Zoo Station to Paris: How the three toughest Tours de France may have spurred the evolution of doping.
The 1999 Tour de France was dubbed the Race of Renewal by its organizers, in a desperate attempt to distance itself from the Festina Affair of the prior year. On July 8, 1998, one of the most powerful teams in cycling, the French-based Festina squad, was caught trying to cross the French-Belgian border with a mobile pharmacy of illegal drugs including hundreds of vials of anabolic steroids, cortisone, and Erythropoietin (EPO). The central figure in 1999 was Lance Armstrong, returning from a near fatal battle with testicular cancer that took him away from racing for the better part of three years between 1996–1999. The perceived ease of Armstrong’s relatively large margin of victory, 7:37, immediately aroused suspicions. Namely, Irish sports journalist, David Walsh, who was instrumental in Armstrong‘s downfall, cited the fact that instead of getting slower, in what was supposed to be a new era of drug-free competition, the average speed was actually the fastest in the history of the race to that point.
Although the average speed of the field in 1999 was faster than during the 1998 Tour (41.43 km/h vs. 41.08 km/h), it was also a shorter course (3870 km vs. 3875 km), with better weather conditions, and no late-night raids of team hotel rooms, or hours-long police interrogations. Indeed, if we look at the relationship between the average speed and the distance of the race over the entire history of the Tour de France (Figure 1), there is a clear inverse relationship, as the Tour has gotten progressively faster over the years, it has also gotten shorter.
Comparing race speeds from year to year is not a good measure of athletic performance or the difficulty of the racecourse. Because of this, average speed is a poor predictor for the presence or absence of performance enhancing drugs (PED). Race speed is influenced by several factors including team tactics, weather conditions, and technological advancements in training, recovery, nutrition, and equipment, and finally, the distance of the race itself.
That said, there was a well-publicized effort on the part of the Tour organizers, Amaury Sport Organisation (ASO) to make the races of the early 2000’s less harrowing than those of the 1990’s. This motivation to deter doping by reducing the perceived need for PEDs raises the obvious question, is there evidence in the historical record of the Tour de France that shows a correlation between changes in the difficulty of the race and the introduction of new doping methods?
To answer this question, an objective measure of race difficulty was needed. For this we turned to the element of the Tour de France that makes it a spectator sport, the mountains. An analysis of television viewership in ten European countries including France, Germany, and Italy, during the 2019 Tour de France shows that the mountain stages drew 49% more viewers than did the flat stages [1]. To determine how “hard” a race is, you need to look at how much climbing the cyclists must do. Tour organizers began using a category system in 1947, that organized climbs into one of five groups with Category 4 being the easiest and Hors Category (HC) being the toughest. But this system is a subjective measure that has changed numerous times since its inception. For example, if a Category 1 climb happens at the end of a long stage, it is considered an HC climb for that day. To generate an objective measure that allows for apples-to-apples comparisons between any two randomly selected climbs from any given year, a novel ranking system was designed. The Climb Score is a unitless measure calculated as the product of the average grade of the climb and its horizontal length in kilometers, from its base to its peak. For example, the average grade of Mont Ventoux is 7.5% and it covers 21.4 km, giving it a Climb Score of 160.5 (7.5% X 21.4 km = 160.5). Like the category system used by AMS, climbs were assigned to one of five categories, (Table 1), with Category 4 being the easiest and Above Category (AC) being the toughest.
A better indicator of the impact of PEDs on race performance can be seen when comparing the Climb Score to the average speed over the 118-year history of the race. Although it would be expected that during races with more climbing, the average speed would decrease, the exact opposite is true (Figure 2). In general, despite climbing more from year to year, the average speed has continually increased. Although some of this can be attributed to improvements in equipment, training, nutrition, and roads, some of these improvements are certainly the result of PEDs. Of particular interest are three spikes in Climb Score in 1951, 1973, and 1987 (Figure 3), each occurred no more than three years before the introduction of a new PED, amphetamines in 1952, cortisone in 1974, and EPO in 1990.
Amphetamines were first synthesized around 1887 and were first marketed as decongestants in 1933. Amphetamines have two modes of action as PEDs, they stimulate the central nervous system and reduce signals of pain and fatigue. The first confirmed use of amphetamines occurred in 1952. Italian legend and two-time Tour winner (1949 and 1952), Fausto Coppi admitted in 1952 that he used amphetamines during that year’s Tour after finishing a disappointing 10th in the 1951 race [2], a route with a Climb Score of 2240, 22% higher than any race before.
Cortisone is an anti-inflammatory steroid hormone that was first produced commercially in 1950. Like amphetamines, cortisone has a mild stimulatory effect, increasing energy and reducing pain. But its primary mode of action is by reducing recovery time and inflammation. A 1977 admission by another two-time winner (1975 and 1977), Bernard Thevenet [3], pins the first usage of these drugs at no later than 1974, the year after the next big jump in Climb Scores (2234).
