www.sciencemag.org/cgi/content/full/325/5938/289/DC1
Supporting Online Material for
Modern Riding Style Improves Horse Racing Times
Thilo Pfau,* Andrew Spence, Sandra Starke, Marta Ferrari, Alan Wilson *To whom correspondence should be addressed. E-mail: tpfau@rvc.ac.uk Published 17 July 2009 Science 325, 289 (2009)
DOI: 10.1126/science.1174605
This PDF file includes: Materials and Methods Figs. S1 and S2 References
Supplementary information (686 words) Modern Riding Style Improves Horseracing Times Thilo Pfau, Andrew Spence, Sandra Starke, Marta Ferrari, Alan Wilson Structure and Motion Lab, The Royal Veterinary College, University of London, Hatfield, AL9 7TA England Race speed data are taken from various sources [S1] and on the few occasions - where race distances had changed in the US - these were compensated for by calculating average times per mile. No such correction was required for the Epsom Derby times (see Figure S1) since racing distance remained unchanged. Epsom Derby races carried out in Newmarket (1915-1918) were excluded from the analysis. Riding style from [S2]. One MTx inertial sensor (Xsens Technologies B.V., Enschede, The Netherlands), was fixed to the saddle cloth underneath the most caudal edge of the saddle. An identical sensor and data logger was fixed onto a neoprene kidney belt worn by the jockey. A GPS speed logger (Trine II, BTGPS, EMTAC, Taiwan) was attached to jockeys hat. Data were processed using custom written software in MATLAB (Natick, MA, USA). Using factory calibration data, raw binary inertial sensor data were calibrated into acceleration (ms-2), angular velocity (degree s-1), and measurements of the earths magnetic field (arbitrary units). GPS speed was used to manually select 15-20 second blocks of data from each run. Inertial sensor data were then processed as previously described [S2] to generate craniocaudal (CC, fore-aft), mediolateral (ML, side to side)
and dorsoventral (DV, vertical) accelerations, velocities and displacements. Briefly, sensor orientation was used to rotate sensor accelerations into a horse reference frame wherein the +x axis is aligned with the heading of the animal, +y is sideways to the left, and +z is vertical upward. Before integrating acceleration respectively velocity data are high pass filtered [S3]. Movement of horse and jockey is expressed relative to the mean, constant velocity horizontal motion. Movement of the jockey relative to the horse (Fig. 1C) was computed as the difference between the jockey and horse displacements. Seventeen routine training sessions of five high quality race horses, with three jockeys (750 strides at 11.4 + 0.9 ms-1, average stride frequency 2.2 Hz) were analysed. Calculation of drag power (Pd) to estimate the difference between traditional upright riding style and modern American riding style (change in frontal area taken from photographs, see for example Figure 1A and Figure S2 for a comparison) are based on the equation for quadratic drag (with =1.293 kg/m3, Cd=0.5,
A=0.12 m2 and v=15 ms-1 (108s/mile) this results in an aerodynamic drag of ~130 W). Metabolic power of a galloping horse at race conditions (maximum exercise [S4] or ~2.5 J/(kg m) [S5]) was estimated at 22 kW. Assuming an overall muscle work efficiency of 30% gives a total muscle power of 6.6 kW and the additional drag related power requirement equates to roughly 2 % of the overall power requirement. Reduced aerodynamic drag would thus only explain part of the observed changes in race times.
References [S1] Race times (Epsom Derby, Kentucky Derby, Belmont stakes and Preakness Stakes) from Cox Library (http://www.thecoxlibrary.com) and also available via various internet sources. [S2] Evidence of riding style changes in the early 20th century taken from photoarchive of Cox Library (The Derby Stakes by Vincent Orchard, 1954) and newspaper picture archives, e.g. Chicago Tribune. National Museum of Racing and Hall of Fame, Jockeys: Tod Sloan, http://www.racingmuseum.org/hall/jockey.asp?ID=222, National Museum of Racing, Saratoga Springs, NY, 191 Union Avenue Saratoga Springs NY 128663566 [S3] T. Pfau, T. H. Witte, A. M. Wilson, J Exp Biol, 208, 2503-2514 (2005). [S4] P. McDonough, C. A. Kindig, C. Ramsel, D. C. Poole, H. E., Erickson, Exp. Physiol., 87, 499-506 (2002) [S5] A. E. Minetti, L. P. Ardigo, E. Reinach, F. Saibene, J. Exp. Biol., 202, 2329-2338 (1999)
67 Figure S1: Race times in the Epsom Derby Stakes from 1846 (when recording of race times began) to the present day. Circles represent time of the winning horse by year and the line shows the best time recorded (record) starting ten years after the first recording.
Figure S2: Photograph of horses showing the transition towards the modern style (finish photo of the 1913 Derby Stakes, taken from The Derby Stakes by Vincent Orchard, published in 1954, courtesy Tim Cox, the Cox Library)