Electromyographic Analysis of the Swim Start

Bilateral Comparison of the Front-weighted and Rear-weighted Track Start from

the OMEGA OSB11 Starting Block

Janna Brit Langholz

, Gunnar Westman

, Magnus Karlsteen

and Christian Finnsgård

1, 2

Centre for Sports Technology, Department of Applied Physics, Chalmers University of Technology, Gothenburg, Sweden

SSPA Sweden AB, Research, Gothenburg, Sweden

Keywords: Swimming, Swim Start, Electromyography, Muscle Activity, Competition Block.

Abstract: Previous swim start studies involving electromyography (EMG) consistently comprised unilateral

measurements and the attachment of the swimmer via cables to a computer. Therefore the present work

aims for an overall picture of the muscle activation pattern during the swim start by conducting bilateral

measurements with minimal restriction of motion. On that account a multichannel surface EMG device with

a wireless Bluetooth connection and videography is utilized in order to assess the nowadays most common

start dive techniques of competitive swimming events - differently weighted track starts from the OMEGA

OSB11 starting block. The data analysis identified that the normalized muscle activation levels were higher

during the front-weighted than during the rear-weighted start - probably caused by shorter block times and

less contribution of the arms. Furthermore the onset of the muscle activation seems to be different in

between start dive techniques, as for instance the muscles of the rear leg commence contracting earlier while

the muscles of the front leg start later in the rear-weighted compared to front-weighted starts. It is highly

likely that this originates in the position of the center of mass relative to the muscles. A general overview

over the coordination of the different muscles could also be obtained: It became obvious that some muscles

are the main drivers of the swim start (vastus lateralis, soleus) whereas others rather exerted supportive

actions (gluteus maximus, semitendinosus, erector spinae longissimus).

1 INTRODUCTION

Fractions of a second often separate the winner from

the rest in elite swim races. For instance, in 50 meter

butterfly races the finishing times of the first and the

third swimmer are only 0.09 s (male) and 0.05 s

(female) apart (Commonwealth Games Delhi 2010,

butterfly finals) (Slawson, 2012). Therefore each

swimmer should try to improve every movement of

the race. In short distance swimming events the start

off the block is of special importance and may

decide on success or defeat. General training

techniques mainly include time measuring or

perfunctory video analysis. Rule of thumb estimates

widely serve as feedback base for the body position

on the block. However, the fundamental unit which

drives the body forward is often neglected - the

muscle. Voluntary muscle contractions originate in

the motor cortex. Thereafter motor neurons transport

the necessary impulse to the muscle of interest and

cause action potentials which travel along the

muscle membrane. The amplitude and frequency of

these action potentials can be measured by

electrodes so that muscle activity levels and timing

can be analyzed - this process is called

electromyography (EMG) (Christensen, 1989).

Maximal off-the block performance can be achieved

if a beneficial muscle activation pattern is used as

force production on the block and the orientation of

the body segments relative to each other are

determined by the activation of the activation of the

involved muscles. Coaches and athletes should be

able to analyze and react to certain - probably

disadvantageous - activation patterns which can in

return lead to better race times. Overall, this work is

designed to gain an insight into the nowadays most

common swim start techniques utilizing an

electromyographic approach to investigate whether

the principle of EMG and the available EMG

technology can aid swimmers in improving their

start dive technique.

310

Langholz, J., Westman, G., Karlsteen, M. and Finnsgård, C..

Electromyographic Analysis of the Swim Start - Bilateral Comparison of the Front-weighted and Rear-weighted Track Start from the OMEGA OSB11 Starting Block.

In Proceedings of the 3rd International Congress on Sport Sciences Research and Technology Support (icSPORTS 2015), pages 310-320

ISBN: 978-989-758-159-5

1.1 Previous Research on the

Competitive Swim Start

The latest changes in the regulations of the

Fédération Internationale de Natation (FINA Rules,

FR 2.7) brought the company Swiss Timing Inc. to

launch their new starting block OMEGA OSB11 in

April 2008 which is now predominantly used in

international competitions (Murrell, 2012). On this

block an additional foot rest (30% incline) is

mounted to the back of the longer and steeper

surface area allowing for an amended track start

called kick start with better performance. Higher

take-off velocities, shorter block times and faster

times at different distances could be recorded

(Honda et al., 2010; Biel et al., 2010; Ozeki et al.,

2012). These facts brought the grab start to mostly

vanish from elite swimming contests (Vint et al.,

2009). Beyond that different studies discovered that

moving the center of mass backwards during the

track start is beneficial compared to the classical

front-weighted track start (Welcher et al., 2008;

Vilas-Boas et al., 2003). A recent study took these

findings and observed the performance of a front-,

neutral- and rear-weighted kick start from the

OMEGA OSB11 over a distance of 15 meters. The

results indicated faster times at 15 meters for the

neutral- and rear-weighted variant of the kick start

from this new starting block (Barlow et al., 2014).

