See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/40028408

Kinematic and Kinetic Comparisons of Elite and Well-Trained Sprinters

During Sprint Start

Article in The Journal of Strength and Conditioning Research · November 2009

DOI: 10.1519/JSC.0b013e3181ad3448 · Source: PubMed

CITATIONS READS

120 7,947

7 authors, including:

Jean Slawinski Guy Ontanon

French Institute of Sport (INSEP) French Institute of Sport (INSEP)
87 PUBLICATIONS 2,728 CITATIONS 8 PUBLICATIONS 243 CITATIONS

SEE PROFILE SEE PROFILE

Raphael Dumas Laurence Cheze

University Gustave Eiffel Claude Bernard University Lyon 1
261 PUBLICATIONS 4,150 CITATIONS 323 PUBLICATIONS 4,437 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Hominin hand preference identification through the analysis of use traces on stone tools (PreMHo) View project

Effects of Acute Physical Fatigue and Expertise on Gaze Behavior and Anticipation Behavior During a Badminton Game View project

All content following this page was uploaded by Jean Slawinski on 18 October 2017.

The user has requested enhancement of the downloaded file.

KINEMATIC AND KINETIC COMPARISONS OF
ELITE AND WELL-TRAINED SPRINTERS
DURING SPRINT START
JEAN SLAWINSKI,1 ALICE BONNEFOY,1 JEAN-MICHEL LEVÊQUE,1 GUY ONTANON,1 ANNIE RIQUET,1
RAPHAËL DUMAS,2 AND LAURENCE CHÈZE2
1
Scientific Expertise Centre, TeamLagardere, Jean Bouin Stadium, Paris, France; and 2University of Lyon 1,
Laboratory of Biomechanics and Mechanics of Shocks, Villeurbanne, France

ABSTRACT p # 0.05). The muscular strength and arm coordination appear

Slawinski, J, Bonnefoy, A, Levêque, JM, Ontanon, G, Riquet, A, to characterize the efficiency of the sprint start. To improve
Dumas, R, and Chèze, L. Kinematic and kinetic comparisons of speed capacities of their athletes, coaches must include in their
elite and well-trained sprinters during sprint start. J Strength habitual training sessions of resistance training.
Cond Res 24(4): 896–905, 2010—The purpose of this study KEY WORDS running, motion analysis, performance training
was to compare the main kinematic, kinetic, and dynamic param-
eters of elite and well-trained sprinters during the starting block
INTRODUCTION
phase and the 2 subsequent steps. Six elite sprinters (10.06–

T
10.43 s/100 m) and 6 well-trained sprinters (11.01–11.80 he analysis of different 100 m finals of international
events (Olympic Games in Seoul 1988, World
s/100 m) equipped with 63 passive reflective markers per-
Championships in Tokyo 1991, Athens 1997,
formed 4 maximal 10 m sprint starts on an indoor track. An opto-
Seville 1999, and Osaka 2007; from International
electronic motion analysis system consisting of 12 digital
Association of Athletics Federations reports) demonstrated
cameras (250 Hz) was used to record 3D marker trajectories. that sprinters reached a maximal velocity of 11.67 m
s21
At the times ‘‘on your marks,’’ ‘‘set,’’ ‘‘clearing the block,’’ and between the 50 m and 60 m of the race. This analysis also
‘‘landing and toe-off of the first and second step,’’ the horizontal demonstrated that, during the first 10 m, the speed gain is
position of the center of mass (CM), its velocity (XCM and VCM), maximal: 8.15 m
s21 at 10 m. More recently, during the last
and the horizontal position of the rear and front hand (XHand_rear Olympic games in Beijing, the spectacular run of Usain Bolt
and XHand_front) were calculated. During the pushing phase on the in the Bird’s Nest confirmed this analysis. Indeed, Bolt
starting block and the 2 first steps, the rate of force development finished in 9.69 seconds, reaching an amazing maximal
and the impulse (Fimpulse) were also calculated. The main results velocity of about 12.2 m
s21 between 40 m and 50 m of the
showed that at each time XCM and VCM were significantly run (13) with a very small number of strides (only 40 com-
pared with the 44 used by the rest of the field). During
greater in elite sprinters. Moreover, during the pushing phase on
the first 10 m of the sprint, the speed gain was also maximal:
the block, the rate of force development and Fimpulse were
9.05 m
s21 at 10 m. Thus, to reach a higher maximal velocity,
significantly greater in elite sprinters (respectively, 15,505 6
we see that the starting block phase and subsequent accelera-
5,397 N
s21 and 8,459 6 3,811 N
s21 for the rate of force tion phase are 2 extremely important phases, which directly
development; 276.2 6 36.0 N
s and 215.4 6 28.5 N
s for generate the results in a 60 m and 100 m sprint.
Fimpulse, p # 0.05). Finally, at the block clearing, elite sprinters Many authors have been interested in the biomechanical
showed a greater XHand_rear and XHand_front than well-trained factors of these 2 phases to explain the key factors of sprint
sprinters (respectively, 0.076 0.12 m and 20.27 6 0.36 m performance (7–9,11,16,18–20,24,25). The starting block
for XHand_rear; 1.00 6 0.14 m and 0.52 6 0.27 m for XHand_front; phase refers to the time during which the sprinter is in
contact with the blocks (24). A closer horizontal projection of
the center of mass (CM) in the ‘‘set position’’ to the starting
line and a shorter block time guarantee a maximal block
Address correspondence to Jean Slawinski, jean.slawinski@teamlagar velocity of the sprinter (9). The subsequent acceleration refers
dere.com. to the first step until the 10th m of the run (10). The efficiency
24(4)/896–905 of this phase depends on the execution of the first step,
Journal of Strength and Conditioning Research particularly the length of the step and the position of the foot
Ó 2010 National Strength and Conditioning Association in the contact phase. In the first 3 steps, the body’s CM has to
the TM

