cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Try the Materials Informatics Toolkit, which is designed to easily handle SMILES data. This and other helpful add-ins are available in the JMP® Marketplace
Choose Language Hide Translation Bar
Can today’s cyclists be trusted or are they pedaling through deceit? A statistical analysis of cheating in the Tour de France's history

As the summer approaches, so does the 111th edition of the Tour de France, which begins on June 29 in Florence, Italy. For those who are not familiar with the Tour de France (aka the Tour), it began in 1903 and takes place annually in late June thru mid-July. The race is broken up into stages that are ridden daily throughout France and surrounding countries. The Tour is one of the most grueling athletic competitions, and the winners are some of the most celebrated and famous athletes on the planet. The quest for Tour de France glory has driven cyclists to transcend the boundaries of human potential. Transcending those boundaries has led some cyclists down the unscrupulous path of winning at any cost. This blog looks at past winners – both scrupulous and unscrupulous – to determine if today’s cyclists can be trusted.

In July 2023, the 110th edition of the Tour de France finished with Danish rider Jonas Vingegaard winning the race, which began in Bilbao, Spain, and ended in Paris after 21 stages. The map of the race is shown below.

 

Figure 1.png

 

In winning the race, Vingegaard completed the 3,404-kilometer route in 82 hours, 5 minutes, and 42 seconds. He bested his closest rival, Tadej Pogacar from Slovenia, by 7 minutes and 29 seconds. My colleague and fellow cycling fan Peter Polito mentioned that the margin of victory was remarkably small, even though most commentary on the victory asserted it was a convincing blowout win by Vingegaard. To sum it up, Vingegaard won by 449 seconds after ridding over 82 hours, which translates to about ~0.15% margin of victory. In other competitions, this would be considered a photo finish; it is equivalent to winning the 100-meter dash by a little over a hundredth of a second. How does this compare to time gaps in past Tours?

 

Figure 2.png

This graph shows the time gaps between the winner and second-place rider throughout the years. The time gap in 2023 is the largest gap since 2014. In fact, the gap between first and second place has only been larger three times since 1984. In Tour de France terms, 2023 really was a blowout win for Vingegaard.

However, this result brought up a few questions about the performance:

  1. Was this version of the Tour longer than usual, thus resulting in bigger time gaps?
  2. How did Vingegaard’s speed over the course of the Tour compare to speeds in the past?
  3. Should fans question the validity of this win with so much controversy in cycling history?

The first question is very easy to answer. In 2023, the race was 3,404 km, which is slightly longer than the 3,328 km of 2022 and 3,383 km of 2021; it was shorter than the 3,483-km race in 2020. While length might be playing a role, it does not appear to be the major driver. It did make me wonder about the history of the distances, as it appears the gap between first and second place finishes had been trending downward. If we look at the average distance for a Tour de France by decade, it is clear that these races have become shorter over the years.

 

Figure 3.png

Let’s look at the second question. Was Vingegaard faster than expected when looking at past results? The graph below shows the average speed of the Tour de France winner each year. The race was not timed from 1905 to 1912; rather, points were given based on finishing position (there is more on the reasoning for that below). There was no race during both World Wars. The doping era (1999-2006) is highlighted, when all the winners during that span were found to have doped so their wins were vacated. The Tour has had many more doping issues, which we will delve into later.

Figure 4.png

Looking at the graph, Vingegaard had the third-fastest average pace of a Tour winner throughout its history, only bested by himself last year and Lance Armstrong in 2005. We will look at this in a little more detail when we answer the last question.

There is a strong trend upward in speed throughout the years. There are plenty of reasons to believe that speeds have improved naturally: improved roads, improved bikes, shorter Tour distances, support vehicles, and sponsored teams that spend all year training for this event. If you look at the picture below from the 1927 Tour de France, you can see a lot has changed over the last hundred years of cycling.