Currently ongoing research takes further parameters

into account including stance width, height of the

center of mass and foot preferences of the swimmer

(Kibele et al., 2015).

In 1964 surface EMG had its debut in the water

environment and introduced a new variable to

swimming movement analysis: muscle activity

(Barthels, 2015). During the last decades

Electromyography found its way into competitive

swimming research. Fatigue analysis of different

swim styles (Conceição et al., 2010; Yasushi Ikuta et

al., 2012) and dry-land exercises (Ganter et al.,

2007; Nazário-de-Rezende et al., 2012) form the

majority of EMG papers in swimming. As most of

the commercially available EMG systems are not

waterproof and rather sensitive to rough under water

movements, measurements have been restricted to

plain swimming motions without regular starts and

turns. Making matters worse, a large number of

EMG systems has to be connected to a computer via

a multitude of wires which may impair the natural

motion of the swimmer.

The swim start move in particular was only

subject to three scientific studies. A first paper

observed electromyographic activity of two different

variants of the backstroke start – feet immerged or

feet emerged. Seven electrodes were placed

unilaterally on arm, trunk and leg muscles (de Jesus

et al., 2011). Another study also focused on the

backstroke start in order to see how different the

muscle activation pattern is in between subjects.

Here eight electrodes were placed on the right side

of the body (Hohmann et al., 2008). A third paper

dealt with the biomechanics of the grab and track

start technique and again eight electrodes were

positioned on upper and lower extremities as well as

the trunk on the right side of the body (Krueger et

al., 2003). All these papers share the fact that the

measurements were conducted unilaterally which

does not allow an overall examination of the

movement. Additionally all subjects were fixed via

cables to the main acquisition unit. This may have

affected the motion during the examined start

technique.

1.2 Purpose of the Paper

Even though EMG measurements are sensitive to

external influences - which especially applies to its

use in the water environment - it has been decided to

advance research regarding the muscular activity

during the block phase of the swim start. It is

assumed that this knowledge may help in optimizing

start techniques as well as suggesting new training

methods and exercises (Clarys et al., 1993). Besides

this the relationship between EMG data and

kinematic variables is said to play a key role in the

evaluation of swimming performance (Conceição et

al., 2013) and should therefore be investigated

further.

2 METHODS

Bilateral measurements with minimal restriction of

motion are to be achieved by carefully selecting the

most important muscles of the swim start and

employing a multichannel surface EMG device

using Bluetooth technology to transfer the measured

data to a computer. In a second step it is analyzed

whether the EMG data can be linked to other

variables of the swim start including angles from the

block and instantaneous horizontal velocities

(kinematic data). It is moreover assessed whether the

results can be connected to the outcome of other

related studies.

The focus is placed on the front- and rear-weighted

kick start (Figure 1) as those techniques are the most

widely applied ones off the OMEGA OSB11 starting

Electromyographic Analysis of the Swim Start - Bilateral Comparison of the Front-weighted and Rear-weighted Track Start from the

OMEGA OSB11 Starting Block

311

block and are subject to the latest swim start studies

(Barlow et al., 2014; Kibele et al., 2015). Due to a

limitation of time, equipment and staff the

observations concentrates on the block phase instead

of including flight and dive phase parameters.

Moreover technology constrains an analysis of the

total start sequence: Bluetooth transmission does not

work under water. This choice is endorsed by the

literature: All subsequent components of the swim

start are influenced by the block phase what gives

each swimmer the task to strive for maximal off-the-

block performance (Mason et al., 2007).

Figure 1: Rear-weighted (left) and front-weighted (right)

initial start position on the block.

2.1 Participants

Three males and three females (five Swedes, one

American) took part in this study. Table 1 shows the

anonymized personal data of each swimmer. All

swimmers were competing on developmental or

national level. Four swimmers preferably had the

right foot on the front edge of the starting block, two

of them the left foot.

