sports
Article
A Three-Dimensional Movement Analysis of
the Spike in Fistball
Andreas Bund 1, *, Saeed Ghorbani 2 and Franziska Rathjens 3
1
2
3
*
Institute of Applied Educational Sciences, University of Luxembourg, Esch-s.-Alzette L-4365,
Luxembourg, Luxembourg
Department of Physical Education and Sport Science, Aliabad Katoul Branch, Islamic Azad University,
Aliabad Katoul, Iran; [email protected]
Institute of Sport Science, University of Oldenburg, Oldenburg D-26129, Germany;
[email protected]
Correspondence: [email protected]; Tel.: +352-46-66-44-9731
Academic Editor: Eling Douwe de Bruin
Received: 16 July 2016; Accepted: 30 November 2016; Published: 2 December 2016
Abstract: Due to its relevancy to point scoring, the spike is considered as one of the most important
skills in fistball. Biomechanical analyses of this sport are very rare. In the present study, we performed
a three-dimensional kinematic analysis of the fistball spike, which helps to specify performance
parameters on a descriptive level. Recorded by four synchronized cameras (120 Hz) and linked to the
motion capture software Simi Motion® 5.0, three female fistball players of the second German league
(24–26 years, 1.63–1.69 m) performed several spikes under standardized conditions. Results show
that the segment velocities of the arm reached their maximum successively from proximal to distal,
following the principle of temporal coordination of single impulses. The wrist shows maximum
speed when the fist hits the ball. The elbow joint angle performs a rapid transition from a strong
flexion to a (almost) full extension; however, the extension is completed after the moment of ball
impact. In contrast, the shoulder joint angle increases almost linearly until the fistball contact and
decreases afterward. The findings can be used to optimize the training of the spike.
Keywords: movement analysis; biomechanics; fistball
1. Introduction
Fistball is a traditional ball game in which two teams with five players play against each other
on a field separated by a net. Similar to volleyball, the idea of the game is to hit the ball with a part
of the arm or the fist in such a way across the net that it is difficult or impossible for the other team
to return it. Within the team, the ball can be played directly or indirectly by three different players,
thus, the standard sequence consists of three elements: defense (trapping the ball), passing and offense
(attacking shot/spike). Fistball is played mainly in Europe and South America, but is becoming more
and more popular in Africa, Asia and North America too. At the World Games as well as at Continental
Championships, national teams play against each other. The organization of these events is done by
the International Fistball Association (IFA) [1], which was co-founded in 1960 by Germany, Brazil,
Switzerland, Austria and Italy.
In recent years, an increasing professionalization on the athletic and organizational levels can
be observed. The field of sport science, however, has paid little attention to fistball so far and very
few scientific papers on the topic are available. Besides older publications on methodological aspects
(e.g., [2,3]), studies often deal with the risk of injury in fistball [4,5]. Runer et al. [5], for example, found
over the course of 12 months an overall injury rate of 12.2 injuries/1000 h of exposure. The most
common types of injury were abrasions and contusions. An exception is provided by Leser [6], who has
developed and evaluated a computer-based observation system for fistball.
Sports 2016, 4, 55; doi:10.3390/sports4040055
www.mdpi.com/journal/sports
Sports 2016, 4, 55
2 of 11
Of particular interest to the present work is a study of Soeser and Schwameder [7]. Focusing
on the hitting side of the body, they performed a three-dimensional kinematic analysis of the fistball
serve. The results can be summarized as follows: (1) The execution of the serve shows in many
characteristics (e.g., ball drop, turning back the hitting side shoulder, and body extension in the
moment of ball impact) a high degree of standardization, i.e., the execution is very similar across
different players and performance levels; (2) A kinematic chain can be seen that typically occurs in
throwing movements, i.e., hip, shoulder, elbow and wrist/fist successively reach their maximum speed
and become slowed down in order to transfer and maximize the impulses from proximal to distal body
segments. As expected, players from the national team reached significantly higher body segments
and ball velocities then players from federal or regional leagues.
Similar to the serve, the spike is a typical hitting movement whose main phase is performed
above the head. It is therefore expected that the results of Soeser and Schwameder [7]—e.g., in terms
of the kinematic chain and the temporal coordination of impulses—are also valid for the spike.
However, it should be noted that the spike starts with a one-legged kick-off jump in the air and, thus,
is performed without ground contact. Consequently, the biomechanical conditions of serve and spike
differ significantly. In addition, the spike usually coordinates with the trajectory of the passed ball
and therefore has to be available and executed under highly variable conditions. Generally speaking,
a good technical execution of the spike requires that the player has the right distance to the ball and
that he or she is able to accelerate strongly and temporally coordinate the segments of the hitting side
of his or her body. At the moment of ball impact, speed and trajectory of the ball are determined by
the velocity, rotation and inclination of the fist, the point of ball contact, and the movement direction of
the hitting arm [7].
The lack of biomechanical and especially kinematic analyses of motor skills used in fistball was
the reason to conduct this study. The following aims were determined:
(1)
(2)
The description of the spike in fistball based on three-dimensional recorded kinematic parameters
with high frequency.
The identification and weighting of performance-determining factors by a comparative analysis
of kinematic parameters of different players.
Unlike Soeser and Schwameder [7], both sides of the body were included in the analysis.
2. Method
2.1. Participants and Task
Three female fistball players aged from 24 to 26 years old who play in clubs in Germany’s 2nd
league on the attack position took part in this study. The players were right-handed, 1.63 to 1.69 m
tall and weighed at the time of the study from 55 to 72 kg. The task was to perform from a short
run-out a spike against a fist ball attached to a rope from the ceiling, just above the fistball net that was
stretched to a height of 1.90 following the competition rules. The ball height varied slightly depending
on the body size of the player and was previously determined in several trials. Table 1 shows the
anthropometric measures of the players as well as the individual ball heights. The procedure of the
study was designed according to the guidelines of the Helsinki declaration. All players provided
informed consent prior to the study.