 

Julien Vervaecke and Maurice Geldhof are smoking during the 1927 Tour de France. Gustaaf van Slembrouck is lightning the cigarette.Julien Vervaecke and Maurice Geldhof are smoking during the 1927 Tour de France. Gustaaf van Slembrouck is lightning the cigarette.

While there are many natural reasons for the increased Tour speeds, the race is also full of controversy and scandal, which means, unfortunately, every win faces extra scrutiny. If you are not familiar with the history of the Tour, the most infamous villain in the cycling world is Lance Armstrong. He won the Tour a record seven times in a row (1999-2005) before he was found guilty of doping and thus stripped of all those titles. Armstrong is not the only cyclist to use doping to gain an advantage. 2006 winner Floyd Landis and 2007 third-place finisher Levi Leipheimer were also found guilty and stripped of their titles, as well as a long list of riders that were caught with positive tests during the races. It is safe to say that the Tour in the late ’90s and early 2000s was heavily influenced by doping.

Doping in the Tour de France started long before the 1990s. Many doping scandals have plagued the Tour through the years, including wide use of alcohol, ether, amphetamines, opioids, steroids, human growth hormone (HGH), and erythropoietin (EPO). In fact, up until the 1966 Tour, it was not illegal to dope in the race and no testing was performed. Testing became mandatory after British rider Tom Simpson died during the 1967 Tour de France with vials of amphetamines found on his body https://en.wikipedia.org/wiki/Tom_Simpson. Even with the advent of testing and the death of a fellow rider, the doping continued, with many of the champions throughout the years being caught or linked to doping. It has been a constant struggle to police doping. Since 1966, nearly half of the Tour winners (25 of the 56 winners) admitted to doping or failed doping tests. https://en.wikipedia.org/wiki/Doping_at_the_Tour_de_France

There has been controversy in the Tour de France long before Lance Armstrong or the doping of champions in the 1960s. All the way back in 1904, all top four finishers and 12 of the 27 finishers were disqualified. The reasons for specific disqualifications were never made public, but allegations of illegal use of cars, trains, and spectator interference have been cited (https://www.rediff.com/sports/2004/jul/06tour.htm). The 1904 Tour was so bad that it led to a change in the rules, creating a point-based system instead of a timed event. This system lasted until the 1912 Tour. The modern-day Tour de France still has two competitions that harken back to the point-based system of those early Tours. The Sprint and King of the Mountain competitions (which award green and polka-dotted jerseys) are based on points awarded throughout the race.

Now let’s look at the last question. Should cycling fans be skeptical of Vingegaard’s 2023 win?

I have been following cycling for years, and it is very hard to give riders the benefit of the doubt. Looking at the winners’ speeds dating back to 1966 when testing for doping started, there is a clear trend. https://bikeraceinfo.com/tdf/tdfstats.html.

Figure 6.png

Since 1966, the average speed of Tour de France winners has increased ~0.12 km/hr each year. The average winning speed today is ~6.8km/hr faster than in 1966. Seeing faster speeds each year is an expected trend. In this graph, the champions who have been caught doping are represented with a red plus sign. If we look at Vingegaard’s speed over the last two years, it doesn’t seem to raise any suspicion. His 2022 ride and 2023 ride are both close to the trend line of improved speed throughout the years. If we look at the trends of the known dopers vs. the clean riders, the story is a little different. The 2023 ride for Vingegaard seems to be well-aligned with clean riders of the past. The 2022 ride is much closer to the known doping riders trend line.

Figure 7.png

 

This is not to say that Vingegaard was doping in 2022 and not doping in 2023. It seems like a very unlikely scenario that a rider would dope to win one year, decide not to dope the following year, and then win again. It appears that Vingegaard is not doping to win the race and is one of the greatest cyclists of all time.

It appears that the 2023 Tour de France was a dominant win for Vingegaard, who had the third-greatest margin of victory since 1984. In winning the Tour, Vingegaard had an average speed of 41.4 km/hr, which translates to the third-fastest Tour of all time. Despite the history of doping in the Tour and the fast pace of Vingegaard, the 2023 Tour seems to be a believable win for a clean rider. As a jaded cycling fan, it wouldn’t surprise me if the recent Tour de France cyclists followed the long tradition of cheating to win the race, but for now it does not appear that is the case. Looking forward to this year’s race!