Table 1: Personal data of all participating subjects (Lat. =

Laterality, Pref. Pos. = preferred weight distribution on the

block).

No. Gender Age cm/kg Lat. Pref. Pos.

S1 Male 19 179/75 right Neutral

S2 Male 20 192/74 right Front

S3 Female 20 173/65 right Front

S4 Female 19 173/75 right Front

S5 Male 22 177/79 right Front

S6 female 22 171/61 right Neutral

2.2 Equipment

The EMG device of choice is the telemetric unit

MQ16 by Marq Medical which weighs 120 g and

has a data buffer memory of 60 MB that collects the

data and a Bluetooth transmitter sends it to a

computer. Preamplifiers amplify the signal close to

the electrodes and conduct analog to digital

conversion. The noise level of the MQ16 is specified

to be less than 3 µV (Meyland et al., 2014). The

default sampling frequency is 1024 Hz which is

sufficient for EMG signals ranging between 20 and

500 Hz. However, it was decided to set the sampling

frequency of the MQ16 to the widely used value of

2048 Hz (Ali et al., 2014; Barlett, 2007). Surface

electrodes are applied in a bipolar configuration with

a common ground and made of Ag/AgCl. As the

device is to be used under wet conditions it has to be

protected by a waterproof casing (iPad®mini™case

by ECase). Transparent film dressing (Tegaderm™,

3M™) is chosen as covering for the electrodes and

will be additionally fixed with foam tape

(Microfoam, 3M™ ). Furthermore an accelerometer

(mounted to the ankle of the rear foot) is used to

track the swimmer’s first motion on the block after

the starting signal and a pressure mat is used to track

the last motion on the block (foot-off moment).

These two event markers are then used for time

normalization of the movement for successive

averaging of trials. The Casio Exilim EX-FH25

video camera (shutter speed 1/500 s, 120 fps) is

mounted to a rigid tripod to allow the recording of

the swimmer in the sagittal plane.

It was positioned

perpendicular to the plane of motion approx. 5 m

from the center of the swimming lane.



2.3 Measurement Protocol

Personal information of all subjects was captured

and they signed a letter of agreement. Thereafter the

electrode placement procedure commenced

considering the SENIAM guidelines (Surface

Electromyography for the Non-Invasive Assessment

of Muscles, EU project) regarding the electrode

location and positioning process. Nine muscles of

the back and the lower limbs were chosen to

represent the most relevant muscles during the start

dive in swimming: erector spinae longissimus,

gluteus maximus, vastus lateralis, semitendinosus

and soleus. The skin area which was supposed to be

covered by the electrodes was shaved, scrubbed with

sand paper and then cleaned with an alcohol wipe.

This was followed by attaching the MQ16 to the

waist of the swimmer and connecting the cables to

the electrodes. Figure 2 shows that the electrodes

including the pre-amplifiers were covered by

waterproof tape with a size of at least 20 x 10 cm.

All edges were additionally fixed by waterproof

foam tape to prevent water induced detachment of

the sealing. The accelerometer was mounted to the

inside of the rear ankle and covered in the same

WPPDSports 2015 - Special Session/Symposium on Weather, Position and Performance data in Outdoor Sports

312

manner as the electrodes. 

Figure 2: Covering of the electrodes and placement of the

accelerometer right above the rear ankle.

The movement of the swimmer was supposed to

be as unrestricted as possible. Consequently it was

decided to supply the swimmers with commercial

tights to fix the cables to the lower limbs. This step

had a beneficial secondary effect: minimization of

cable motion. Figure 3 illustrates that pink markers

with a diameter of approximately 4 cm were placed

on the four landmarks defining the body segments:

ankle, knee, hip and shoulder (Krueger et al., 2003).

Maximum voluntary contractions (MVC)

performed on a gym bench are used to normalize the

amplitude of the EMG signal. Normalization makes

it possible to conduct an analysis which is not based

on absolute values. Thereby comparisons between

trials of the same subject or between different

individuals can be made. Three maximal

contractions against an insuperable resistance are

performed for each muscle. A duration of six

seconds and a short period of rest in between is

chosen (Masso et al., 2010; Halaki et al., 2012).

From different papers the most feasible MVC tasks

for the muscles of interest in the present study were

selected (Halaki et al., 2012; Konrad, 2005). For the

reason that varying electrode placement might

change the measured EMG signal, the MVC

recordings for normalization were conducted with

the same configuration as the subsequent EMG start

dive measurements. As the wet pool environment

might impair the electrode fixation, the MVC

measurements were performed beforehand in a dry

environment. Prior to executing the MVC tests, the

swimmers gave feedback on their positioning and

the static resistance they had to work against. Only if

the positioning was suitable for the measurement

and comfortable for the swimmer, the recording was

started.