896 Journal of Strength and Conditioning Research

 the TM

Journal of Strength and Conditioning Research | www.nsca-jscr.org

rise gradually in a vertical direction to enable the maximi- sprinters started the run using starting blocks. All trials were
zation of the anterior component of velocity of the CM (9). used for further analysis.
However, even though numerous kinematics studies have
Subjects
covered the subject of the sprint start, little data have been
The sprinters gave their informed written consent to
published for elite sprinters. Nevertheless, detailed informa-
participate in the study. Their age, body mass, height, and
tion on the starting block phase and on the first meters of the
personal best times over 100 m are presented in the Table 1.
run could be of great importance for coaches to better
Performance times of the elite sprinters over 100 m ranged
understand the specific movements of both these phases and
between 10.06 and 10.43 seconds. This group was called
to develop them. The purpose of this study was to compare
‘‘elite’’ because each had already participated in an in-
the major kinematic and kinetic parameters of elite and well-
ternational competition and belongs to the group of the 20
trained sprinters during these phases of the sprint start. This
best French sprinters. This study conforms to the recom-
analysis will describe the most relevant kinematic and kinetic
mendations of the Declaration of Helsinki and was approved
parameters that contribute positively to the efficiency of the
by the local ethics committee.
block phase and the subsequent acceleration in elite sprinters.
Procedures
METHODS The reaction time (RT) was measured with Reactime
Experimental Approach to the Problem (Microgate, Bolzano, Italy). The time at 5 m and 10 m
Six elite sprinters and 6 well-trained sprinters performed 4 (T5 and T10) was recorded using photocells (Microgate,
maximal 10 m sprint starts on an indoor track (Figure 1). The Bolzano, Italy).

Figure 1. Experimental design.

TABLE 1. Characteristics of sprinters.

Height (cm) Weight (kg) Age (yr) Half squat, max force (N) 100-m time (s)

Elite sprinters (6SD) 179.2 6 6.2 79.5 6 10.5 24.5 6 3.2 2,192 6 145 10.27 6 0.14
Well-trained sprinters (6SD) 175.3 6 4.0 66.3 6 5.5† 21.3 6 3.0 1,845 6 69* 11.31 6 0.28*
*p # 0.0001.
†p # 0.05.
‡Half squat maximal force was measured with isokinetic ergometer from Ariel Dynamics, Inc. (Trabucco Canyon, CA , USA).

VOLUME 24 | NUMBER 4 | APRIL 2010 | 897

Kinematic and Sprint Start

An opto-electronic motion analysis system (Santa Rosa,

CA, USA) consisting of 12 digital cameras (250 Hz) was used
to record the 3D marker trajectories. The subject was
equipped with 63 passive reflective markers (16 mm diameter)
(Figure 2). The markers were glued to the skin and were
assumed to follow the movement of the bony landmarks. The
markers have to be placed on points that are easily
identifiable on all participants, as close as possible to the
bony parts of the landmarks. This is required to minimize the
influence of the soft tissues that may result in movement
errors by way of the skin and fat tissues. The 3D trajectories
of the passive reflective markers were computed and then
corrected by a low-pass filter (Butterworth, fourth-order,
with a cut-off frequency of 8 Hz). Three coordinate systems
were determined on each body segment based on the
markers. The orientation of their axes was carefully carried
out using ISB recommendations (31,32). The rotation
sequences proposed by the International Society of Bio-
mechanics (ISB) were used to describe the lower- and the
upper-limb joint movements.

The RTand the Time at 10 m Were Used to Compute the Velocity of

the Sprinter. Segment kinematics during the starting block
phase (defined as the phase during which at least 1 foot is in
contact with the starting block) and the 2 first steps of the
Figure 2. Body modelled with 63 passive reflective markers.
acceleration phase were reconstructed from the spatial trajec-
tories of the markers according to ISB recommendations. Angle
values presented in this study correspond to the instantaneous
rotation value about the z-axis (i.e., angles of flexion-extension).
Moreover, from the reconstructed spatial trajectories of the

Figure 3. Five critical instants used to analyse kinematics data.

the TM

898 Journal of Strength and Conditioning Research

 the TM

Journal of Strength and Conditioning Research | www.nsca-jscr.org

markers, the segment mass,

position of the CM, and inertia
tensor were estimated with scal-
ing equations (12).
To analyse the movement of
the athletes, 3 critical phases
were identified during the block
phase (on your mark, set posi-
tion, and block clearing), and
the first and the second steps
were used to describe the acce-
leration phase (Figure 3). For
each phase, different kinematics
and kinetics parameters were
calculated using Matlab (Math-
works, Natick, MA, USA) and
Excel software (Microsoft,
Figure 4. Typical acceleration of center of mass (CM) curve during block phase, first and second step.
Redmond, WA, USA). The
referential origin was placed
on the middle of the start lane.