Table 1. Anthropemetric measures of the participants and individual ball heights.
Participant
Body Weight (kg)
Body Height (m)
Ball Height (m)
Player 1
Player 2
Player 3
55
63
72
1.62
1.64
1.69
1.94
1.96
1.96
Sports 2016, 4, 55
Sports 2016, 4, 55
3 of 11
3 of 11
2.2.
2.2. Data
Data Recording
Recording
The
spike(including
(including
the run-out)
was recorded
by four synchronized
mutually
The spike
the run-out)
was recorded
by four synchronized
and mutuallyand
perpendicular
perpendicular
positioned
cameras
1.65 m(Basler
high tripods
Scout)rate
withofa 120
frame
of 120
positioned
cameras
on 1.65
m highon
tripods
Scout) (Basler
with a frame
Hz.rate
Based
on Hz.
the
Based
on
the
dimensions
of
a
volleyball
court,
four
cameras
with
an
oblique
view
filmed
the
player,
dimensions of a volleyball court, four cameras with an oblique view filmed the player, two cameras
two
from
thetwo
front
andthe
twoback.
from the
The recorded
space
was calibrated
fromcameras
the front
and
from
Theback.
recorded
space was
calibrated
using using
a cubea cube
with
with
2 ×m.
2.5Figure
m. Figure
1 shows
the exact
position
the cameras
and
the distances
among
each
2 × 22××2.5
1 shows
the exact
position
of theofcameras
and the
distances
among
each other.
other.
Recordings
were
made
in
the
gymnasium
of
the
University
of
Oldenburg
(Germany).
Each
Recordings were made in the gymnasium of the University of Oldenburg (Germany). Each player
player
completed
4 tohitting
6 trialsthe
hitting
the
ball in a dynamic
fluent movement.
For the subsequent
completed
4 to 6 trials
ball in
a dynamic
and fluentand
movement.
For the subsequent
kinematic
kinematic
totalpoints
of 7 body
points
of the
players
knee, hip,
elbow,and
shoulder,
and
analysis, a analysis,
total of 7 abody
of the
players
(foot,
ankle,(foot,
knee,ankle,
hip, elbow,
shoulder,
wrist) were
wrist)
were
bilaterally
covered
with
light-reflective
markers.
Five
of
these
markers
were
affixed
bilaterally covered with light-reflective markers. Five of these markers were affixed directly to the
directly
to the
the foot
skin,and
onlythe
thehip
foot
and thewere
hip markers
were
on the
shoe and ashirt,
tight-fitting
shirt,
skin, only
markers
placed on
theplaced
shoe and
a tight-fitting
respectively.
respectively.
The exactoflocalization
markersinisTable
described
The
exact localization
the markersofisthe
described
2. in Table 2.
Figure
Figure 1.
1. Measurement
Measurement setup.
setup.
Table
2. Body
points included
included in
in the
the kinematic
kinematic analysis.
analysis.
Table 2.
Body points
Body Point
Body Point
Foot
Foot
Ankle
Ankle
Knee
Knee
Hip
Hip
Elbow Elbow
ShoulderShoulder
Wrist
Wrist
Body Side
Body Side
bilaterally
bilaterally
bilaterally
bilaterally
bilaterally
bilaterally
bilaterally
bilaterally
bilaterally
bilaterally
bilaterally
bilaterally
bilaterally
bilaterally
Localization
Basis Metatarsales V
Basis Metatarsales
MalleolusVlateralis
Malleolus lateralis
Epicondylus
Epicondylus
lateralis lateralis
femoris femoris
Trochanter
Trochanter
major major
Epicondylus
lateralis lateralis
humeri humeri
Epicondylus
Acromio-clavicular
join
Acromio-clavicular
join
Os capitatum
Os capitatum
Localization
2.3. Data
Data Processing
Processing
2.3.
The trajectories
of the
body points
digitized using
using the
the
The
trajectories of
the above-mentioned
above-mentioned body
points were
were automatically
automatically digitized
motion
capture
software
Simi
Motion
5.0
(Simi
Reality
Motion
Systems,
Unterschleißheim,
Germany).
motion capture software Simi Motion 5.0 (Simi Reality Motion Systems, Unterschleißheim, Germany).
The
of the
the third
third last
last step
step (ground
(ground contact)
contact) was
was defined
defined as
as starting
starting point
point of
of the
the movement;
movement;
The beginning
beginning of
the seventh
frame after
In order
order to
to compare
the
seventh frame
after the
the ball
ball contact
contact was
was considered
considered as
as the
the end
end point.
point. In
compare the
the
recordings
of
the
various
trials
and
players,
they
were
standardized
in
the
post-processing
on
each
recordings of the various trials and players, they were standardized in the post-processing on each
100
100 frames
frames exactly.
exactly. Specifically,
Specifically,the
thefollowing
followingkinematic
kinematicparameters
parameterswere
werecalculated:
calculated:
•
Path velocities of the hip, shoulder, elbow and wrist (unilateral, for the hitting side of the body)
Sports 2016, 4, 55
4 of 11
Sports 2016, 4, 55
•
•
•
••
••
4 of 11
Path velocities of the hip, shoulder, elbow and wrist (unilateral, for the hitting side of the body)
Joint angles of elbow, shoulder (unilateral, for the hitting side of the body), knee and hip
Joint angles of elbow, shoulder (unilateral, for the hitting side of the body), knee and hip (bilateral)
(bilateral)
Height
Heightof
ofthe
thehip
hipand
andshoulder
shoulder(bilateral)
(bilateral)
Center
of
gravity
(CoG)
Center of gravity (CoG)
The temporal
temporal coordination
coordination of
of path
path velocities
velocities and
and accelerations
accelerations of
of the
the hip,
hip, elbow,
elbow,shoulder
shoulder and
and
The
wrist of
ofthe
thehitting
hitting side
side of
ofthe
thebody
body isispivotal
pivotal to
tothe
theimpulse
impulse transfer
transfer from
from the
the wrist
wrist to
to the
the ball
ball and
and
wrist
largely determine
determine its
its speed.