After all preparatory procedures were

undertaken, the swimmer was ready to conduct three

front-weighted and three rear-weighted start dives.

Figure 3 sows subject S4 in a rear-weighted starting

position. With respect to various existing start jump

techniques the swimmers included a few additional

start jumps of the requested techniques into their

regular training during the previous weeks. In the

front-weighted position they were asked to distribute

the majority of their weight on the front foot whilst

still preserving a stable stance on the block and in

the rear-weighted position they were instructed to

allocate the majority of their weight to the rear foot

and at the same time maintain a stable position on

the block.

Figure 3: Swimmer with fixated electrodes and markers

and the MQ16 on the back. OSB11 cover with pressure

mat placed under the front foot.

The subjects were directed to propel themselves

off the block with maximal effort like they would do

in a competition. In order to prevent the

waterproofing bag and the electrode sealing from

damage, the swimmers were directed to not perform

any powerful swim strokes but return to the edge of

the pool by exerting cautious arm strokes. They were

randomly asked to apply one or the other technique

and after each trial they had approximately five

minutes of rest. All trials were performed in the

same way: When the Bluetooth connection between

the MQ16 and the computer was established and the

swimmer was in readiness, the EMG recording was

started. Right after that the MQ16 transmit trigger

(connected to pressure mat) was enabled. The

Electromyographic Analysis of the Swim Start - Bilateral Comparison of the Front-weighted and Rear-weighted Track Start from the

OMEGA OSB11 Starting Block

313

camera operator then started the video recording,

gave the prestart command "On your mark!" and

then activated the start signal.

2.4 Data Analysis

As the swimmer’s movement off the starting block

only comprises a duration of around a second, it is

assumed that the fatigue effect (frequency analysis)

of the involved muscles can be neglected. Therefore

the subsequent data analysis focused on the analysis

of EMG amplitudes and the timing of the muscle

activation of the different muscles relative to each

other. The MATLAB R2013b software was utilized

for this purpose.

Figure 4: Processed signal (vastus lateralis rear, S3) with

mean and peak RMS values and markers depicting the

start and end of the block phase.

The processing steps included rectification of the

signal, correction of a possible DC offset and

determination of the linear envelope or root-mean-

square (RMS) to specify EMG amplitude

(Lamontagne, 2000). Figure 4 shows an example of

an EMG amplitude analysis. Additionally on- and

offset set times of muscular activity were calculated

and the time stamps from the accelerometer and the

pressure mat on the starting block were found.

The video data was predominantly processed

with the software Tracker (Tracker software, version

4.8x) which helps in calculating times, angles and

velocities after tracking the markers (Figure 5).

Filtering of the digitized landmarks was again done

in MATLAB R2013b. Differences in between the two

start techniques were tested by means of paired t-

tests using SYSTAT 13.

Figure 5: Video scaling & digitizing of the hip marker

(approx. centre of mass) with the software Tracker.

3 RESULTS

Many start dive recordings were accompanied by

intrusion of water under the electrode sealing, which

resulted in erroneous data sets. An averaging of

different trials of the same position and subject

would then lead to unnaturally high or low voltage

values, which do not reflect the actual activation.

Therefore it was decided to neglect the time

normalization and successive averaging of the three

front-weighted and three rear-weighted recordings of

each subject but select the best front-weighted and

the best rear-weighted data set of each subject. For

that reason only five front-weighted and six rear-

weighted data sets enter the final data analysis.

As the subjects have been introduced to the

concept of recording maximum voluntary

contractions prior to the test day, meaningful MVC

recordings were obtained. Additional time during the

MVC tests to adjust the position on the gym bench

or the fixation of the limbs, also contributed to the

positive outcome of the MVC recordings.

Contraction periods could clearly be separated from

rest periods, which facilitated proper RMS

calculations and identification of the maximal

voltage value for each muscle.

However, as also depicted in other papers, MVC

recordings may not be appropriate to normalize

dynamic movements as they represent the ability of

the muscle to contract in a static setting. In dynamic

movements the synchronization of motor units is

different. This constraint must be kept in mind.

