Podcast, Science, Training

How World Tour cyclists train, and a discussion on training load with Teun van Erp, PhD | EP#250

 September 7, 2020

By  Mikael Eriksson


Teun van Erp

In this Episode you'll learn about:

  • Different measures of training load and how they stack up in scientific research
  • Pros and cons of using Training Stress Score (TSS), Lucia's TRIMP (heart rate based measure), and session-RPE to quantify training load
  • Comparing the training characteristics of male and female professional cyclists
  • Training Intensity Distribution
  • The types of moderate and high intensity training used by professional cyclists
  • The use of different cadences in training
  • Advice to amateur cyclists and triathletes for improving cycling performance

Sponsored by:

Precision Hydration

Precision Hydration
One size doesn't fit all when it comes to hydration. Take PH's free Triathlon Sweat Test to get personalised hydration advice tailored to what you're training for. Use the promo code THATTRIATHLONSHOW15 to get your first box for free!

The finest triathlon wetsuits, apparel, equipment, and eyewear on the planet. Trusted by Lucy Charles, Javier Gómez-Noya, Flora Duffy, Mario Mola, and others. Visit roka.com/tts for 20% off your order.



04:15 -

  • I worked with the Sunweb Pro Cycling Team and it’s predecessors for nine years as a sports scientist.

    During my time with the team, I collected tons of training data, which I came to analyze as part of my Phd.

    As I was conducting my Phd I realized that I had a great passion for research and science and hence I switched job and currently I am doing my postdoc at Stellenbosch University.

Quantifying training load

06:40 -

  • My main subject of my Phd thesis was different ways of measuring and quantifying training load.
  • One of my main findings of my thesis was that the correlation between all of the most common ways of quantifying training load is very high, which depends on that the time factor in each of the different methods (either based on kilojoules expenditure, intensity factor, heart rate or rate of perceived exertion) for measuring training load plays a dominant part in the algorithm.

    However, as the correlation was found to be very strong between these methods in training, a weaker correlation was seen when quantifying the training stress from competition.

Different methods for quantifying training load


  • Training stress score (TSS) is a way of quantifying training load based on the intensity factor (IF) and duration, where a 100 TSS points corresponds to riding 1h at your FTP.

    The TSS algorithm heavily emphasizes the intensity factor of the workout, yielding high TSS points for high intensity rides and massively lower points for lower intensity rides.

    For instance, if you ride at an IF of 0.35 for four hours, then you get 48 TSS points, but if you would ride at an IF of 0.93 for 1.5h, then you would collect 130 TSS points, and the total energy expenditure (kilojoules burned) would be exactly the same for both of these rides.
  • Heart rate based quantification of training load (TRIMP) uses HR to estimate training load, which major issue is that during big training blocks or Grand Tours, heart rate becomes suppressed towards the end of such blocks or stage races, and hence, it looks like you didn’t put much stress on your body then.
  • Quantifying training load based on rate of perceived exertion (RPE) is a really simple way of quantifying training stress, which also is super easy to collect.

    However, the major issue with the method is that the algorithm is based both on the RPE score (typically a number between 0-10 where 0 corresponds to no effort and a 10 is an all out effort) and the duration, which is problematic since the duration aspect is an important determinant of how hard the athlete will rate the session, and consequently the time factor is doubly emphasized in the formula.
  • In summary, all the methods have pros and cons and therefore I don’t recommend anyone in particular, rather, I would advocate collecting them all and utilize them depending on the situation.

    It’s hard to know exactly how to utilize the training load data from the different methods, but one example is that over the career of a pro cyclist, the chronic training load should steadily increase over the years (be higher in a 30-35 years old cyclist compared to a 20 years old athlete), which can be tracked by either one of these methods.

    When I worked with Team Sunweb, we also utilized the training load data to ensure so the chronic and acute training load is in the right place before for instance a certain time of the season or a race.
  • There aren’t too much other research done on this topic, but a colleague of mine found that the TSS system was the method best correlated to performance.

Differences between male and female professional cyclists

25:20 -

  • There are several differences between female and male cycling races as well as between the cyclists.