 

 

 

 

 

Last Modified: Jun 3, 2024 9:00 AM
Comments
eclaassen
Staff

For some reason the 1989 Tour hit my radar (I'd never even heard of competitive cycling, I don't think, being a teenager in middle America), but I became absolutely enamored with that competition between LeMond and Fignon. I'm pretty sure I sat at our kitchen table watching the final time trial on the little TV we had there until the final 8 second margin of victory was set. All those hours, and only 8 seconds difference in the end! 7 and a half minutes is an eternity in comparison!

Peter_Polito
Staff

This is a fantastic read, @Peter_Hersh! I would posit one other opinion on Vingegaard's '23 win and that was Pogacar's broken wrist. I think there is some credence to the suggestion that he wasn't able to train as much as he would have liked having broken his wrist in the spring classics. The large margin of victory last year really came down to two stages--the time trial and the mountain stage (was it the next day?). During the time trial, Jonas was out of this world and on top of that, Tadej made a bike change. In the mountain stage, Vingegaard didn't even win the stage, he was more than a minute back on, I think it was Jai Hindley, but regardless, Pogacar broke and finished something like 4 minutes back. Had Pogacar fought him on the mountain stage the margin of victory would have been more than halved. 

 

But what is real interesting is the roles are reversed this year. Vingegaard's crash was massive in the Basque Tour and if he is able to line up at the start of the Tour, he'll be in a worse position than Pogacar last year. Meanwhile, Pogacar just turned in one of the most dominant Giro's in recent memory with a 10 minute advantage over second place. I'm getting off topic now but if Vingegaard lines up, we'll get to watch him, Pogacar, Evenpeol, and Roglic all go at. It's gonna be fantastic. 

Peter_Hersh
Staff

@eclaassen oh Greg LeMond was the reason I started watching cycling.  If you haven't seen it I would recommend watching slaying the badger it is a great docu-series on LeMond and how he won his first tour taking down Hinault.

eclaassen
Staff

Thanks for the recommendation, @Peter_Hersh ! I'll have to track it down.

Peter_Hersh
Staff

@Peter_Polito Yes that is a great point on Pogacar's injured wrist it is pretty amazing that even with that he finished second.  The gap between the two in 2022 was less than 3 minutes with both relatively healthy.  If anything Progacar's win in the Giro gives me more pause that he will win this year's tour no rider has won both events in the same year since Marco Pantani did it in 1998.  From his EPO/blood doping scandal I think it is safe to assume Pantani was not clean in winning those races.  It seems like these grand tours are so challenging that you have to time it up to peak at the right time.  Then again no one has won the Giro that dominantly since the 1960s.  Anyway it should be a great race I hope that Vingegaard is close to peak form.  I am also hoping to see the Sepp Kuss ride a good race.  He might be in contention if Vingegaard is not close to 100% as they are teammates.

Peter_Hersh
Staff

Race Starts Saturday.  Looking forward to this year.

czcarroll
Staff

Thanks Peter for that very interesting analysis.  In Ireland we have a stage race called the Ras Tailteann http://www.rastailteann.com/event/fbdras_1323.shtml.  I rode it five times in the 80's when it was  9 days  Saturday to Sunday (without a rest day) , so have some skin in the game from a rider perspective, even as an also ran.  Average speeds in the 80's were around 40 KPH .  There was always one day when the riders agreed  to take it easy. There is no doubt the average speeds are going up, driven by sheer professionalism and data being available. 

Would be interesting to look at all the time trial stages over the years in the TDF.  

Peter_Hersh
Staff

@czcarroll I agree the time trails by themselves would be really interesting.  Also looking forward to the final stage of this years tour Monaco to Nice a 35km individual time trial.  It will be very strange to not have the tour end up in Paris, but could be very exciting if the race is close.