speed. The
The joint
joint angles
angles were
were calculated
calculated in
in the
the clockwise
clockwise direction
direction through
through the
the
largely
SimiMotion
Motionsoftware
softwareby
byapplying
applyingEquation
Equation(1)
(1)on
oneach
eachsingle
singleframe.
frame. The
The basis
basis was
was the
the x-coordinates
x-coordinates
Simi
ofthe
thebody
bodypoints,
points,that
thatis,
is,the
theangles
angleswere
wereanalyzed
analyzedin
inthe
thesagittal
sagittalplane.
plane. The
The elbow
elbow joint
joint angle
angle was
was
of
determined by
by the
the elbow
elbow as
as apex
apex and
and the
the wrist
wrist and
and shoulder
shoulder as
as end
end points
points of
of the
the corresponding
corresponding
determined
triangle-legs. The
The shoulder
shoulder joint
joint angle
angle was
wasdefined
definedby
bythe
theshoulder
shoulderas
asapex
apexand
andthe
theelbow
elbowand
andhip
hipas
as
triangle-legs.
endpoints.
points.The
Thehip
hipangle
anglewas
wasmeasured
measuredwith
withthe
thehip
hipas
asapex
apexand
andthe
theshoulder
shoulderand
andknee
kneeas
asend
endpoints.
points.
end
Finally,the
theknee
kneewas
wasused
usedas
asapex
apexand
andthe
theankle
ankleand
andhip
hipasasend
endpoints
pointstotodetermine
determinethe
theknee
kneejoint
joint
Finally,
angle. Figure
Figure 2 demonstrates the joint
angle.
joint angles
angles and
andtheir
theircorresponding
correspondingvectors.
vectors.
Figure
Figure 2.
2. Joint
Joint angles
angles calculated
calculated in
in the
the study.
study.
joint
jointangle
angle= =acos(scalar
acos(scalarproduct(v1,
product(v1,v2)/(norm(v1)
v2)/(norm(v1)×× norm(v2)))
(1)
(1)
The
The hip
hip and
and shoulder
shoulder heights
heights were
were calculated
calculated from
from the
the z-coordinates,
z-coordinates, i.e.,
i.e., in
in the
the frontal
frontal plane,
plane,
and
and indicate
indicate the
the players
players posture
posture during
during run-up,
run-up, take-off
take-offand
andhitting
hittingphases.
phases. The
The assessment
assessment of
of the
the
CoG
CoGwas
wasmade
madeby
bythe
theSimi
SimiMotion
Motionsoftware
softwareaccording
accordingto
tothe
theHanavan
Hanavanhuman
humanbody
bodymodel.
model.The
TheCoG
CoG
parameter
thethe
run-up
velocity
(horizontal
CoG-velocity
before
the take-off)
and
parameterwas
wasused
usedtotodetermine
determine
run-up
velocity
(horizontal
CoG-velocity
before
the take-off)
the
heightheight
(maximum
CoG-height
in flight
at last ground
contact).contact).
Finally,
andjump
the jump
(maximum
CoG-height
inphase—CoG-height
flight phase—CoG-height
at last ground
the
initialthe
path
velocity
the ball
calculated
as a key parameter
of the spike
movement.
Finally,
initial
path of
velocity
ofwas
the ball
was calculated
as a key parameter
of the
spike movement.
Sports 2016,
2016, 4,
4, 55
55
Sports
of 11
11
55 of
3. Results
3. Results
3.1.
3.1. Kinematic
Kinematic Description
Description of
of the
the Spike
Spike
Figure
Figure 33 shows
shows as
as an
an example
example the
the kinegram
kinegram of
of player
player 1.
1. The
The sequence
sequence shows
shows how
how the
the player
player
performs
the
run-up,
the
take-off,
the
backwards
torsion
of
shoulder
and
arm
on
the
hitting
of
performs the run-up, the take-off, the backwards torsion of shoulder and arm on the hitting sideside
of the
the
body,
the
subsequent
acceleration
of
these
body
segments
in
the
direction
of
the
ball,
the
ball
body, the subsequent acceleration of these body segments in the direction of the ball, the ball impact
impact
with simultaneous
body extension
the preparation
of the landing.
The run-up
velocities
with simultaneous
body extension
and theand
preparation
of the landing.
The run-up
velocities
differ
differ
only
minimally
and
range
for
all
players
from
3.31
m/s
to
3.49
m/s.
The
maximum
jump
only minimally and range for all players from 3.31 m/s to 3.49 m/s. The maximum jump heightheight
varies
varies
between
andm.0.43
Theintervals
time intervals
between
take-off,
ball impact
and landing
are
between
0.32 m0.32
andm
0.43
Them.time
between
take-off,
ball impact
and landing
are again
again
relatively
they range
s to 0.16
s and
to 0.20
s, respectively.
Thus,
similar
relatively
stable;stable;
they range
fromfrom
0.26 s0.26
to 0.16
s and
0.350.35
s tos0.20
s, respectively.
Thus,
similar
to
to
the
fistball
serve
[7],
we
also
found
for
the
spike
only
small
inter-individual
variance
regarding
the fistball serve [7], we also found for the spike only small inter-individual variance regarding the
the
kinematic
of movement
movement phases.
phases.However,
However,ititshould
should
noted,
that
players
who
participated
in
kinematic of
bebe
noted,
that
thethe
players
who
participated
in this
this
study
had
a
very
similar
level
of
performance.
study had a very similar level of performance.