The history of doping in the Tour de France is divided into two eras, before and after the introduction of EPO into the Peloton. The ESPN, 30 for 30 documentary, “Lance”, provides an excellent perspective of the evolution of doping from the use of what Lance Armstrong labelled “Low octane” doping techniques including amphetamines and cortisone, to the “High octane, rocket fuel,” that was EPO. Whereas amphetamines and cortisone provided improvements in performance that are estimated to be between 1–2%, EPO improved performance by 8–10%, a staggering value that turned formerly unknown domestiques into Grand Tour contenders from one year to the next.
EPO stimulates the bone marrow to produce more red blood cells, boosting the delivery of oxygen to the muscles during high intensity aerobic exercise. EPO was approved by the FDA for commercial use in 1989, a year before its first confirmed use in the Tour de France. In 2009, Dutch cyclist, Steven Rooks who finished second in the 1988 Tour admitted to using EPO starting at the end of the 1989 season [4]. He rode the Tour five more times between 1990–1994. As with the introduction of amphetamines and cortisone, the arrival of EPO was preceded by a large spike in Climb Score. The 1987 race had a score of 2860, 28% higher than the value for the 1951 Tour, and an unbelievable 76% higher than the average Climb Score for all Tours between 1950 and 1986 (1623). But the Climb Score can be inflated by including several smaller, relatively easy climbs in succession. That wasn’t the case in 1987. The average number of AC climbs in the Tours between 1950–1986 was approximately 8. The 1987 Tour de France had 19 AC level climbs, more than any other race before or after (Figure 4).
The 1987 Tour de France represents a confluence of several factors. In honor of the 750th anniversary of its founding, the race began in West Berlin, on the other side of the Iron Curtain. In addition to having the highest Climb Score and the most AC climbs of any Tour, the 1987 race had 25 stages strung over 26 days, more than any Tour before or since. It was 4231 kilometers in length, the longest race since 1970, and longer than any race since.
The counterargument to this hypothesis is that PEDs will be used once they become available. But that narrative doesn’t fit the chronology of the introduction of either amphetamines or cortisone into the Peloton, both of which had been available for decades before their uses became widespread on the Tour. Even with EPO, this argument is shaky. Increasing the number of red blood cells as a method of athletic performance enhancement didn’t begin with EPO, but rather, with blood transfusion. This practice has been used in sports since the late 1960’s. Indeed, 1980 Tour de France winner, Joop Zoetemelk, admitted to receiving a blood transfusion during the 1976 Tour [5]. Interestingly, once a reliable blood test for EPO was developed around 2000, most cyclists who had been using it turned back to the old practice of reinfusing their own blood during the race. EPO just made it faster, easier, and much less dangerous to increase blood oxygen levels.
During the drug-fueled free-for-all that marked the two decades from 1990 to 2010, when numerous athletes either tested positive or admitted to doping, the Climb Scores stayed relatively stable, averaging 1928. The one anomaly was the 1993 Tour (2420), when all three podium finishers were implicated in the use of EPO at some point during, or after, their careers. But, of concern is what can be seen in the Climb Score during the last decade, from 2011–2021. During this time, the Climb Score has steadily increased, averaging 2254, with significant jumps in 2016 (2555) and 2020 (2836). The 2020 Tour de France marked the first time in many years that there was a steady chorus of suspicion about the outcome. Many people associated with the race, both competitors and observers, felt that the domination exhibited by the Jumbo Visma team was reminiscent of how Lance Armstrong’s U.S. Postal-Discovery Channel teams set blistering paces at the front of the Peloton for seven years. Although no one tested positive for PEDs during the 2020 race, absence of failed doping controls is not the same as a clean race. No one riding for U.S. Postal-Discovery Channel ever tested positive for PEDs while an active member of those teams. But by 2012, no fewer than 12 former members had admitted to EPO use between 1998–2006. Race organizers should consider the potential consequences of making the most beautiful bicycle race in the world into a gladiatorial battle of attrition and suffering. Everyone wants to see the best cyclists in the world duke it out over the Aubisque and the Tourmalet (a name that translates to “Evil Trip”), but we should take heed from the lessons of history that show us what can happen when we push our athletes too far. As is the case with many things in life, more is not always better.
References:
1. Reeth, D.D.V., Road Cycling TV Viewing Report 2019. 2019, Katholieke Universiteit Leuven: Brussels, Belgium.
2. Format. Quando Volava l’Airone. 1952.
3. Mondenard, J.-P.d., Dopage — l’imposture des performances. 2000, Paris, France: CHIRON.
4. Kröner, H., Rooks admits to EPO use, in Cycling News. 2009.
5. McKay, F., A history on blood transfusions in cycling, part 2, in Cycling Newa. 2013.