    The female races are much shorter, which drives up the intensity and the dynamics of the race, for instance will a break away almost never get more than 1-2 mins gap before the peloton.

    The fitness level of the female riders in a professional race are also expressing a much greater variety, meaning that a large number of the riders in the peloton are cycling at a significantly higher intensity than the top riders.

    In a male pro cycling race, almost all riders in the peloton ride at a fairly low intensity.
  • When it comes to differences in training between male and female riders, the biggest difference is in volume.

    The male pro riders are laying down a much greater number of training hours, while the total volume of intensity between male and females are quite the same.

Advice to amateur athletes

30:35 -

  • My main advice to amateur athletes would be to have a more polarized approach to their training, many age groupers spend a lot of time in zone 3 or ”the grey zone” instead of either doing really low intensity (zone 1 and 2) or very high intensity (zone 5).

    I see many amateurs going way too hard on their endurance rides, the long and easy ride should predominantly be in zone 1 with maybe a little bit of zone 2 in there.


35:45 -

  • I think periodization is pretty important.

    When I was at Sunweb, in November and December we did a lot of long endurance rides and stayed away from too much intensity at this time of the year since it tends to get the athletes in really good shape too early (January and February) instead of in March and April when the most important races (the most prestigious classics) take place.
  • Consequently, we implement a fairly traditional approach to periodization with an increased intensity as closer to the season we get, hence, in January we did a fair amount of zone 3 work and then closer to March and April we increased intensity towards zone 5.

    Looking at an average training week in late January and February, we typically had one long and easy endurance rides per week, but all the other rides comprised some kind of intensity.


43:25 -

  • I was never the trainer for Team Sunweb, but I know that the coaches were playing around with riding at different cadences a lot, addressing both ends of the spectrum (high and low cadence).

    Low cadence emphasizes the strength component and high cadence brings the heart rate up much more quickly and gets the riders accustomed to the peloton where they often ride at high cadence.


44:20 -

  • Nutrition is a huge part of professional cycling.

    The training response will differ depending on what the athlete eat before, during and after a training session.

    By reducing carbohydrate intake before a session, the fat metabolism is stimulated to a higher extent, which is a very important factor during long races.

    On the other hand, athletes need to be good at utilizing carbohydrates during high intensity sessions as well.

    During races we keep track on the energy expenditure and make sure that the athletes are getting in the right amount of energy before, during (usually 90 grams of carbohydrate per hour) and after the race.

Performance testing

47:55 -

  • As Team Sunweb had quite a low budget, we only conducted power curve field tests to get a power profile of the cyclist.

    The advantage with this is that it is very easy to conduct but as of now, I would have preferred to also use laboratory testing due to its higher degree of certainty and the ability to see every athlete’s strengths and weaknesses in a more telling manner.

Top three advice to age group athletes

50:40 -

  1. Be diligent when it comes to your nutrition (especially) in racing and training.
  2. Don’t train too much in the ”grey zone”/zone 3, keep the effort easy when supposed to and really high during the high intensity sessions.
  3. Get a power meter!

Rapid fire questions

51:40 -

  • What is your favorite book, blog or resource related to endurance sports? The data that I have collected from the cyclists in Team Sunweb.
  • What is your favorite piece of gear or equipment? The power meter.
  • What is a personal habit that has helped you achieve success? You need to have some luck but also see the opportunities out there and take them, for instance I worked one year for free at Team Sunweb just to show them what a sport scientist could do for them.


Mikael Eriksson

I am a full-time triathlon coach, founder of Scientific Triathlon, and host of the top-rated podcast That Triathlon Show. I am from Finland but live in Lisbon, Portugal.

Please contact me if you have feedback on the podcast or want to make suggestions for improvement or send in a question for a Q&A episode.

If you are a long-time listener and appreciate the value the podcast brings, please consider taking a couple of minutes for leaving a rating and review on iTunes/Apple Podcasts, or wherever else you can think of leaving a rating and review.

{"email":"Email address invalid","url":"Website address invalid","required":"Required field missing"}

Explore our products and services