recommend avoiding increases in training load once race season begin. To maintain a stable load when racing is introduced it may be useful to reduce training intensity somewhat.
Finally, these U23 riders display an emphasis on moderate training that deviates significantly from that which is often reported in successful cyclists. In my opinion, it is probably advisable to pursue higher volumes of low intensity training than what is reported in this study.
The original paper is published open access, in which you can find all the details on the power profile of the riders:
Cover photo credit: Norges Cykleforbund
- Leo P et al. Training characteristics and power profile of professional U23 cyclists throughout a competitive season. Sports, 2020;8:167
- Chorley A and Lamb KL. The application of critical power, the work capacity above critical power (W’), and its reconstitution: A narrative review of current evidence and implications for cycling training prescription. Sports, 2020;8:123
- Pinot J and Grappe F. A six-year monitoring case study of a top-10 cycling Grand Tour finisher. Journal of Sports Sciences, 2015;33(9):907-914
- Seiler S. What is best practice for training intensity and duration distribution in endurance athletes? International Journal of Sports Physiology and Performance, 2010;5:276-291
- Seiler S and Tønnessen E. Intervals, thresholds, and long slow distance: The role of intensity and duration in endurance training. Sportscience, 2009;13:32-53
- Stöggl T and Sperlich B. The training intensity distribution among well-trained and elite endurance athletes. Frontiers in Physiology, 2015;6:295
- Bourgois JG et al. Perspectives and determinants for training-intensity distribution in elite endurance athletes. International Journal of Sports Physiology and Performance, 2019;14:1151-1156
- van Erp et al. Training characteristics of male and female professional road cyclists: A 4-year retrospective analysis. International Journal of Sports PHysiology and Performance, 2019;15(4):534-540
- Scantlebury S et al. Understanding the relationship between coach and athlete perceptions of training intensity in youth sport. Journal of Strength and Conditioning Research, 2018;32:3239-3245
- Brink MS et al. Coaches’ and players’ perceptions of training dose: Not a perfect match. International Journal of Sports Physiology and Performance, 2014;9:497-502