Position ‘‘On Your Marks’’. Verti-

cal and anterior components of TABLE 2. Kinematic and kinetic data for center of mass during ‘‘on your marks
the CM (YCM and XCM) were position.’’*
calculated.
On your marks position Elite sprinters (6SD) Well-trained sprinters (6SD)
Position ‘‘Set’’. YCM and XCM, XCM (cm) 225.7 6 2.1 231.7 6 4.5†
anterior-posterior position of YCM (cm) 49.7 6 3.1 51.3 6 5.4
the shoulder (Xshoulder), and
*XCM = anterior component of center of mass; YCM = vertical component of center of
front and rear knee flexion- mass.
extension angles (FKA and †p # 0.05.
RKA) were calculated.

Clearing of the Block. This phase

refers to the time when the front
foot leaves the starting block.
We measured YCM and XCM,
the anterior-posterior position TABLE 3. Kinematic and kinetic data for center of mass during ‘‘set position.’’*
of the rear and front hand Set position Elite sprinters (6SD) Well-trained sprinters (6SD)
(XHand_rear and XHand_front), the
2D angle between the ground XCM (cm) 222.9 6 1.5 227.81 6 2.8†
and the body and the flexio- YCM (cm) 65.7 6 3.8 62.6 6 3.9
Xshoulder (cm) 10.7 6 2.7 4.0 6 5.5†
n-extension of the rear and Front knee angle (°) 110.7 6 9.3 106.1 6 13.7
front hip (RHA and FHA), Rear knee angle (°) 135.5 6 11.4 117.3 6 10.1†
and RKA. The term ‘‘rear’’
refers to the side of the body *XCM = anterior component of center of mass; YCM = vertical component of center of
mass; Xshoulder = anterior-posterior position of shoulder.
that is associated with the rear †p # 0.05.
foot in the blocks. The term
‘‘front’’ refers to the side of the
body that is associated with
the front foot in the blocks. The
vertical velocity and the norm of the velocity of the CM were from the first movement of the set position to the clearing
also calculated on clearing the block (VYCMclear and block. We measured the duration of this phase (time block),
VCMclear). All these parameters are detailed in Figure 3. the pushing time on the rear block (PTRB), the percentage of
To characterize the efficiency of the block phase, we took the pushing time on the rear block (%PTRB), the time flight
into account a pushing phase. This phase comprises the time between the rear block and the first step (Tf_rear block to

VOLUME 24 | NUMBER 4 | APRIL 2010 | 899

Kinematic and Sprint Start

first step), the average speed and

acceleration of the CM during
the pushing phase (VblockMean, TABLE 4. Kinematic and kinetic data for center of mass during ‘‘pushing phase.’’*
AblockMean), the first and sec- Block phase Elite sprinters (6SD) Well-trained sprinters (6SD)
ond peak acceleration (Apeak1,
Apeak2), and the rate of force Time block (s) 0.352 6 0.018 0.351 6 0.020
development (RFD). The RFD PTRB (s) 0.154 6 0.017 0.140 6 0.026
%PTRB (s) 43.5 6 3.8 39.8 6 8.1
was calculated as the product
Tf_rear block to first step (s) 0.292 6 0.021 0.300 6 0.029
between the slope of the first VblockMean (m
s21) 1.94 6 0.09 1.87 6 0.14
acceleration (SApeak1) and the AblockMean (m
s22) 9.5 6 0.4 8.8 6 0.8
body mass. The average force Apeak1 (m
s22) 16.7 6 1.9 13.7 6 3.1
impulse (Fimpulse) was also cal- Apeak2 (m
s22) 13.6 6 1.8 12.6 6 1.3
RFD (N
s21) 15505 6 5,397 8459 6 3,811†
culated. Fimpulse is the product of
Fimpulse (N
s) 276.2 6 36.0 215.4 6 28.5†
the resultant force applied on
the ground and the duration of *PTRB = pushing time on rear block; Tf_rear block to first step = time flight between rear
application of this force. In block and first step; VblockMean = average speed of center of mass during pushing phase;
AblockMean = acceleration of center of mass during pushing phase; Apeak1 = first peak
other words, Fimpulse is the acceleration; Apeak2 = second peak acceleration; RFD = rate of force development; Fimpulse =
integral of the resultant force average force impulse.
applied on the ground during †p # 0.05.
the block phase. Thus, we
calculated Fimpulse as
Z t
flight between the block clearing and the first step (Tf - Block
F impulse ¼ Fdt where F ¼ ðF y  P Þ þ F x þ F z
0 to first step), and the time flight between the first step and the
Z t second step (Tf - first to second step). During stance, we also
F impulse ¼ M gdt calculated the average velocity of the CM (Vmean), the average
0
F impulse  VCMclear 3 M eq:1 and maximal acceleration of the CM (Amean and Amax), the
slope of the first acceleration (SAmax) and the RFD (Figure 4),
In this equation, F is the norm of the resultant reaction force the stride rate (SR), and the stride length (SL).
applied on the ground (Fy, Fx, and Fz are the vertical,
horizontal, and lateral components of the resultant force), P is Statistical Analyses
the weight, g is the norm of the resultant acceleration of the All data are presented as means plus or minus SD. After
CM, VCMclear is the norm of the velocity of the CM at the a normality test, comparison of the kinematics and kinetics
block clearing (calculated with
the motion analysis system),
and M is the body mass. TABLE 5. Kinematic and kinetic data for center of mass during ‘‘block clearing.’’*