MachineFinch739
Staff

This is a fantastic article and analysis. Thank you! I was fortunate enough to see the final stage of the Tour de France last year and cheer on Jumbo Visma. Seeing these professional cyclists in person drives home how incredibly fast they ride. I hope to see the final stage in Nice this year. Great to see Jonas back on his bike after such a horrific crash. Sad to see Sepp Kuss out of the Tour. 

Peter_Hersh
Staff

@MachineFinch739, So cool seeing a stage is on my bucket list.  The ending in Nice should be amazing this year.  So sad about Sepp Kuss really seemed like he had a chance to have a top 10 or even better. 

Mandy_Chambers
Staff

this is so interesting @Peter_Hersh !  your posts are pretty remarkable and very interesting and educational. Love the use of JMP too of course!

I cycle myself.. but not like these folks!   Will be fun to watch all the summer games and events! 

Peter_Hersh
Staff

@Mandy_Chambers Thanks I am glad you enjoyed it.

Françoise
Level V

bonjour,

 

merci pour votre article.

 

quand on voit la photo de 1927 et les coureurs aujourd'hui, il y a un écart dans leurs poids !

je pense qu'en 1927, ils avaient dans leur sac de ravitaillement une bouteille de vin ! 

l'alimentation des sportifs a changé depuis.

 

le duel entre les 2 favoris a commencé hier dans la dernière ascension !

 

à surveiller: mardi dans le Galibier !!

haleyyaremych
Staff

This is so interesting! Thank you for sharing this @Peter_Hersh. I'm also an avid cycling fan, and had always wondered if anyone had thought to conduct an analysis like this, especially when these questions started swirling after Jonas's wildly good time trial at the 2023 Tour. Looking forward to seeing how the race unfolds this year! 

Benonolan22
Staff

Really nice analysis of the tour on a subject cycling fans no doubt think about often! Vingegaard certainly had a dominant team to put him in a position to win in 2022 and 2023 but this year all of the Jumbo team have been involved in crashes in the lead up to the tour so this should have an impact, Wout van Aert is +20 minutes back on the leaders for example and has been seen as a potential GC winner for years. Pogacars UAE team are seen to have strengthened also so it should be close. The doping trend line seems interesting but not conclusive enough to prove anything untoward, I have always suspected Vingegaards performance in the time trial in 2023 which seemed an order of magnitude above the rest of the pack, his power to weight ratio for the 2nd part of the TT climb was 7.4w per KG vs 6.7W per KG for pogacar for example, which is significantly higher over 13 minutes despite being on a TT bike with Pogacar having changed to a lighter bike. Pogacar had crashed earlier in the season though so could it just be a bad day + wear and tear after 2 weeks on the tour, who knows. Could other factors like sleep, nutrition and bike technology account for such a difference? Interested to see the speed improvement that you believe can be attributed to bike tech and improved aerodynamics etc. 

Peter_Hersh
Staff

@Françoise ,   

Oui, un début de course incroyable.

Pogacar semble être dans une forme incroyable.

La chute de Vingegaard était peut-être dans son esprit puisqu'il a perdu 30 secondes dans la descente du Col du Galibier.

 

See how Google Translate did.

Peter_Hersh
Staff

@haleyyaremych and @Benonolan22,  Yes that time trial last year was a huge win for Jonas.  I personally think it had to do with some excellent team strategy using a unique bike for time trialing that actually had two chain rings instead of the traditional single also no disc for the rear wheel and no water bottle to save weight.  This enabled him to climb the final section of the course and not have to change bikes staying on the more aerodynamic TT bike.  I think the main difference was Pogacar coming off an injury.  The first time trial of this year's tour is going on as I write this.  It should be interesting to see how everyone performs.  Really great tour so far Cavendish for number 35 also love that they had a mountain stage at stage 4 and a time trial at stage 7.  Kudos to the race organizers.