Discussion

The consideration of non-steady flow effects is essential when calculating mechanical work or
efficiency of self-induced propulsion at a given physical competence. This study using a test closely
related to natural swimmers situation (surfing at water level). Caspersen et al. (2010) used an
oscillating 2.8 m long bar with handles and with springs, while researching the relative added mass
(Ma%) for boys, women and men and concluded that the added mass in human swimmers, in
extended gliding position, is approximately 1/4 of the subjects' body mass. A direct comparison to the
results of this study concentrating on the effects when the acceleration is positive is not possible.
Since during cyclic activities in aquatic space deceleration also occurs the magnitude of 'added mass'
is expected to be different because of different flow events. The mathematical treatment is still a
challenge for experts to approach a more accurate model of flow effects due to cyclic interaction of
human body and displaced water mass.

Literature
Caspersen C, Berthelsen PA, Eik M, Pakozdi C, Kjendlie P.L. (2010). Added mass in human swimmers: Age and
gender difference. JoBiom, Vol43, 12:2369-2373.

Cureton, T.K. {1971) Biomechanics of swimming with inter-relationship to fitness and performance. In: L. Lewillie,
J.P. Clarys (Eds.) Proc. First Intern. Symposium on 'Biomechanics of Swimming'. Universite Libre de Bruxeles,
31-52.

Klauck, J. (1976) Der Wasserwiderstand des menschlichen Korpers bei beschleunigter Bewegung. W. Decker, M.
Laemmer (Eds) Kolner Beitrage zur Sportwissenschaft (5) 25-35.

Klauck, J. {1999) Man's water resistance in accelerated motion: An experimental evaluation of the added mass
concept. In: K.L. Keskinen, P.V. Komi, A.P. Hollander (Eds.) BMS VIII, Univeritiy of Jyvaskyla, 83-88.

Lighthill M.J. {1969). Hydromechanics of Aquatic Animal Propulsion- Annual Review of Fluid Mechanics Vol. 1:
413-446

Matsuuchi K., Miwa T., Nomura T., Sakakibara J., Shintani H., Ungerechts B. E. (2009). Unsteady flow field around
a human hand and propulsive force in swimming. Journal of Biomechanics Vol. 42, Issue 1: 42-47.

Ungerechts, B. (1988). The relation of peak body acceleration and phases of movements in swimming. In: B. E.
Ungerechts, K. Wilke and K. Reischle (eds.), Swimming Science V, Human Kinetics Publishers, Champaign:61-
66.

Ungerechts, B. E.; Buckwitz, R. & Baehr, H. (2003) Principles of non-stationary swimming- a preliminary attempt.
In: J C Chartard (eds.) BM IX, University of St. Etienne, France, 45-50

Ungerechts B E. & Klauck J. (2006). Consequences of non-stationary flow effects for functional attribution of
swimming strokes. In: J.P. Vilas-Boas, F. Alves, A. Marques (eds.), BMS X, Portuguese J of Sport Sci, Vol. 6,
Suppl. 2: 109-111.

Wirtz, W. (1996). Apparative Entwicklung und Anwendung eines dreidimensionalen kinematischen Verfahrens im
Kraulsprint. Shaker Verlag Aachen.

Inter-individual variability of body angles during swim start: analysis of

preferential and non-preferential techniques for expert swimmers
1 2 2 1 1
Julien Vantorre , Joao-Paulo Vilas-Boas , Ricardo J Fernandes , Didier Chollet , Ludovic Seifert
1 2
CETAPS EA 3832, Faculty of Sports Sciences, University of Rouen, France, LABIOMEP, Faculty of Sport, University
of Porta, Portugal

Keywords: biomechanics, variability, expertise, preferential technique, swimming start

Introduction
Most of the biomechanical studies of start time have used kinetic and kinematical analyses to
compare the two main start techniques used in competition: the grab and the track starts. Using a
track start, swimmers tend to leave the block quicker (Ayalon, Van Gheluwe & Kanitz 1975) and to

BMS2014-PROCEEDINGS 281
make a flatter flight trajectory due to higher horizontal velocity (Costill, Maglischo & Richardson
1992). With the grab start, swimmers spend more time on the block (lssurin & Verbitsky 2003). The
above-referred studies, as many other studies, tried to observe which is the best starting technique to
increase performance. Indeed, studies that compared the two start techniques showed the
comparison between grab starters and track starters, each swimmer in his preferential technique.
These studies were particularly interested in the starting technique effect. However, studies that
analyzed differences between start techniques and that take in consideration the swimmers start
preference are scarce as Vilas-Boas et al. showed us (2003) or more recently with Vantorre et al.
(2011) on angular momentum. Knowing that the use of preferential technique may induce better
performance to 1Sm, as well as other differences like a higher variability in some parts of the
movement. Indeed, some studies showed that expert swimmers were able to exploit movement
variability to achieve the dual-task goal of the swimming start: dive as far as possible to minimise
resistances and to make a forward rotation to enter into the water properly (Seifert et al. 2010;
Vantorre, Seifert, Fernandes, Vilas Boas & Chollet 2010; Vantorre, Seifert, Fernandes, Vilas-Boas &
Chollet 2010). Consequently, it is challenging to investigate the impact of the non-preferential
technique in comparison to the preference one.