First and Second Step. We calcu- Clearing of block Elite sprinters (6SD) Well-trained sprinters (6SD)
lated the anterior component of
XCM (cm) 37.0 6 2.0 30.2 6 5.9†
the CM when the foot hit the
YCM (cm) 82.8 6 3.3 81.2 6 4.1
ground (XCMland), the anterior Yfoot (cm) 22.9 6 4.2 19.1 6 4.5
component of the CM when VCMclear (m
s21) 3.48 6 0.05 3.24 6 0.18†
the foot left the ground VYCMclear (m
s21) 0.52 6 0.06 0.51 6 0.14
(XCMto), the distance travelled XHand_rear (cm) 7.5 6 12.0 227.1 6 36.5†
XHand_front (cm) 99.8 6 13.8 52.3 6 27.0†
by the CM during stance
RKA (°) 297.5 6 7.0 282.1 6 10.2†
(DCM), the difference between Body - ground angle (°) 34.7 6 1.4 34.3 6 2.0
the anterior-posterior position RHA (°) 79.3 6 8.8 73.7 6 11.0
of the foot when the foot FHA (°) 171.2 6 8.4 173.1 6 12.5
hit the ground and XCMland
*XCM = anterior component of center of mass; YCM = vertical component of center of
(XCMland – Xfoot), the highest mass; Yfoot = highest vertical position of free foot; VCMclear = norm of velocity of center of
vertical position of the free foot mass; VYCMclear = vertical velocity of center of mass; XHand_rear = horizontal position of rear
hand; XHand_front = horizontal position of front hand; RKA = rear knee flexion-extension
(Yfoot), the velocity norm and
angle; RHA = flexion-extension of rear hip; FHA = flexion-extension of front hip.
the vertical velocity of the CM †p # 0.05.
at the toe-off (VCMto, VYCMto),
the time contact (Tc), the time
the TM

900 Journal of Strength and Conditioning Research

 the TM

Journal of Strength and Conditioning Research | www.nsca-jscr.org

RESULTS
Reaction Time and Time at
TABLE 6. Kinematic and kinetic data for center of mass during ‘‘first step.’’*
5 and 10 Meters
First step Elite sprinters (6SD) Well-trained sprinters (6SD) The average RT of the elite
sprinters was not significantly
XCMland (cm) 68.5 6 4.7 58.0 6 8.1†
XCMto (cm) 137.1 6 9.0 120.8 6 8.7†
different from that of the well-
DCM (cm) 68.6 6 5.1 62.8 6 4.4 trained sprinters (respectively,
Yfoot (cm) 22.7 6 8.6 32.7 6 6.7† 0.151 6 0.016 s vs. 0.158 6
VCMto (m
s21) 4.69 6 0.15 4.42 6 0.11† 0.033 s; p = 0.7). When the time
VYCMto (m
s21) 0.35 6 0.03 0.42 6 0.09 at 5 m and 10 m minus the RT
XCMland 2 Xfoot (cm) 17.5 6 3.0 19.4 6 4.5
Tc(s) 0.173 6 0.010 0.167 6 0.011
was considered, the elite sprint-
Tf 2 Block to first step (s) 0.093 6 0.009 0.087 6 0.021 ers ran faster than the well-
Tf 2 first to second step (s) 0.067 6 0.008 0.083 6 0.031 trained sprinters: the times at
VMean (m
s21) 4.06 6 0.09 3.87 6 0.12† 5 m and 10 m were, respec-
AMean (m
s22) 7.5 6 0.4 7.0 6 0.4 tively, 1.20 6 0.04 seconds and
Amax (m
s21) 15.2 6 1.3 13.6 6 2.3
RFD (N
s21) 13,570 6 4,126 9,804 6 2287
1.88 6 0.03 seconds for the elite
SR (Hz) 3.78 6 0.24 3.97 6 0.34 sprinters and 1.25 6 0.03
SL (cm) 137.1 6 9.0 120.8 6 8.7† seconds and 1.97 6 0.05 sec-
Fimpulse (N
s) 104.8 6 16.5 78.6 6 6.3† onds for the well-trained
*XCMland = anterior component of center of mass when foot hit the ground; XCMto =
sprinters (p # 0.05). When
anterior component of center of mass when foot left the ground; DCM = distance travelled by the total time was considered,
center of mass during stance; Yfoot = highest vertical position of free foot; VCMto = vertical only the time at 10 m was lower
velocity of center of mass at toe-off; VYCMto = vertical velocity of center of mass at toe-off;
XCMland 2 Xfoot = difference between the anterior-posterior position of foot when foot hit the
for elite sprinters (2.03 6 0.04 s
ground; Tc = time contact; Tf 2 Block to first step = time flight between block clearing and first vs. 2.12 6 0.06 s; p # 0.05). The
step; Tf 2 first to second step = time flight between first step and second step; VMean = average velocity at 10 m was
average velocity of center of mass; AMean = average acceleration of center of mass; Amax =
maximal acceleration of center of mass; RFD = rate of force development; SR = stride rate;
also higher for elite sprinters
SL = stride length; Fimpulse = average force impulse. (4.93 6 0.09 vs. 4.71 6
†p # 0.05. 0.14 m
s21; p # 0.05).

