In stage 8 of the Tour de France (2016), probably the most remarkable event was the way in which Chris Froome descended the Peyresourde towards the finish line, took a 13” lead on Nairo Quintana, Adam Yates, Bauke Mollema and his other competitors, from which he broke away just before the top of the Peyresourde, and won the stage. Figure 1 shows Chris Froome during his descent, sitting on the top tube, with his chest on the handlebar. During the descent he was alternatively pedaling and holding the legs fixed. The descent covered 15.5 km, from an altitude of 1569 m to 632 m. Froome reached an average speed of 62.5 km/h and a maximum speed of 90.9 km/h.
This rather uncommon descent position received huge media coverage, and many have attributed the 13” lead to the more aerodynamic position of Froome compared to that of his competitors. While many have applauded Team Sky and Chris Froome for adding spice to the race, others have criticized them for implicitly having promoted this rather dangerous position in professional (and maybe even recreational) cycling. Indeed, this position puts the majority of the weight on the front wheel and severely reduces steering capability.
However, the question is: was it really this position on the bike that led to the win of the stage and the 13” lead? We try to give the answer below.
In the past week, researchers from Eindhoven University of Technology in the Netherlands, KU Leuven and University of Liege in Belgium and ANSYS International have joined forces to aerodynamically analyze the “Froome position”, as we will call it, and compare it to other positions.
First, scanning was performed to provide the geometry of a cyclist in four different positions: first the two main positions of interest: the “Froome position” and a safer but also very aerodynamic position, with the cyclist on the saddle, which we will simply call “safer descent position”. These two positions are shown in Figure 2a and 2b. In addition two reference positions were added: the time trial position with arms stretched and hands and arms on the time trial bar, generally considered to be most aerodynamic, and the upright position with hands on top of the handlebar, generally considered to be least aerodynamic. Figure 2c and 2d show the latter two positions.
Next, numerical simulations with Computational Fluid Dynamics (CFD) with ANSYS Fluent were performed. First, high-resolution computational grids were generated for the four cyclist positions. The steady Reynolds-averaged Navier-Stokes equations were solved with the standard k-epsilon model for closure. Near-wall modeling was performed with scalable wall functions. This methodology has been extensively validated by us in other cycling aerodynamics studies in the past using wind tunnel measurements (Defraeye et al. 2010, Blocken et al. 2013, Blocken and Toparlar 2015, Blocken et al. 2016).
The results are presented in Figure 3, which shows the ranking of the positions from most to least aerodynamic. The subfigures in the left column show the pressure coefficient contours on the cyclist bodies, the middle column shows quantitative information on the increase in aerodynamic drag relative to the time trial position, and the right column shows the photographs of the four positions. The following observations can be made. The “Froome” position is not aerodynamically better, but slightly worse than the “safer descent position”, albeit with a very slight margin of 0.6% in terms of air resistance (drag force). The “safer descent position” is only slightly worse than the time trial position. Note however that the time trial cyclist is sitting quite upright and his aerodynamic position is less favorable. The upright position is clearly the worst, with 19.9% more air resistance than the time trial position.
The next logical question is: what then was the reason why Chris Froome took a 13” lead in his descent? The first part of the answer is stating that this question is poorly posed. Indeed, he did not take the full 13” lead in only the descent, but he already took a rather big lead by accelerating before the top of the Peyresourde, as shown in Figure 4. And if you start descending while others are still climbing, you can develop quite a lead at the finish line.
Actually, it is very remarkable that his competitors made the mistake to allow him to take this lead. Froome used the element of surprise, in a masterful way. Next, he took additional time in the largest part of the descent. Given the results of the CFD simulations, this would mean that the position of the leader chaser, Nairo Quintana included, was less aerodynamic than the “Froome position”, hence also less aerodynamic than the “safer descent position” in Figure 3. Indeed, Figure 5 shows the very limited available footage of the chasers, where the lead chaser seems to be in a more upright position, certainly not as aerodynamic as the “safer descent position”, which corresponds to a horizontal position of the back. Finally, at the very end of the race, the chasers hesitated for a while, after which they took some time back, yielding the final gap of 13”.
Answering the question in the title: “Fantastic downhill from Chris Froome; but was it worth it? The answer is both “YES” and “NO”, which indicates that also this question was poorly posed.
YES, because it was worth to take the lead in the descent and win the stage this way, for several reasons. Froome had the reputation of being less good downhill, and he provided quite the spectacle here to demonstrate his skills. In addition, Team Sky had imposed on itself the mission to add spice to the race, which they did successfully at the end of stage 8. Last but not least, a Tour de France can be won and lost based on a few seconds – although this will probably not be the case this year – and Froome did not yet wear the yellow jersey at the start of stage 8.
NO, because the “Froome position” – as opposed to common perception – is not more aerodynamic than a more common and safer position with the cyclist sitting on the saddle and bent forward with the head close to the handlebar. The distance between cyclist body and bicycle frame and between cyclist body and the ground is not a factor of importance in terms of aerodynamics.
Finally, some brief calculations can be made based on the CFD simulation results to link percentages in drag reduction to time differences. Assuming an average speed of 62.5 km/h over the 15.5 descent, 13” corresponds to an aerodynamic drag difference of 2.5 to 3%. This seems to make sense and in line with the results in Figure 3 given the fact that the position of the lead chaser in Figure 5 is more upright than the “safer descent position”.
- Interestingly, it appears that also others were inspired by the “Froome position” to perform an investigation. Although less scientific, the field test by the Global Cycling Network (GCN) is well presented and… it provides similar conclusions as our study. Note that there was no correspondence between the GCN and our team and that we independently arrived at similar conclusions:
- Although we called it the “Froome position”, this position was already used by other riders in the past, including Tony Martin in the Vuelta of 2013.
- The cyclist that was scanned and involved in the simulations was neither Chris Froome, nor any other professional cyclist acting in the Tour de France. The cyclist we scanned has similar body height and weight as Chris Froome, but a slightly different position on the bike, with a more flat back, as opposed to the larger curvature in the back of Chris Froome. Therefore, it is not unlikely that the use of other cyclist bodies would result in slightly different percentages than shown in Figure 3. However, we believe that the “Froome position”, taken be any athlete, will generally be less aerodynamic than the safer descent position.
This research was performed by a team composed of:
- Prof.dr.ir. Bert Blocken, Eindhoven University, the Netherlands and KU Leuven, Belgium (www.urbanphysics.net)
- Ir. Thijs van Druenen, Eindhoven University of Technology, the Netherlands
- Ir. Yasin Toparlar, Eindhoven University of Technology, the Netherlands
- Dr.ir. Thomas Andrianne, Université de Liege, Belgium
- Ir. Thierry Marchal, ANSYS International