The aim of this study was to analyse inter-individual variability on body angles and velocity between
preferential and non-preferential start techniques during aerial phase of swim start.

Method
Five expert swimmers (age: 23.2 ± 1.5 years, size: 1.8 ± 0.1 m, weight 78.6 ± 8.2 kg) male sprinters
specialists of freestyle voluntarily participated in this study. The skill level was expressed in
percentage of the world record time(% of RM} for a 100-m crawl in 50-m pool and was 89.3 ± 3.0%.
The target time was expected to be within more or less 2.5% of the race time. In a 25m swimming
pool, each swimmer performed six randomised 25m front crawl at the 50m race pace, being three
repetitions using the track start and three using the grab start techniques. Grab start technique was
their preferential technique.

Kinematics
Three cameras (two above water and one underwater) with rapid shutter speed (1/1000 s) were used
to follow the swimmer from the block to the entry of the feet in the water. The first camera (50 Hz,
Sony® DCR-HC42E) was placed from the edge of the pool and videotaped the leave block and flight
phases. Two other cameras (50Hz, JVC GR-SX1 SVHS-C PAL) were mounted on a specially designed
support placed at the lateral wall 3m from the edge of pool deck (one 30cm above and the other
30cm under the water surface), videotaping the entry phase. A fourth camera was placed in front of
the 15-m mark and videotaped the swimmer from the moment when the head broke the surface of
the water to the end of the 15-m. Kinematical analysis was processed using APAS (Ariel Performance
Analysis System, Ariel Dynamics Inc. 2001). The spatial model was composed by 20 anatomical
landmarks digitalised in each frame, defining 14 body segments model (De Leva 1996). Images, once
digitised, allowed us to obtain the instantaneous velocity of the center of gravity of the swimmer
throughout the movement and limbs angles. Concerning limbs organisation of the swimmer during
the different phases of the movement, a recalculation of the center of gravity of the arms and of the
legs allowed us to summarise a segment in a point using Matlab (Matlab 7.1, Mathworks lnc 2012).
Indeed, angles were calculated for the arm between the center of gravity (CG) of the upper limb parts
(hand, forearm and arm) and the trunk (CG arm- trunk) and concerning legs between the center of
gravity of the lower limbs (foot, leg and thigh) and trunk (CG leg- trunk). These measurements were
determined on all the movement from start signal and feet entry in the water. To compare the trials,
trials were normalised on one hundred values.

Inter-individual variability
Concerning inter-individual variability, the temporal dynamics of limbs angles and velocity of the
center of gravity during movement were studied. From the fastest trial for each subject, its duration

282 BMS20 14-PROCEEDINGS

was normalised in 100 points. The average standard deviation of each variable was calculated by
averaging the 100 standard deviations calculated point by point by taking all the subjects. This
represented an overall indicator of the inter-individual variability. After that, local indicators of inter-
individual variability were calculated at four key points: (1) first measured value; (2) time when the
feet leaved the block; (3) time when the hands enter into the water; (4) time when the feet enter into
the water. To do that, the average of standard deviations was calculated based on the 5 values before
and 5 values after each key point.

Statistical analysis
A normal distribution (Ryan Joiner test) and the homogeneity of variance (Bartlett test) were verified
and authorised parametric statistics (Minitab 15.1.0.0, Minitab Inc. 2006). Two-way ANOVA's (fixed
factor: technique; random factor: subject) were applied for all parameters (angles, velocity of the
center of gravity, standard deviations of these parameters) to analyze the differences between
techniques. Statistics were performed on mean values but also on the average standard deviations
(aSD). The significance level was set at p <.OS.

Results
Means standard deviations were calculated for the angles and the velocity of the center of gravity. On
these values, only the average standard deviation of the velocity of the center of gravity is
significantly different between the two techniques (Table 1).

Table 1 Means and standard deviations of the measured variables for preferential and non-preferential
technique
Variables Experts Non experts

CG Arm-Trunk 1 (") 93.22 ± 10.12 108.24 ± 4.42 (a.b)

CG Leg-Trunk 1 (") 30.09 ±4.18 27.36 ± 5.22 (a)
lnstV 1 (m.s- 1 ) 0.09 ±0.09 0.05 ± 0.04
CG Arm-Trunk 2 40.70 ± 78.40 66.35 ± 99.29 (a.b)
CG Leg-Trunk 2 169.32 ± 10.20 167.62 ± 8.22
lnstV 2 4.15 ± 0.24 4.12 ± 0.33
CG Arm-Trunk 3 173.16± 7.85 169.97 ± 8.07(a)

CG Leg-Trunk 3 152.15 ± 15.53 154±6.23(b)

lnstV 3 4.99 ± 0.31 4.94±0.25
CG Arm-Trunk 4 166.59 ± 16.49 170.24 ± 8.74(a,b)

CG Leg-Trunk 4 166.64 ± 6.15 166.4 ± 10.15(b)

lnstV 4 3.94 ± 0.32 3.72 ± 0.38(a)

CG Arm Trunk aSD 29.02 34.15(a)

CG Leg-Trunk aSD 9.47 9.61

lnstV mSD 0.27 0.25

15-mTime 6.44 ± 0.36 6.51 ± 0.32(a)

a: significant difference compared to preferential technique on averages, b: significant difference compared to preferential
technique on standard deviations at p <0.05.