Figure
Figure 3.
3. Kinegram
Kinegram of
ofthe
thefistball
fistball spike
spike (player
(player 1),
1), with
with timeline
timeline (s).
(s). higher
higherresolution
resolution picture.
picture. Comma
Comma
should
should be
be changed
changed into
into dot.
dot.
3.2. Path Velocities
3.2. Path Velocities
Figures 4–6 show the path velocities of the shoulder, elbow and wrist joint on the hitting side of
Figures 4–6 show the path velocities of the shoulder, elbow and wrist joint on the hitting side of
the body. Aside from inter-individual similarities, the data indicate various individual characteristics
the body. Aside from inter-individual similarities, the data indicate various individual characteristics
of the players’ movement. During run-up, the path velocity of the right shoulder rises and falls
of the players’ movement. During run-up, the path velocity of the right shoulder rises and falls
according to the individual running cadence of the player. The cycle varies considerably, and the
according to the individual running cadence of the player. The cycle varies considerably, and the
differences between the velocity minima and maxima are significant. Player 1 shows clear differences
differences between the velocity minima and maxima are significant. Player 1 shows clear differences
concerning these parameters. Common to all players is that they rapidly accelerate the shoulder just
concerning these parameters. Common to all players is that they rapidly accelerate the shoulder just
before the ball impact in order to reach maximum speed. After the impact, the shoulder is strongly
before the ball impact in order to reach maximum speed. After the impact, the shoulder is strongly
slowed down in a short time period.
slowed down in a short time period.
The path velocities of the elbow and the wrist joint can be described analogous in essential points.
The path velocities of the elbow and the wrist joint can be described analogous in essential
During the run-up phase, the path velocity of the elbow again shows a cyclical pattern. Overall, the
points. During the run-up phase, the path velocity of the elbow again shows a cyclical pattern.
cycles vary inter-individually less than in the shoulder; however, the minimum–maximum
Overall, the cycles vary inter-individually less than in the shoulder; however, the minimum–maximum
differences are significantly smaller for player 3 than for players 1 and 2. The very fast acceleration
differences are significantly smaller for player 3 than for players 1 and 2. The very fast acceleration
of the elbow joint as well as the deceleration after the ball contact are also very similar between the
of the elbow joint as well as the deceleration after the ball contact are also very similar between the
players. Just before the ball impact, the players reach an almost equal velocity maximum. In contrast
players. Just before the ball impact, the players reach an almost equal velocity maximum. In contrast
to this, the speed of the wrist reaches its maximum exactly when the fist hits the ball; before and after
to this, the speed of the wrist reaches its maximum exactly when the fist hits the ball; before and after
that moment a strong positive or negative acceleration can be observed. While running-up towards
that moment a strong positive or negative acceleration can be observed. While running-up towards the
the net, the path velocity of the wrist again oscillates according to the players’ individual rhythm. As
net, the path velocity of the wrist again oscillates according to the players’ individual rhythm. As with
with the elbow joint, these cycles occur “flatter” for player 3 than for players 1 and 2. Furthermore,
the elbow joint, these cycles occur “flatter” for player 3 than for players 1 and 2. Furthermore, player 3
player 3 does not achieve such a high maximum speed as the other two players.
does not achieve such a high maximum speed as the other two players.
Sports 2016, 4, 55
Sports 2016, 4, 55
Sports 2016, 4, 55
Sports 2016, 4, 55
6 of 11
6 of 11
6 of 11
6 of 11
Figure 4. Inter-individual comparison of the path velocities of the right shoulder joint (100 frames;
Figure
4. Inter-individual
velocitiesofofthe
theright
rightshoulder
shoulder
joint
(100
frames;
Figure
4. Inter-individualcomparison
comparisonof
of the
the path
path velocities
joint
(100
frames;
dotted
line
= ball impact). comparison of the path velocities of the right shoulder joint (100 frames;
Figure
4.
Inter-individual
dotted
lineline
= ball
impact).
dotted
= ball
impact).
dotted line = ball impact).
Figure 5. Inter-individual comparison of the path velocities of the right elbow joint (100 frames; dotted
Figure 5. Inter-individual comparison of the path velocities of the right elbow joint (100 frames; dotted
Figure
5.
Inter-individual
comparison
of the
theright
rightelbow
elbowjoint
joint
(100
frames;
dotted
line
= ball
impact).
Figure
5. Inter-individual
comparisonofofthe
thepath
pathvelocities
velocities of
(100
frames;
dotted
line
= ball
impact).
line line
= ball
impact).
= ball impact).
Figure 6. Inter-individual comparison of the path velocities of the right wrist joint (100 frames; dotted
Figure 6. Inter-individual comparison of the path velocities of the right wrist joint (100 frames; dotted
line
= ball
impact).
Figure
6. Inter-individual
comparisonofofthe
thepath
pathvelocities
velocities of
(100
frames;
dotted
line
= ball
impact).
Figure
6.
Inter-individual
comparison
of the
theright
rightwrist
wristjoint
joint
(100
frames;
dotted
line = ball impact).
line = ball impact).
Sports 2016, 4, 55
7 of 11
Sports 2016, 4, 55
7 of 11
The velocity maxima of the captured body points and the resulting ball speed are shown in
The velocity maxima of the captured body points and the resulting ball speed are shown in Table
Table 3. For all players, the velocity maxima continuously increase from proximal (hip) to distal (wrist)
3. For all players, the velocity maxima continuously increase from proximal (hip) to distal (wrist)
body points. This data suggests that in this case a kinematic chain is operating, in which the speed
body points. This data suggests that in this case a kinematic chain is operating, in which the speed
maxima of the involved body segments are placed in sequence so that the (distal) end link of the chain
maxima of the involved body segments are placed in sequence so that the (distal) end link of the
experiences maximum acceleration.
chain experiences maximum acceleration.