Kinematics and Kinetics

All results are presented in
data between the elite and well-trained sprinters was Tables 2, 3, 4, 5, and 6 and Figures 4 and 5. In the position
performed with an unpaired Student’s t-test. All significant ‘‘on your marks,’’ XCM was greater in elite sprinters (Table 2).
differences reported are at p # 0.05 unless otherwise noted. In the ‘‘set position,’’ XCM, Xshoulders, and RKA of the elite
sprinters were greater than
those of the well-trained sprint-
ers (Table 3). During the push-
ing phase on the block, only the
RFD and Fimpulse were greater
in the elite sprinters (Table 4).
At the clearing of the block,
XCM and VCMclear were greater
in elite sprinters (Table 5). The
elite sprinters positioned their
rear and front hands closer to
the finish line than the well-
trained sprinters. During the
first step, XCMland, XCMto,
VCMto, VMean, Fimpulse, and SL
were greater in the elite sprint-
ers (Table 6). Table 7 shows
that XCMto, VCMto, and Fimpulse
were greater in the elite sprint-
Figure 5. Horizontal position of center of mass (XCM) at 5 critical instants (*p # 0.05). ers and that VYCMto was lower
in the elite sprinters.

VOLUME 24 | NUMBER 4 | APRIL 2010 | 901

Kinematic and Sprint Start

the well-trained sprinters would

be 44 cm behind the elite
TABLE 7. Kinematic and kinetic data for center of mass during ‘‘second step.’’* sprinters at the 10 m mark.
Second step Elite sprinters (6SD) Well-trained sprinters (6SD)
Moreover, the comparison of
the present data with the time at
XCMland (cm) 168.2 6 11.3 156.9 6 12.4 10 m performed during inter-
XCMto (cm) 243.6 6 13.9 224.9 6 12.0† national events (the Olympic
DCM (cm) 75.3 6 5.72 68.4 6 7.9
Games in Seoul 1988, the World
Yfoot (cm) 24.4 6 6.2 26.7 6 7.2
VCMto (m
s21) 5.50 6 0.26 5.25 6 0.13† Championships in Tokyo 1991,
VYCMto (m
s21) 0.35 6 0.05 0.45 6 0.07† Athens 1997, Seville 1999, and
XCMland 2 Xfoot (cm) 6.1 6 4.6 7.3 6 9.0 Osaka 2007) confirmed that
Tc (s) 0.138 6 0.031 0.145 6 0.016 the best sprinters started faster.
21
VMean (m
s ) 5.07 6 0.19 4.84 6 0.09†
22 The average time at 10 m of
AMean (m
s ) 6.2 6 0.5 5.6 6 1.2
Amax (m
s21) 13.8 6 2.8 12.7 6 5.9 17 sprinters with a time over
RFD (N
s21) 12,172 6 3,494 10,092 6 1,069 100 m of less than 10 seconds is
SR (Hz) 4.61 6 0.16 4.40 6 0.27 1.84 6 0.06 seconds (including
SL (cm) 106.6 6 5.9 105.3 6 6.3 RT). This time at 10 m corre-
Fimpulse (N
s) 75.0 6 15.8 55.9 6 9.4†
sponds to an average speed of
*XCMland = anterior component of center of mass when foot hit the ground; XCMto = 5.44 6 0.19 m
s21, which is
anterior component of center of mass when foot left the ground; DCM = distance travelled by 110% greater than the 1 of the
center of mass during stance; Yfoot = highest vertical position of free foot; VCMto = vertical
velocity of center of mass at toe-off; VYCMto = vertical velocity of center of mass at toe-off;
elite sprinters from this study.
XCMland 2 Xfoot = difference between the anterior-posterior position of foot when foot hit the In other words, at the 10 m
ground; Tc = time contact; Tf 2 Block to first step = time flight between block clearing and first mark, the elite sprinters from
step; Tf 2 first to second step = time flight between first step and second step; VMean =
average velocity of center of mass; AMean = average acceleration of center of mass; Amax =
our study would be 90 cm
maximal acceleration of center of mass; RFD = rate of force development; SR = stride rate; behind a sprinter with a per-
SL = stride length; Fimpulse = average force impulse. sonal best over 100 m of less
†p # 0.05.
than 10 seconds. More recently,
the importance of the first 10 m
in achieving a good perfor-
mance during the 100 m race
is confirmed by the work of
Eriksen et al. (13), which mod-
elled the race of Usain Bolt
during his world record race.
They found a time at the 10 m
mark of 1.75 seconds (including
RT), which corresponds to an
average speed of 5.66 m
s21.
Thus, at the 10 m mark, the
sprinters with a personal best
over 100 m would be about
50 cm behind Usain Bolt.
These first 10 m are influ-
enced by the RT and the first
steps of the run. The study of
the RT during international
championships demonstrated
Figure 6. Evolution of velocity of center of mass during pushing phase and 2 first steps. that the RT is generally lesser
than 0.2 seconds (between 0.12
and 0.17 s) and that it increases
DISCUSSION in proportion to the length of the race distance (2,14,22). The
The above results show that the elite sprinters started faster results of the present study demonstrate that the RT for elite
than the well-trained sprinters. At 5 m, the elite sprinters, and well-trained runners was, respectively, 0.151 seconds and
compared with the well-trained sprinters, had a gain of 0.05 0.158 seconds, and no significant differences in RT existed
seconds and a gain of 0.09 seconds at 10 m. This means that between the 2 populations. This result was confirmed by
the TM