WPPDSports 2015 - Special Session/Symposium on Weather, Position and Performance data in Outdoor Sports

314

3.1 Activation Levels

The main actors (vastus lateralis front/rear, soleus

front/rear) predominantly register the highest

activation levels, ranging between 48.6% and

191.0% for mean RMS. Additionally the standard

error is rather high for these strong, rapidly acting

muscles. The gluteus maximus rear and

semitendinosus rear settle just below with an

average activation of 38.1% and 34.1% respectively.

The back muscles (erector spinae longissimus) as

well as the gluteus maximus front and

semitendinosus front show lower mean RMS values

18.9% and 29.4% with smaller standard errors than

the main actors.

3.2 Activation Timing

After the reaction time has passed the muscles begin

to take up their work at different times. It can clearly

be seen that the erector spinae longissimus is on

average active before any motion can be detected

(accelerometer/video). Almost synchronously the

gluteus maximus and semitendinosus of the rear leg

take up their work and also the semitendinosus front

begins contracting almost simultaneously with the

two aforementioned muscles. Shortly after the vastus

lateralis rear and soleus rear show their activation

onset. Figure 6 illustrates that, after almost half of

the block time has passed, the gluteus maximus front

comes into the game marking the start of the pushing

phase of the front leg. Lastly the S front and the

vastus lateralis front perform the last powerful action

off the block with an onset at 54.8% and 57.1% of

total block time respectively.

Generally the rear leg comes into action before

the front leg. When observing the two different

initial positions it can be seen that the muscles of the

rear leg during the rear-weighted start generally start

contraction earlier than in front-weighted starts. On

the other hand, the muscles of the front leg seem to

have a later onset during the rear-weighted than

during the front-weighted start.

3.3 Kinematic Parameters

Reaction times (start signal until start of motion) of

0.245 ± 0.013 s have been identified by means of

Tracker from the video recordings. It can be seen

that the reaction times for the rear-weighted start are

often shorter (0.236 s) compared to the front-

weighted start (0.254 s). Beyond that it was

observed that the swimmers remained on average

0.074 s longer on the starting block in rear-weighted

track starts.

Figure 6: EMG signals [V] of all nine investigated

muscles during a front-weighted start dive (S1). The pink

line depicts the start of motion (accelerometer signal) and

the blue line represents the foot-off moment (pressure mat

signal).

Electromyographic Analysis of the Swim Start - Bilateral Comparison of the Front-weighted and Rear-weighted Track Start from the

OMEGA OSB11 Starting Block

315

This leads to a longer average start-to-flight time

(start signal until foot-off) for the rear-weighted start

(0.841 ± 0.034 s) off the OMEGA OSB11 starting

block - even though the detected reaction times were

shorter than for the front-weighted starts. For a

front-weighted start the swimmers required on

average 0.783 ± 0.045 s. Generally the men achieved

faster starting times than the women.

The calculated angles from the block for the six

subjects depict a certain pattern with regard to the

initial position on the block: rear-weighted starts

seem to lead to smaller angles. Yet it cannot be

proven that the angles during the front-weighted

start are significantly higher than during the rear-

weighted start dive (p-value = 0.097). Most angles

measured per subject per position are close together,

but the angle from the block ranges largely in

between subjects depicting values ranging from 25.1

to 48.6 degrees.

The instantaneous, horizontal velocities also

indicate a particular pattern: the velocities off the

block are significantly higher for rear-weighted

starts (p-value = 0.013). The angle also seems to

have a certain influence on the velocity as it is

frequently detected that the smaller the angle, the

higher the horizontal velocity at foot-off becomes.

An average instantaneous, horizontal velocity of

2.166 ± 0.112 s is found for front-weighted and

2.369 ± 0.110 s for rear-weighted starts. The

velocity recorded might however not represent the

maximal velocity throughout the jump. For different

subjects higher horizontal velocities were found

shortly before or after foot-off, depending on the

individual stretching pattern during this phase of the

start.

4 DISCUSSION

The EMG data acquisition during the start dive

formed the core of the present investigation. Each

subject conducted three front-weighted and three-

rear weighted kick starts off the OMEGA OSB11

starting block. As favored the subjects did not report

any impairment of motion due to the cables on their

limbs or the MQ16 on their back. When the

electrodes remained dry under the waterproof tape,

they were able to measure the muscular activation

during the start dive. This leads to the assumption

that the selected electrode positions as well as the

skin preparation chosen, were appropriate for the

current study. The noise of the components has been

found to be negligible and a certain variability in the

data due to variations in between subjects has to be

accepted. However, certain events negatively

influenced the EMG data acquisition in the water

environment – detachment of the electrodes from the

skin and disturbance of clamps and amplifiers by the

water.