Table 3. Maxima of path velocities and resulting speed of the ball (m/s).
Table 3. Maxima of path velocities and resulting speed of the ball (m/s).
Title
Player 1
Player 2
Player 3
Hip Title
Player 1
4.41
Player 2
4.22
Player 3
3.82
Hip
Shoulder
Shoulder
4.415.23
4.224.65
3.82
4.95
5.23
4.65
4.95
Elbow
Elbow
8.98
8.98
8.28
8.28
8.37
8.37
Wrist Wrist
Ball
11.79 11.79
12.45
11.43 11.43
12.19
8.65
10.89
8.65
Ball
12.45
12.19
10.89
Of course,
course, the
the existence
existence of
of aakinematic
kinematic chain
chain becomes
becomes clearer
clearer ifif the
the path
path velocities
velocities of
of the
therelevant
relevant
Of
body points
This
is done
in Figure
7 using
the example
of player
who achieved
body
points are
arepresented
presentedtogether.
together.
This
is done
in Figure
7 using
the example
of1,player
1, who
the highest
speedball
with
12.45with
m/s.
achieved
theball
highest
speed
12.45 m/s.
Figure
Figure 7.
7. Path
Path velocities
velocitiesof
ofthe
thebody
bodypoints
pointson
onthe
thehitting
hittingside
sideof
ofthe
thebody
bodyof
of player
player11 (100
(100 frames;
frames;
dotted
dotted line
line ==ball
ballimpact).
impact).
3.3.
3.3. Joint
Joint Angles
Angles
Compared
Compared to
to the
the velocity
velocity data,
data, the
the course
course of
of shoulder
shoulder and
and elbow
elbow angles
angles (Figures
(Figures 88 and
and 9)
9)clearly
clearly
reveals
greater
inter-individual
differences.
This
concerns
in
particular
the
players’
run-up
and
takereveals greater inter-individual differences. This concerns in particular the players’ run-up and take-off
off
phases.
Player
2, example,
for example,
her playing
(“hitting”)
arm during
the run-up
to the
phases.
Player
2, for
keepskeeps
her playing
(“hitting”)
arm during
the run-up
close toclose
the body
so
body
so
that
the
shoulder
angle
slightly
opens
and
closes
according
to
the
running
cadence.
During
that the shoulder angle slightly opens and closes according to the running cadence. During and after
and
after atake-off,
a very
fast extension
of the shoulder
joint
to anofangle
of◦ca.
130°place,
takes which
place, which
take-off,
very fast
extension
of the shoulder
joint to an
angle
ca. 130
takes
just as
just
as quickly
into a flexion
strong flexion
at the moment
of ball contact.
For1player
1 and especially
quickly
evolvesevolves
into a strong
at the moment
of ball contact.
For player
and especially
player 3,
player
3, the
angle
course issimilar,
basically
similar,the
however,
theand
extension
flexion
of the
shoulder
the angle
course
is basically
however,
extension
flexion and
of the
shoulder
angle
occurs
angle
occursrange
in a smaller
and less
abruptly.player
Concerning
player
3, Figure
7 illustrates
that
the
in a smaller
and lessrange
abruptly.
Concerning
3, Figure
7 illustrates
that
the playing
arm
is
◦
playing
arm
is
stretched
far
out
when
she
is
moving
toward
the
net
(shoulder
angle
about
40°).
That
stretched far out when she is moving toward the net (shoulder angle about 40 ). That might be partly
might
be partly
responsible
the low
jump
of 0.34
cm,
becausethe
inarm
this can
position
the
can
responsible
for the
low jumpfor
height
of 0.34
cm,height
because
in this
position
hardly
bearm
used
as
hardly
be used
as an additional
element.
In players,
contrastthere
to the
other
players,flexion
there of
is the
no
an additional
swinging
element. Inswinging
contrast to
the other
is no
significant
significant
flexion
of the
shoulder
angle
whenthe
shehip
hits
the decreases
ball. Instead,
the hip angle
shoulder angle
when
she hits
the ball.
Instead,
angle
considerably
at thisdecreases
moment,
considerably
moment,
thus,
player
is perform
using thethe
entire
upper
body
to spike.
perform the hitting
thus, player 3atis this
using
the entire
upper
body3 to
hitting
phase
of the
phaseInofrespect
the spike.
to the elbow angle, even greater differences occur between the players, only during
In
respect
to
elbow
even greater
the players,
during
and after the ball the
impact
do angle,
the profiles
tend to differences
match eachoccur
other.between
Previously,
the arm only
posture
was
and
the to
ball
do the
profiles
each relation
other. Previously,
theand
armtake-off
postureaction.
was
veryafter
specific
theimpact
individual
and
notablytend
hadto
nomatch
consistent
to the run-up
very specific to the individual and notably had no consistent relation to the run-up and take-off action.
Player 1 successively reduces the elbow angle during the run-up phase; only the take-off leads to a
Sports 2016, 4, 55
8 of 11
Sports 2016, 4, 55
Sports 2016, 4, 55
8 of 11
8 of 11
Player 1 successively reduces the elbow angle during the run-up phase; only the take-off leads to
temporary
a temporaryextension.