Curves and standard deviations of body angles and velocity of the CG during the aerial phase of the
start were summarised in the following figure (Figure 1) with key points in bold.

BMS2014-PROCEEDINGS 283
 CG Arm-Trunk Preferential CG Leg-Trunk Prefrential CG Velocity Preferential

%of the starting action %ofthestartingaetlon o/oofthestartingaeHon

CG Arm- Trunk Non Preferential CG Leg-Trunk Non Preferential

CG Velocity Non Preferential

%of !he starting action %ofthestarUngactlon

%ofthestartmgaeffon
"'

Figure 1 Curves and SO of body angles in in preferential and non-preferential technique

A higher inter-individual variability of CG Arm-Trunk (29.02 vs 34.15 for average standard deviations)
and no significant difference at hand entry showed an adaptation to the task of starting in the two
techniques by the expert swimmers (Vantorre, Seifert, Fernandes, Vilas Boas et al. 2010). However,
lower alignment of upper limb (lower CG Arm-Trunk angle) during its entry induced a lower efficiency
at entry confirmed with a higher decrease in velocity (Seifert et al. 2010). Concerning CG Leg-Trunk,
experts had significant higher variability at arms entry (with 15.53 versus 6.23 respectively for expert
and non-expert swimmers) but results were inverted at feet entry (with 6.15 versus 10.15 respectively
for expert and non-expert swimmers) showing a process of adaptation during the transition between
rotating in the air and enter in the water in a streamlined position. Results also showed a use of
various body angles-especially in preferential technique- in the aerial phases (as previously showed
by Seifert et al. 2010) suggesting that several behavioural profiles enable to reach effectively the task-
goal.

Conclusion
These results were fundamental to link with the study of the same technique done on a traditional
block versus the OSB11 starting block (Honda, Sinclair, Mason & Pease 2012; Takeda, Takagi &
Tsubakimoto 2012). Indeed, this study showed the interest of considering movement variability as a
way of individualisation and exploration during the training process, suggesting that movement
variability could be viewed as functional and adaptive. One promising way could be to examine the
adaptation of swimmers on the new blocks.

References
Ayalon, A., Van Gheluwe, B. & Kanitz, M. (1975). A comparaison of four racing styles of racing strat in swimming.
In Biomechanics and Medicine in Swimming 11 (pp. 233-241).

Costill, D., Maglischo, E. & Richardson, A. (1992). Handbook of sports medicine and science. Swimming (p. 214).
Oxford: Blackwell Scientific Publications.

De leva, P. (1996). Adjustments to Zatsiorsky-Seluyanov's segment inertia parameters. Journal of Applied

Biomechanics, 29(9), 1223-1230.
Honda, K., Sinclair, P., Mason, B. & Pease, D. (2012). The effect of starting position on elite swim start
performance using an angled kick plate. In 30th Annual Conference of Biomechanics in Sports (pp. 72-75).
Melbourne.

284 BMS2014-PROCEEDINGS
lssurin, V. & Verbitsky, 0. (2003). Track Start vs Grab Start: evidence from the Sydney Olympic Games. In J.
Chatard (Ed.), Biomechanics and Medicine in Swimming IX (pp. 213-217). Saint Etienne: Universite de Saint
Etienne.

Seifert, L., Vantorre, J., Lemaitre, F., Chollet, D., Toussaint, H. & Vilas-Boas, J. (2010). Different profiles of the
aerial start phase in front crawl. The Journal of strength and conditioning research, 24(2), 507-516.

Takeda, T., Takagi, H. & Tsubakimoto, S. (2012). Effect of inclination and position of new swimming starting
block's back plate on track-start performance. Sports Biomechanics, 11(3), 37Q-381.

Vantorre, J., Seifert, L., Fernandes, R., Vilas Boas, J. & Chollet, D. (2010). Comparison of grab start between elite
and trained swimmers. International journal of sports medicine, 31(12), 887-93.

Vantorre, J., Seifert, L., Fernandes, R., Vilas-Boas, J. & Chollet, D. (2010). Kinematical profiling of the front crawl
start. International Journal of Sport Medecine, 31, 16-21.

Vantorre, J., Seifert, L., Vilas-boas, J., Fernandes, R., Bideau, B., Nicolas, G. & Chollet, D. (2011). Biomechanical
analysis of starting preference for expert. Portuguese Journal of Sport Sciences, 11(2), 415-418.

Vilas-Boas, J., Cruz, J., Sousa, F., Conceic;ao, F., Fernandez, R.& Carvalho, J. (2003). Biomechanical analysis of
ventral swimming starts: comparaison of the grab-start with two track-start techniques. In J. Chatard (Ed.),
Biomechanics and Medicine in Swimming IX (pp. 249-253). Saint Etienne: University of Saint Etienne.

BMS2014-PROCEEDINGS 285