902 Journal of Strength and Conditioning Research

 the TM

Journal of Strength and Conditioning Research | www.nsca-jscr.org

Mero et al. (24), who suggested that RT does not correlate condition to improve the sprint start. However, more studies
with performance level. Other parameters (e.g., position or have to be conducted to better understand the role of the
velocity of the CM) may be more important than RT in arms in the sprint start and their influence on the position of
explaining the differences observed in the 5 m and 10 m times the CM at the block clearing.
between the 2 populations. The kinematic and kinetic At the block clearing, a greater VCM for elite sprinters
analysis allows us to understand better why the elite sprinters (3.48 6 0.05 m
s21 for elite and 3.24 6 0.18 m
s21 for
showed faster times at 5 m and 10 m than the well-trained well-trained sprinters) conforms to the results of other
sprinters. studies, which demonstrate that the horizontal velocity of the
The results of the present study show that elite sprinters CM at the block clearing of highly skilled participants (10.02–
put their CM closer to the start line than the well-trained 10.79 s/100 m time) ranged from 3.46 to 4.11 m
s21. Less-
sprinters (Figure 5). Positioning the CM as close as possible skilled sprinters (11.5–11.85 s/100 m time) have displayed
to the start line is important in reducing the distance by lower block velocities of 2.94 to 2.95 m
s21 (9,18). Thus, the
which the athlete must displace his CM and in creating ability of a sprinter to leave the blocks at a high velocity
a greater velocity of the CM during the pushing phase. increases with his sprint performance over 100 m (18).
Baumann (3) also found that the horizontal distance from the Equation 1 (see Methods) suggests that a greater impulse
start line to the CM decreased with increasing performance explains this ability to leave the blocks at a high velocity.
level. Harland and Steele (18) confirmed these observations Indeed, during the pushing phase, the impulse is greater in
and found that the horizontal distance of the CM from the the elite sprinters (276.2 6 36.0 N
s for elite and 215.4 6 28.5
start line (XCM) was 0.16 m for fast (10.35 s), 0.20 m for N
s21 for well-trained sprinters). In accordance with classic
medium (11.11 s), and 0.27 for slow sprinters (11.85 s). In the mechanical physics, the impulse of a movement is defined as
present study, we found a greater XCM for elite (0.23 m) and the area under the force-time curve. The size of this area
well-trained sprinters (0.28 m). These differences could be depends on 3 main parameters: the duration of force
related to the method of calculation of the CM. In fact, in this application, the RFD, and the maximal force reached. No
study, we used new scaling equations (12) to compute the difference in the duration of force application has been found
segment mass, the position of the CM . This greater XCM for between elite and well-trained sprinters. Indeed, the time
elite sprinters is a result of a greater rear knee angle. Indeed, block was not significantly different between elite and well-
the elite sprinters, when compared with the well-trained trained sprinters (Table 4). The RFD was greater in elite
sprinters, presented the same front knee angle but a greater sprinters (Table 4). The RFD exerted within the early phase
rear knee angle. A greater rear knee angle allowed the of rising muscle force can be defined as the rate of rise in
sprinters to position the shoulders further forward and to contractile force at the onset of contraction and is generally
move the CM closer to the start line. These data are called ‘‘explosive muscle strength’’ (17,29,30). During fast
confirmed by Harland and Steele (18), who summarized knee limb movements, the short contraction time may not allow
angles of skilled sprinters in the set position measured in maximal muscle force to be reached. As a result, any increase
different studies. They showed that the front knee angle in in contractile RFD becomes highly significant because it
the set position ranges from 90° to 110°, and the rear knee allows a higher level of muscle force to be reached in the
angle ranges from 115° to 130°. This greater rear knee angle early phase of muscle contraction (e.g., within the initial
and the further forward position of the shoulder suggested 0.100–0.200 s of contraction). During the pushing phase on
that the strength of the arms is a key factor in reducing this the starting block, the maximal acceleration (Apeak1) was
horizontal distance from the start line to the CM. The reached when the rear foot pushed on the rear block and in
stronger the arms are, the shorter the horizontal distance a very short time, less than 0.150 seconds. Thus, RFD was an
from the start line to the CM (24). extremely important parameter in the ability to leave the
At the block clearing, the elite sprinters had greater XCM blocks at a high velocity. In addition to RFD, another
than well-trained sprinters (Figure 5). XCM depends on the important muscular strength parameter is the maximal force
position of the body segments (trunk, arms, legs, etc.). The that can be produced within a given contraction time. This
results of this study showed that the elite runners placed both maximal force produced during the pushing phase is the
hands (front and rear) further forward than the well-trained product between the body mass and the maximal acceler-
sprinters (Table 5). Thus, for elite sprinters, the ability to ation of the CM of the body (e.g., Newton’s second law of
move the CM further forward than well-trained sprinters motion). The maximal acceleration reached during the
depends partially on the movement of the arms. Many pushing phase (Apeak1) was not significantly different between
studies have investigated the coordination between legs and elite and well-trained sprinters. Only the body mass was
arms of explosive leg extensions movement such as squat higher in elite sprinters. Thus, the difference in impulse
jump (5,6,15,23). However, few studies have been interested between elite and well-trained sprinters was directly asso-
in the analysis of arm movements in the sprint start (4,26). ciated with body mass and RFD.
Bhowmick and Bhattacharyya (4) suggested only that the These differences in RFD and body mass between the elite
vertical component of the arm movement creates a favorable and the well-trained sprinters could be explained by the level