It was found that the mean EMG amplitude

during the rear-weighted start is significantly smaller

than the amplitude during the front-weighted start

technique. This is mainly caused by longer block

times in rear-weighted starts as the center of mass

(CoM) needs to travel a longer distance until foot-

off than in the front-weighted version. Additionally

the arms are capable of contributing a large part to

the forward motion by pulling the swimmer out of

the initial rearward position. The standard error was

found to be much higher for the forceful main actors

of the start dive (vastus lateralis, soleus) than for the

other muscles. This may originate from the

dynamics of the start dive leading to high voltage

peaks in the EMG curve of these strong muscles.

A comparison of the onset times for front-

weighted and rear-weighted start dives

predominantly revealed an earlier muscle onset

within the rear leg in rear-weighted starts. Certain

muscles of the front leg, however, start contracting

later than in front-weighted starts. This discrepancy

leads to a non-significant difference between the

muscular onset of the front-weighted and rear-

weighted track start technique when conducting a

statistical analysis. A reason for this might be that

the initial position places a high weight on the rear

foot allowing an immediate force production in the

beginning whereas the CoM has to be moved a long

distance before the muscles of the front leg can start

executing their function.

The parameters obtained from the video

recordings support the assumption that the selected

subjects are suitable representatives for aspiring

swimming professionals. When opposed to the

outcome of other swim start studies, the subjects of

this study show similar results - however, not with

the same magnitude as elite swimmers (Barlow et

al., 2014).

Visually connecting the EMG measurements to

the video recordings ultimately revealed the motion

sequences during which the different muscles are

predominantly active (Figure 7). The interplay of

different muscle groups was identified, for instance

the almost simultaneous activation of the gluteus

maximus and the vastus lateralis muscles and the

dissimilar activation of semitendinosus and soleus in

the same leg. The collaboration of the gluteus

maximus and the vastus lateralis was also identified

previously (de Jesus et al., 2011).

WPPDSports 2015 - Special Session/Symposium on Weather, Position and Performance data in Outdoor Sports

316

Figure 7: Images of a rear-weighted start dive (S4)

indicating the different onset times of the recorded

muscles.

The mentioned dissimilar activation of

Hamstring and Calf muscles can be explained by the

differing required performance conditions. During

the start dive the Hamstrings influence the

swimmer’s hip and knee extension at smaller knee

angles whereas the Calves seem to take up their

work at larger knee angles conducting the last

powerful movement on the block.

Besides, the semitendinosus of the front leg

counterbalances the forces which are generated

when the soleus of the rear leg pushes off the foot

rest on the back of the block in order to preserve a

secure stance on the block. After a leg has left the

block the semitendinosus of the same leg might

show another peak related to the injury prevention

mechanism of the knee joint; At that point in time

the soleus of that leg already shows strongly

decreasing activation levels.

Moreover the back muscles as well as the gluteus

maximus help in extending the swimmer into the

favored streamline position as they show activation

at the beginning of motion and right before and after

foot-off. This role is confirmed by different swim

start publications which assigned the same

responsibilities to the abovementioned muscles (de

Jesus et al, 2011, Hohmann et al., 2008). As the

upward rotation of the trunk commences first with

the help of the back muscles and leg extensors

causing vertical acceleration of the trunk and

simultaneously a force pointing downwards in the

direction of the lower limbs, the extension of the

legs only starts when the hips begin to be

unweighted. A possible drop of activity in the hip

extensors (gluteal muscles and Hamstrings) after the

first part of the push phase might be related to the

termination of the upward rotation of the trunk.

Furthermore the fact that the muscular activation of

the ST of the front leg peaks later than the one of the

rear leg seems to be linked to the positioning of the

muscle relative to the CoM meaning that the muscle

in the front leg cannot contribute its part to the hip

extension and stance stabilization until the CoM

reached a certain location further to the front of the

starting block.

After foot-off the muscle activation profiles vary

remarkably in between subjects as each swimmer

conducts different corrective and stabilizing

movements during the flight phase (Hohmann et al.,

2008). Some swimmers still show elevated EMG

profiles in their back muscles whereas others

continue to contract their soleus muscles for a

beneficial feet position. It was neither aimed nor

possible to make statements about later swim start

Electromyographic Analysis of the Swim Start - Bilateral Comparison of the Front-weighted and Rear-weighted Track Start from the

OMEGA OSB11 Starting Block

317

phases as it is a highly individual issue and the

Bluetooth connection vanishes upon water entry.