extension.Concerning
Concerningthe
thecountermovement
countermovementofofthe
theplaying
playingarm,
arm,the
the elbow
elbow angle
angle first
temporary extension. Concerning the countermovement of the playing arm, the elbow angle first
oscillates slightly followed
followed by an explosive stretching
stretching until
until ball contact;
contact; however, the ball contact is
oscillates slightly followed by an explosive stretching until ball contact; however, the ball contact is
made shortly before the maximum extension of the arm is reached. In contrast to this, player 2 starts
made shortly before the maximum extension
of the arm is reached. In contrast to this, player 2 starts
◦ ), which
(<120°),
with an extended elbow angle (<120
which is
is then
then gradually
gradually flexed
flexed on
on after
after the
the take-off
take-off and
and finally
finally
with an extended elbow angle (<120°), which is then gradually flexed on after the take-off and finally
opened again to hit the ball. Striking
Striking (and
(and certainly
certainly unusual)
unusual) in
in this
this case
case is the short-term
short-term (slight)
(slight)
opened again to hit the ball. Striking (and certainly unusual) in this case is the short-term (slight)
“intermediate” extension
extension during the jumping phase (about frame 75). Finally, player 3 shows a third
“intermediate”
“intermediate” extension during the jumping phase (about frame 75). Finally, player 3 shows a third
variant: During run-up and take-off the elbow is bent more and more. The extension of the arm takes
variant: During run-up and take-off the elbow is bent more and more. The extension of the arm takes
and is
is aa bit
bit less
less pronounced
pronounced than
than for
for the
the players
players 11 and
and 2.
2.
place only before and after the ball impact and
place only before and after the ball impact and is a bit less pronounced than for the players 1 and 2.
Figure
8.
Inter-individual
comparison
of the
joint
angles
of
the
right
shoulder
joint
(100
frames;
dotted
Figure 8.
8. Inter-individual
Inter-individual comparison
comparison of
of
the joint
joint angles
angles of
of the
the right
right shoulder
shoulder joint
joint (100
(100 frames;
frames; dotted
dotted
Figure
the
line
=
ball
impact).
line == ball
ball impact).
impact).
line
Figure 9. Inter-individual comparison of the joint angles of the right elbow joint (100 frames; dotted
Figure
9.
comparison
Figure
9. Inter-individual
Inter-individual
comparison of
of the
the joint
joint angles
angles of
of the
the right
right elbow
elbow joint
joint (100
(100 frames;
frames; dotted
dotted
line
= ball
impact).
line
line == ball
ball impact).
impact).
Are joint angles and path velocities systematically linked with each other, e.g., at the time of ball
Are joint angles and path velocities systematically linked with each other, e.g., at the time of ball
Are Higher
joint angles
andof
path
velocities
systematically
with each
other,
e.g.,inathigher
the time
of
contact?
degrees
extension
and,
thus, greater linked
body angles
should
result
path
contact? Higher degrees of extension and, thus, greater body angles should result in higher path
ball contact?
Higher
extension
thus,the
greater
body
angles
result
in higher
velocities
of the
distal degrees
points ofofthe
body. Ofand,
course,
question
cannot
be should
answered
reliably
with
velocities of the distal points of the body. Of course, the question cannot be answered reliably with
path
velocities
of the distal
the body.
Of course,
the question
cannot be
answered
reliably
the very
small sample
used points
in this of
study;
however,
the available
data supports
this
assumption:
For
the very small sample used in this study; however, the available data supports this assumption: For
with
the very
small
sample
usedwith
in this
the available
supports wrist
this assumption:
example,
player
1 hits
the ball
an study;
elbow however,
angle of 129°
achievingdata
a maximum
velocity of
example, player 1 hits the ball with an elbow angle of 129° achieving
a maximum wrist velocity of
For example,
player
1 hits
the ball the
with
an elbow
ofwhich
129◦ achieving
maximum
wrist
velocity
11.79
m/s, while
player
3 stretches
elbow
only angle
to 115°,
leads to aawrist
velocity
of only
8.37
11.79 m/s, while player 3 stretches the elbow only to 115°, which◦ leads to a wrist velocity of only 8.37
of
11.79
m/s,
while
player
3
stretches
the
elbow
only
to
115
,
which
leads
to
a
wrist
velocity
of
m/s.
m/s.
only 8.37 m/s.
Sports 2016, 4, 55
Sports 2016, 4, 55
9 of 11
9 of 11
3.4. Height of Hip, Shoulder and CoG
3.4. Height of Hip, Shoulder and CoG
Overall, the bilateral measurement of the height profiles of hip and shoulder shows only small
Overall, the bilateral measurement of the height profiles of hip and shoulder shows only small
inter-individual variance: During run-up, hip and shoulder move on both sides at the same level.
inter-individual variance: During run-up, hip and shoulder move on both sides at the same level.
During
the take-off of the player, both pivot points are raised; however, on the (right) hitting side of
During the take-off of the player, both pivot points are raised; however, on the (right) hitting side of
thethe
body
they
are
raised
due to
tothe
theincreasing
increasingextension
extensionofof
this
part
body
they
are
raisedhigher
higherthan
thanon
onthe
the left
left side
side due
this
part
of of
thethe
body.
At
the
moment
of
ball
contact,
the
right
hip
and
shoulder
reach
the
highest
point
in
all
players.
body. At the moment of ball contact, the right hip and shoulder reach the highest point in all players.
However,
thethe
difference
sideand
andthe
theleft
leftside
sideofofthe
the
body
vary
from
However,
differencebetween
betweenthe
theheight
height on
on the
the right
right side
body
vary
from
player
to
player.
In
relation
to
the
hip,
for
player
1
this
was
ca.
9
cm,
for
player
2,
ca.
16
cm
and
player to player. In relation to the hip, for player 1 this was ca. 9 cm, for player 2, ca. 16 cm and forfor
player
3, 3,
only
ca.ca.5 5cm.
moretilted
tiltedduring
duringthe
thehitting
hitting
phase
than
player
only
cm.InInother
otherwords,
words, player
player 2 is clearly
clearly more
phase
than
thethe
other
two
players.
theright
rightand
andleft
leftshoulder
shoulder
about
to 18
other
two
players.The
Theheight
heightdifference
difference between
between the
is is
about
16 16
to 18
cm.cm.
