I try to keep on top of the sports science literature on skiing. Most of it does not have direct application to ski technique, but here’s one that does:
R. Pozzo, A. Canclini, G. Baroni, D. Benedetti and S. D’Ottavio. 3-D kinematic and kinetics analysis of slalom in elite skiers at the Bormio World Cup ski finals in 2008. In E. Muller, editor, Science and Skiing V (Meyer & Meyer, London), 2012, pp 355-363.
This paper was presented at the Fifth International Congress on Science and Skiing, which was held at St. Christoph am Arlberg, Tyrol, Austria, in December 2010.
The full paper is not available online, but the first five pages (355-359), which contain most of the key results, are available through Google preview:
http://books.google.com/books?id=zZvxTQ ... 08&f=false
What makes this study so interesting is that the authors weren’t simply asking how WC racers ski. Rather, they used 3D motion capture and timing to determine what movement characteristics separate the best-of-the-best from the best. Specifically, they set up a set of cameras covering five gates in the middle of the first run at the 2008 World Cup Finals in Bormio, and analyzed the performance, over that section, of the top 8 skiers in that run (top eight for the run, not top eight for the race). In order of run placing, from first to eighth, they were: Manfred Moelgg, Felix Neurether, Reinfreid Herbst, Baptiste Grange, Marcel Hirscher, Marc Bethod, Ivica Kostelic, and Benjamin Raich. [Incidentally, the three top overall SL skiers for ’07-’08 were Moelgg (531 pts.), Grange (512 pts.) and Herbst (450 pts.).] It turns out that two fastest skiers over that section (Neurether and Moelgg), as well as the two slowest over that section (Raich and Kostelic), were also the fastest and slowest for the run, suggesting that this section was representative of overall run performance. [I was unable to determine if the run finish order for all eight was the same as the performance order in this section, since the authors don't provide numerical velocity values for all eight, just a bar graph -- the paper is not as well-presented as it could be.]
Using their motion-capture system, the authors measured (or calculated from those measurements) the following over the entire section: the velocity and path of the skiers' centers of gravity (CG); the angles of inclination of both legs of each skier (defined by the lines from ankle to hip joint) to the snow surface; and the angles of both knee joints. Here are their key findings:
1) The CGs of the fastest skiers tended to show less vertical displacement (i.e., less up-and-down motion -- only 0.5 m – 0.6 m between the highest and lowest points, which presumably are the transition and the point of maximum angulation, respectively). Relatedly, their CGs also had slower lowering velocities, i.e., the rate at which their CGs moved closer to the snow between the transition and the point of maximum angulation. The latter result is expected, since their CGs, having smaller vertical displacements, had less distance to travel. But another possible contributing factor (the authors didn’t comment on this) could be that the faster skiers also spent more time in the first phase of the turn (from initiation to the point of maximum angulation).
Interestingly, the rates at which the skiers raised their CGs, in coming out of the turn, didn’t show a consistent pattern.
2) The CGs of the faster skiers tended to have the largest minimum turning radii (i.e., they didn’t go straight at the gate and turn sharply) (note this is the turning radius of their CGs, not of their skis). Correspondingly, they experienced the lowest maximum centripetal accelerations. Likely because of this, their skis had the lowest maximum ground reaction forces (GRFs). E.g., Neurether and Raich didn’t differ much in their minimum GRFs (1.97 BW vs. 2.17 BW, respectively; BW is body weight), but their maximum GRFs were 3.33 BW and 4.04 BW, respectively. I.e., Raich was briefly pulling 4 g’s. This value is consistent with measurements found in earlier studies. A lower GRF makes for less frictional losses and a ski that is less likely to break out of its carve. It also puts less physical demand on the skier.
3) The knee angle measurements were presented in a somewhat confusing manner, but the faster skiers had a more extended stance leg (less bend in the knee) at the midpoint of the turn (i.e., when the turning radius was at a minimum), at least for the two turns where the slope fell away to that side (these two turns also showed the most leg inclination). They also measured angular velocities of the knee joint (i.e., how fast the knee flexed and extended), but didn’t appear to find a clear trend.
4) By contrast with 1-3, above, the authors found that the maximum angles of inclination of both the inside and outside legs, which they presumed corresponded to the angles of the skis to the snow, did not vary significantly within this group.
Given that they captured the inclination angles and knee joint angles of both the inside and outside legs, I assume they also have the video data to measure stance width at the transition. It’s too bad they didn’t report this – it would have been interesting to see if it showed a correlation to performance. Maybe I’ll see if I can track down one of the authors and ask them this.
It should be emphasized that none of these athletes were skiing poorly – note that even the worst of this lot was the eighth best in the world on that run! From that perspective, they all skied at an extraordinarily high level. Yet, as is obvious from the race times, even among such an elite group one finds important differences.