VOLUME 24 | NUMBER 4 | APRIL 2010 | 903

Kinematic and Sprint Start

of resistance training. Indeed, resistance training induces an From a practical point of view, a resistance training
increase in explosive muscle strength associated with neural program aimed at inducing morphologic and neural changes
adaptations (1) and muscle hypertrophy (21). The elite in sprinters will always constitute a mixture of different
sprinter population used in this study have carried out much methods. These methods are related to one another in
more resistance training than well-trained sprinters. Indeed, a hierarchical way. First, power training is performed. This
the elite sprinters completed a resistance training session power training is characterised by a large number of sets of
3 times a week, and they presented very high strength repetitions with submaximal loads of 60–80%, where 100% is
capacities (they were able to squat approximately 2.5–3 times the maximum weight that can be handled in a dynamic
their body weight for a half squat). Thus, the improvement situation. The execution of the movement is performed at
of strength capacity in sprinters could be a key factor for maximal velocity. The exercise stops when the velocity of the
improving VCM during the block phase (28). movement decreases. Second, resistance training must be
However, differences between the elite and well-trained performed with power training. Methods to improve strength
sprinters are not only a result of differences in strength, muscle use a small number of sets of repetitions with maximal loads of
mass, or resistance training. Indeed, other factors such as 80–100% (or, in the case of eccentric actions, supramaximal
experience, technique, or innate capacity may be involved in loads). In these exercises, the movement velocity of the load is
these differences. relatively low, but the neuromuscular action is maximal. As in
At the landing of the first step, elite sprinters had a greater sprinting, high-velocity contraction is of major importance.
XCMland than well-trained sprinters. The greater velocity at Therefore, power and strength training must be followed by
the clearing block and the greater SL between the rear foot in speed training. Third, speed training methods demand loads
the block and the first step explain this result. At the toe-off that the athlete can displace with a high velocity of movement
of the first step, XCMto was also greater in elite sprinters execution. These loads are lower than 60% of the maximal
(Figure 5). At the landing of the second step, elite sprinters load. Delecluse (10) suggested that, to reduce the de-
did not have a greater XCMland. Compared with the landing celeration induced by the stop of the load at the end of an
of the first step, the difference in XCMland between elite and explosive concentric movement, these movements can be
well-trained was reduced (18% to 7%) (Tables 6 and 7). This replaced by ballistic movements including stretch-shortening
result is surprising because VCMto and Vmean during the cycle, bench throwing, or squat jumping.
contact of the first step were greater in elite sprinters. Both To establish a transfer between resistance training and fast
parameters allow projection of the CM further forward than running, there is still a need for specific sprint exercises that
as was so in the well-trained sprinters. The fact that XCMland include a strength component. Two groups of methods have
was not greater in elite runners could be explained by the bad been developed: overspeed and overload running (10). To
coordination of elite sprinters during the flight phase that improve the efficiency of the block phase and the subsequent
follows the first step. acceleration, overload running is suggested. The classic
As shown in Figure 6, the elite sprinters presented a overload training technique is incline running. Indeed, there
greater VCM at the toe-off of the first and the second steps. is a mechanical similarity between the starting block phase
The greater impulse in the elite sprinters during both steps and subsequent acceleration and incline running. Incline
explains this result. Here, the greater impulse in the elite running induces an increase of the push-off time that is also
sprinters is associated only with their greater body mass and observed during the acceleration phase of a sprint. Incline
not with the RFD as was the case in the block phase. They sprint running (with slope greater than 3%) could be
also had a lower VYCMto, which improves the orientation of beneficial to performance as related to the specific acceler-
the vector speed compared with the well-trained sprinters. ation phase (27).
To conclude, the present study shows that, to start faster,
elite sprinters placed their CM as close as possible to the finish REFERENCES
line (Figure 5). It appears that their greater ‘‘explosive muscle 1. Aagaard, P, Simonsen, EB, Andersen, JL, Magnusson, P, and Dyhre-
strength’’ and better arm coordination allowed them to have Poulsen, P. Increased rate of force development and neural drive of
a greater RFD and impulse and thus a greater velocity of their human skeletal muscle following resistance training. J Appl Physiol
93: 1318–1326, 2002.
CM from the block phase until the toe-off of the second step
2. Ae, M and Ito, A. The men’s 100 meters. New Stud Athletics 7: 47–52,
(Figure 6). 1992.
3. Baumann, W. Kinematic and dynamic characteristics of the sprint
start. In: Biomechanics IV. Komi, PV, ed. Baltimore, MD: University
PRACTICAL APPLICATIONS Park Press, 1976. pp. 121–125.
The present results demonstrate that coordination and 4. Bhowmick, S and Bhattacharyya, AK. Kinematic analysis of arm
strength capacity are key factors to improve the efficiency movements in sprint start. J Sports Med Phys Fitness 28: 315–323,
1988.
of the starting block phase and subsequent acceleration phase.
5. Bobbert, MF, Gerritsen, KG, Litjens, MC, and Van Soest, AJ. Why is
Coordination and strength capacity can be improved countermovement jump height greater than squat jump height? Med
considerably with resistance training (10). Sci Sports Exerc 28: 1402–1412, 1996.
the TM