Figure
1010
illustrates
left and
andright
rightshoulder
shoulderofofplayer
player1.1.
Figure
illustratesasasananexample
examplethe
theheight
height profile
profile of the left
10. Height profies of right and left shoulder of player 1 (100 frames; dotted line = ball
FigureFigure
10. Height
profies of right and left shoulder of player 1 (100 frames; dotted line = ball impact).
impact).
The
measurement
onthe
thehorizontal
horizontalvelocity
velocity
while
running-up
The
measurementofofthe
theCoG
CoGprovides
provides information
information on
while
running-up
and
onon
thethe
jumping
velocitiesjust
justbefore
beforethe
thetake-off
take-offare
areatat
3.43
m/s,
and
jumpingheight
heightofofthe
theplayers.
players. The
The run-up
run-up velocities
3.43
m/s,
3.49
m/s
and
3.31
m/s
and
thus
very
little.
Compared
with
this,
the
differences
in
the
jump
heights
3.49 m/s and 3.31 m/s and thus very little. Compared with this, the differences in the jump heightsare
greater:
0.43 m,0.43
0.32m,
m 0.32
and m
0.34
m.0.34
Duem.
to the
it issize,
not possible
to determine
a correlation
are greater:
and
Duesample
to the size,
sample
it is not possible
to determine
a
between
run-up
velocity
andvelocity
jump height.
In addition,
CoG height
profile
is profile
very similar
correlation
between
run-up
and jump
height. Inthe
addition,
the CoG
height
is veryfor
allsimilar
players.
for all players.
4. Discussion
4. Discussion
Based
onon
thethe
fact
that
only
very
little
sport
science
research
onon
fistball
is is
available
toto
date,
the
aim
Based
fact
that
only
very
little
sport
science
research
fistball
available
date,
the
of aim
the present
study was
to was
carrytoout,
on out,
a descriptive
level, a level,
three-dimensional
kinematic
analysis
of the present
study
carry
on a descriptive
a three-dimensional
kinematic
spike in
fistball.
Using camera-based
motionsoftware
capture software
(Simi 5.0),
Motion
5.0),
of analysis
the spikeofinthe
fistball.
Using
camera-based
motion capture
(Simi Motion
kinematic
kinematic such
parameters
such as
velocity,
angle and of
trajectory
of body
different
body
points
were bilaterally
parameters
as velocity,
angle
and trajectory
different
points
were
bilaterally
recorded,
recorded,
and finallyBased
evaluated.
on the comparison
of several
players, weand
identified
digitized
anddigitized
finally evaluated.
on theBased
comparison
of several players,
we identified
weighted
and
weighted
parameters
that
significantly
affect
the
spike
performance.
The
results
can
be
parameters that significantly affect the spike performance. The results can be summarized as follows:
summarized
as
follows:
The kinematic pattern of movement phases shows only a small range of inter-individual variance.
kinematic
pattern of
movement
phases take-off
shows only
a small are
range
of inter-individual
Run-upThe
velocity
and temporal
sequence
of run-up,
and landing
relatively
similar across
variance.
Run-up
velocity
and
temporal
sequence
of
run-up,
take-off
and
landing
are relatively
the players.
similar across the players.
1.
The analysis of the path velocities of several body points demonstrates individual styles in the
1. The analysis of the path velocities of several body points demonstrates individual styles in the
execution of the fistball spike; they concern, in particular, timing and range of velocity changes.
execution of the fistball spike; they concern, in particular, timing and range of velocity changes.
2.
In terms of a kinematic chain, the hip, shoulder, elbow and wrist joint reach their maximum
2. In terms of a kinematic chain, the hip, shoulder, elbow and wrist joint reach their maximum
speed
Thedeceleration
decelerationofofeach
eachmore
more
proximal
joint
speedsuccessively
successivelyand
andtemporarily
temporarily coordinated.
coordinated. The
proximal
joint
causes the impulse to be (partly) transferred to the next joint or limb, so that the wrist, as the end
of the chain, receives the highest speed.
Sports 2016, 4, 55
3.
4.
10 of 11
The courses of the joint angles, especially of that of the hitting arm, differ considerably from
player to player. As expected, extended elbow angles tend to be associated with greater speed of
distal body points.
The height profile of hip, shoulder and CoG is basically similar for all players. Due to the
extension during the hitting phase, the points on the hitting side of the body are higher than
the corresponding points on the opposite side. The hip axis is clearly more inclined than the
shoulder axis.
Overall, our results are in line with those of Soeser and Schwameder [7] on the fistball serve,
taking into account that the latter of course found higher values for joint and ball speed due to the
gender (male) and level (semi-professional) of their participants (e.g., ball speed: 24.69–28.01 m/s
compared to 12.19–12.69 m/s in our study). The impulse transfer within a kinematic chain was clearly
demonstrated in both studies. It occurs typically in such throwing movements. As expected, fistball is,
in this regard, comparable to other sports in which an object has to be maximally accelerated by
a strike (short contact) or a throw (long contact): the biomechanical principle of optimal coordination
of impulses applies [8]. Specifically, this is also documented for the spike in volleyball [9], the goal
shot in handball [10–12], and the tennis serve [13]. Another commonality between the present study
and that of [7] is the relatively strong lateral inclination of the upper body during the hitting phase,
which is reflected in significant height differences between left and right hips and shoulders.
In connection with the coordination of impulses, positive relationships between certain parameters
are to be expected, firstly between path velocities and resulting ball speed, and secondly, between joint
angles and path velocities. In both cases, our findings indicate such a positive relationship; however,
it is not supported by statistical evidence because of the small sample size. This reflects a serious
weakness of the present study: The comparative analysis had to remain on the descriptive level;
an inferential statistical analysis was not possible. A larger sample consisting of players of different
performance levels, as used by Soeser and Schwameder [7], would have been desirable in order to
get statistical evidence about the relationship of various kinematic parameters and their influence
on the fistball spike performance. Another problem comes from the limited number of body points
that were included in the analysis. Although we have measured the trajectories of the main joint
angles on both sides of the body, some important aspects of the spike movement were not considered.