Taking the EMG recordings and the video data

into account one might see that the rear leg is the

initiator and main driver of the observed track start

technique. It sets the body in motion and moves the

CoM to the front of the starting block. Changing the

initial position changes the activation levels and

timing in both legs but the rear leg generally

performs the bigger part of the work which can be

seen in predominantly higher activation levels.

However, without the stabilizing actions of the front

leg, which form the basis for a solid stand on the

block, and the last push off the block, which defines

the positioning of the swimmer in the air, the start

jump would not be as effective and efficient as it is.

5 CONCLUSION

Muscles vary largely in volume and structure

leading to a different base of operation for each

muscle and restricting the comparability of

activation levels and timing in different muscles.

Moreover some of the leg muscles observed belong

to the group of biarticular muscles (e.g. gluteus

maximus, Hamstrings) meaning that they traverse

two joints and thereby have another principle of

operation than monoarticular muscles (Bobbert et

al., 1988). Additionally it has to be noted, that in

complex movements like the swim start which

include all body segments, muscles influence the

joints they cross as well as other joints (Bobbert et

al., 1988). Another issue is the link between EMG

activation and produced tension: After a muscle has

been stimulated, a delay of 20 to 100 ms occurs

between a recorded muscular activity onset and the

resulting mechanical output which can be observed

in video recordings (Bartlett, 2007; Bobbert et al.,

1988).

The complexity of the human body as well as the

ambiguous relationship between EMG and

kinematic data unfortunately limit the information

one can extract from EMG measurements and a

combination of those with video data. Solely activity

levels and timing can be detected, but not the

muscle’s exact task within a movement (Bartlett,

2007). In order to avoid speculative interpretation,

the present work did not assign a particular muscle

to a certain function, body position or joint angle but

gave a suggestion regarding the contribution of a

muscle to a certain motion sequence during the

swim start.

The physiological and mechanical considerations

made within the scope of this work are not the only

ones influencing the muscular activation during the

start jump in swimming. In future investigations

further muscles or muscle properties should be

addressed and the location of the CoM or different

body segments varied (Bobbert et al., 2007). Besides

altering the scope of the investigation, improvements

should be made regarding the equipment, for

instance using wireless electrodes in order to

facilitate better sealing, finding a solution for the

Bluetooth loss between air and water and

strengthening the synchronization of all applied

measurement devices. Additionally trying to supply

the swimmers and coaches with quickly processed,

meaningful data would be a desirable advancement

(Hamill, 2010).

It is assumed that a step in the right direction was

made by pointing out the importance of the

understanding of the muscular tasks exerted by the

swimmers body. This can aid in developing training

methods and prospective start dive techniques.

Unfortunately the process of recording EMG with

the current technology is highly sensitive and

protracted so that an application in everyday swim

training sessions is - for the time being - not feasible.

ACKNOWLEDGEMENTS

The authors wish to acknowledge the support from

all swimmers of MASS and Göteborg Sim who

participated in the measurements.

REFERENCES

Ali, A., Sundaraj, K., Ahmad, R.B., Ahamed, N.U., Islam,

A., 2014. Recent observations in surface

electromyography recording of triceps brachii muscle

in patients and athletes, Applied Bionics and

Biomechanics, 11, (p. 105–118).

Barlow, H. et al., 2014. The effect of different kick start

positions on OMEGA OSB11 blocks on free swimming

time to 15 m in developmental level swimmers,

HumanMovement Science, 34 (2014), (p. 178-186).

Barthels, Katharine M., 2015. Biomechanical Research in

Swimming: Past, Present and Future, Physical

Education Department, California Polytechnic State

University, https://ojs.ub.uni-konstanz.de/cpa/article/

download/906/.

Bartlett, R., 2007. Introduction to Sports Biomechanics –

Analysing Human Movement Patterns, Taylor &

Francis e-Library, Second edition.