904 Journal of Strength and Conditioning Research

 the TM

Journal of Strength and Conditioning Research | www.nsca-jscr.org

6. Bobbert, MF and van Ingen Schenau, GJ. Coordination in vertical 21. Kraemer, WJ and Ratamess, NA. Fundamentals of resistance
jumping. J Biomech 21: 249–262, 1988. training: progression and exercise prescription. Med Sci Sports Exerc
7. Boisnoir, A, Decker, L, Reine, B, and Natta, F. Validation of an 36: 674–688, 2004.
integrated experimental set-up for kinetic and kinematic three- 22. Martin, D and Buoncristiani, J. Influence of reaction time on athlétic
dimensional analyses in a training environment. Sports Biomech performance. New Stud Athletics 10: 67–79, 1995.
6: 215–223, 2007. 23. Mathiyakom, W, McNitt-Gray, JL, and Wilcox, R. Lower
8. Čoh, M, Jošt, B, Škof, B, Tomažin, K, and Dolenec, A. Kinematic and extremity control and dynamics during backward angular impulse
kinetic parameters of the sprint start and start acceleration model of generation in forward translating tasks. J Biomech 39: 990–1000,
top sprinters. Gymnica 28: 33–42, 1998. 2006.
9. Čoh, M, Tomažin, K, and Štuhec, S. The biomechanical model 24. Mero, A, Komi, PV, and Gregor, RJ. Biomechanics of sprint running.
of the sprint start and block acceleration. Phys Educ Sport A review. Sports Med 13: 376–392, 1992.
4: 103–114, 2006. 25. Mero, A, Kuitunen, S, Harland, M, Kyrolainen, H, and Komi, PV.
10. Delecluse, C. Influence of strength training on sprint running Effects of muscle-tendon length on joint moment and power during
performance. Current findings and implications for training. Sports sprint starts. J Sports Sci 24: 165–173, 2006.
Med 24: 147–156, 1997.
26. Ropret, R, Kukolj, M, Ugarkovic, D, Matavulj, D, and Jaric, S. Effects
11. Delecluse, C, Coppenolle, H, Diels, R, and Goris, M. A model for of arm and leg loading on sprint performance. Eur J Appl Physiol
the scientific preparation of high level sprinters. New Stud Athletics Occup Physiol 77: 547–550, 1998.
7: 57–64, 1992.
27. Slawinski, J, Dorel, S, Hug, F, Couturier, A, Fournel, V, Morin, JB,
12. Dumas, R, Cheze, L, and Verriest, JP. Adjustments to McConville and Hanon, C. Elite long sprint running: a comparison between
et al. and Young et al. body segment inertial parameters. J Biomech incline and level training sessions. Med Sci Sports Exerc 40:
40: 543–553, 2007. 1155–1162, 2008.
13. Eriksen, HK, Kristiansen, JR, Langangen, O, and Wehus, IK. Velocity 28. Sleivert, G and Taingahue, M. The relationship between maximal
dispersions in a cluster of stars: how fast could Usain Bolt have run? jump-squat power and sprint acceleration in athletes. Eur J Appl
arXiv 2: 1–5, 2008. Physiol 91: 46–52, 2004.
14. Ferro, A, Rivera, A, and Pagola, I. Biomechanical analysis of the 7th 29. Sleivert, GG and Wenger, HA. Reliability of measuring isometric and
World Championship in Athletics, Seville 1999. New Stud Athletics isokinetic peak torque, rate of torque development, integrated
16: 25–60, 2001. electromyography, and tibial nerve conduction velocity. Arch Phys
15. Giatsis, G, Kollias, I, Panoutsakopoulos, V, and Papaiakovou, G. Med Rehabil 75: 1315–1321, 1994.
Biomechanical differences in elite beach-volleyball players in vertical 30. Thorstensson, A, Karlsson, J, Viitasalo, JH, Luhtanen, P, and
squat jump on rigid and sand surface. Sports Biomech 3: 145–158, Komi, PV. Effect of strength training on EMG of human skeletal
2004. muscle. Acta Physiol Scand 98: 232–236, 1976.
16. Guissard, N, Duchateau, J, and Hainaut, K. EMG and mechanical 31. Wu, G, Siegler, S, Allard, P, Kirtley, C, Leardini, A, Rosenbaum, D,
changes during sprint starts at different front block obliquities. Whittle, M, D’Lima, DD, Cristofolini, L, Witte, H, Schmid, O, and
Med Sci Sports Exerc 24: 1257–1263, 1992. Stokes, I. ISB recommendation on definitions of joint coordinate
17. Hakkinen, K and Komi, PV. Training-induced changes in neuro- system of various joints for the reporting of human joint motion.
muscular performance under voluntary and reflex conditions. Eur J Part I. Ankle, hip, and spine. International Society of Biomechanics.
Appl Physiol Occup Physiol 55: 147–155, 1986. J Biomech 35: 543–548, 2002.
18. Harland, MJ and Steele, JR. Biomechanics of the sprint start. Sports 32. Wu, G, van der Helm, FC, Veeger, HE, Makhsous, M,
Med 23: 11–20, 1997. Van Roy, P, Anglin, C, Nagels, J, Karduna, AR, McQuade, K,
Wang, X, Werner, FW, and Buchholz, B. ISB recommendation
19. Korchemny, R. A new concept for sprint start and acceleration on definitions of joint coordinate systems of various joints
training. New Stud Athletics 7: 65–72, 1992. for the reporting of human joint motion. Part II.
20. Kraan, GA, van Veen, J, Snijders, CJ, and Storm, J. Starting from Shoulder, elbow, wrist and hand. J Biomech 38: 981–992,
standing; why step backwards? J Biomech 34: 211–215, 2001. 2005.

VOLUME 24 | NUMBER 4 | APRIL 2010 | 905

View publication stats