For example, the prono-supination of the hitting arm—i.e., the internal-external rotation of the arm
during the execution of the spike—was not taken into account. Neither was the angle between the
player’s body and the hall floor as a measure for the body inclination before and during the ball contact
determined. Our analysis is therefore not complete and in terms of reliability the results should be
seen as an approximation.
5. Conclusions
Nevertheless, we believe that from the kinematic description done in this study valuable insights
for training and competition arise, especially as the analysis—apart from the ball passing—was carried
out under typical game conditions. It has become clear, for example, how important it is to temporally
coordinate the single impulses during the hitting phase in order to maximize the terminal speed of
wrist and ball. Therefore, during practice, special attention should be taken that elbow, shoulder and
wrist are accelerated in quick sequence, one after the other. Since the first impulse for the spike is
produced in the torso, it seems reasonable to consider strengthening exercises for the core muscles as
a regular part of the training regime. Vigorous torso and shoulder muscles can also help to prevent
muscle injuries, which are quite common in fistball due to the partially strong inclinations and torsions
of the upper body [4] (as demonstrated in this study.) Electromyography methods would provide
further insight in this case. As the run-up is usually very short, the run-up velocity is believed to
play only a minor part in the jump height. The jump height is mainly the result of the explosive
strength of the legs and the coordinative quality of the jumping motion (e.g., fluent connection of up
and downward movement, use of initial force, and arm movement). In addition, the run-up in fistball
Sports 2016, 4, 55
11 of 11
is performed less in the up-front direction than, for example, in Handball. Of course, dynamometric
measurements using force plates would be interesting in this context; combined with our findings,
they would lead to a better understanding of the biomechanical structure of the fistball spike.
The progressive professionalization in fistball and the increasing popularity of this sport in many
countries should be reflected in the future by a stronger interest from sport science. This study as well
as the work of [4–7] are a starting point.
Author Contributions: Andreas Bund and Saeed Ghorbani conceived and designed the experiment;
Saeed Ghorbani and Franziska Rathjens performed the experiment; Andreas Bund analyzed the data;
Andreas Bund and Saeed Ghorbani wrote the paper.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
IFV. Faustball. In IFA—50 Years Young/50 Jahre Jung; Meuter-Druck: Düsseldorf, Germany, 2010.
Keil, B. Fistball (Faustball); Sport-Verlag: Berlin, Germany, 1983. (In German)
Holzheuer, M. Fistball—The unknown sport (Faustball—Die unbekannte Sportart). Sportpraxis 1986, 6, 13–14.
(In German)
Schmitz, A. Sports Injuries and Sports Damages in Fistball with a Special Focus on Shoulder Injuries of
Offense Players (Sportverletzungen und Sportschäden beim Faustball unter Besonderer Berücksichtigung
von Schulterverletzungen bei Angriffsspielern). Ph.D. Thesis, University of Münster, Münster, Germany,
2002. (In German)
Runer, A.; Runer, F.; Neunhäuserer, D.; Ring-Dimitriou, S.; Resch, H.; Moroder, P. A 1-year prospective
analysis of injuries in amateur and elite fistball. Scand. J. Med. Sci. Sports 2014, 24, 188–194. [CrossRef]
[PubMed]
Leser, R. On the use of a computer-based game observation system in competitive fistball (Zum Einsatz
eines computergestützten Spielbeobachtungssystems im Leistungs-Faustball. In Sporttechnologie Zwischen
Theorie und Praxis IV; Witte, K., Edelmann-Nusser, J., Sabo, A., Moritz, E.F., Eds.; Shaker: Aachen, Germany,
2006; pp. 209–214. (In German)
Soeser, K.; Schwameder, H. A three-dimensional kinematic analysis of the stand- and jump serve in
fistball. In Proceedings of the XXIII International Symposium on Biomechanics in Sports, Beijing, China,
22–27 August 2005; China Institute of Sport Science: Beijing, China, 2005; pp. 160–163.
Wank, V.; Menzel, H.J.; Wagner, H. Throwing and putting (Werfen und Stoßen). In Handbuch Sportbiomechnik;
Gollhofer, A., Müller, E., Eds.; Hofmann: Schorndorf, Germany, 2009; pp. 282–316. (In German)
Kuhlmann, C.H. Identification of Performance-Related Criteria for the Biomechanical Performance Diagnosis
by the Example of the Volleball Spike (Identifizierung Leistungsrelevanter Parameter für die Biomechanische
Leistungsdiagnostik am Beispiel des Angriffsschlages im Volleyball). Ph.D. Thesis, Technical University of
Chemnitz, Chemnitz, Germany, 2010. (In German)
Van den Tillaar, R.; Ettema, G. A three-dimensional analysis of overarm throwing in experienced handball
players. J. Appl. Biomech. 2007, 23, 12–19. [CrossRef] [PubMed]
Wagner, H.; Pfusterschmied, J.; von Duvillard, S.P.; Mueller, E. Performance and kinematics of various
throwing techniques in team-handball. J. Sports Sci. Med. 2011, 10, 73–80. [PubMed]
Wagner, H.; Pfusterschmied, J.; Klous, M.; von Duvillard, S.P.; Mueller, E. Movement variability and skill
level of various throwing techniques. Hum. Mov. Sci. 2012, 31, 78–90. [CrossRef] [PubMed]
Subijana, C.L.; Navarro, E. Kinetic energy transfers during the tennis serve. Biol. Sport 2010, 27, 279–287.
[CrossRef]
© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC-BY) license (http://creativecommons.org/licenses/by/4.0/).