WPPDSports 2015 - Special Session/Symposium on Weather, Position and Performance data in Outdoor Sports

318

Biel, K., Fischer, S., Kibele, A., 2010. Zur Effektivität des

neuen Startblocks (OSB 11) beim Schrittstart im

Schwimmen, Sportverlag Strauß (Köln):

Biomechanische Leistungsdiagnostik im Schwimmen

– Erfahrungen im Leistungssport und Ableitungen für

die Ausbildung von Studierenden. Beiträge zum dvs-

Symposium Schwimmen, (p. 113–117).

Bobbert, M. F., Jaspers, R. T., Sijpkens, I. W. T., 2007.

Muscle activation patterns in squat jumps from

different initial positions, Journal of Biomechanics, 40

(S2).

Bobbert, M. F., Van Ingen Schenau, G. J., 1988.

Coordination in vertical jumping, Journal of

Biomechanics, 21(3), (p. 249-262).

Christensen, H., 1989. EMG Analysis, IEEE Engineering

in Medicine & Biology Society, 11th annual

international conference, (p. 1468-1469).

Clarys, J., Cabri, J., 1993. Electromyography and the

study of sports movements: a review, Journal of Sports

Sciences, 11, (p. 379-448).

Conceição, A., Silva, A., Barbosa, T.M., Louro, H., 2013.

Observation and Technical Characterization in

Swimming: 200mBreaststroke, Rev. Bras.Med.

Esporte, Vol. 19, No. 1.

Conceição, A., Silva, A., Palma, S., Silva, H., Gamboa,

H., Louro, H., 2010. Electromyography in Front

Crawl Technique - Case Study, The Open Sports

Sciences Journal, 3 (2010), (p. 67-69).

de Jesus, K., de Jesus, K., Figueiredo, P., Goncalves, P.,

Pereira, S., Vilas-Boas, J. P, Fernandes, R., 2011.

Electromyographic analysis of two different feet

positions in back stroke start, Portuguese Journal of

Sport Sciences, 11 (Suppl. 2), 2011, (p. 191-194).

Ganter, N., Witte, K., Edelmann-Nusser, J., Heller, M.,

Schwab, K., Witte, H., 2007. Spectral parameters of

surface electromyography and performance in swim

bench exercises during the training of elite and junior

swimmers European Journal of Sport Science, 7, 3, (p.

143-155).

Halaki, M., Ginn, K., 2012. Normalization of EMG

Signals: To Normalize or Not to Normalize and What

to Normalize to? Computational Intelligence in

Electromyography Analysis – A Perspective on

Current Applications and Future Challenges, (p. 175-

194).

Hamill, J., 2010. Real-TimeMeasurement using EMG

andMotion Capture Systems, XXVIII International

Syposiumof Biomechanics in Sports.

Hohmann, A., Fehr, U., Kirsten, R., Krueger, T., 2008.

Biomechanical analysis of the backstroke start

technique in swimming, E-Journal Bewegung und

Training, 2 (2008), (p. 28-33).

Honda, K. E., Sinclair, P. J., Mason, B. R., Pease, D. L.,

2010. A Biomechanical Comparison of Elite

Swimmers Start Performance Using the Traditional

Track Start and the New Kick Start, Biomechanics and

Medicine in Swimming XI. Proceedings of the XI th

International Symposium for Biomechanics and

Medicine in Swimming, (p. 94–96).

Kibele, A., Biel, K., Fischer, S., 2015. SWIM START

STANDPOINTS ON THE OSB11 STARTING

BLOCK, Institute for Sports and Sport Science,

University of Kassel, Germany.

Konrad P., 2005. The ABC of EMG - A Practical

Introduction to Kinesiological Electromyography,

Version 1.0, April 2005.

Krueger, T., Wick, D., Hohmann, A., El-Baharawi, M.,

Koth, A., 2003. Biomechanics of the Grab and Track

Start Technique, Biomechancis and Medicine in

Swimmng IX, (p. 219-223).

Lamontagne, M., 2000. Application of Electromyography

in Movement Studies, 18 International Symposium on

Biomechanics in Sports.

Mason, B., Alcock, A., Fowlie, J., 2007. A kinetic analysis

and recommendations for elite swimmers performing

the sprint start, Proceedings of the 25th conference of

the international society of biomechanics in sports

(Brazi), (p. 192–195).

Massó, N., Rey, F., Romero, D., Gual, G., Costa, L.,

Germán, A., 2010. Surface electromyography

applications in the sport, ApuntsMed Esport, 45(165),

(p. 121-130).

Merletti, R., Parker, P., 2004. ELECTROMYOGRAPHY

Physiology, Engineering, and Noninvasive

Applications, JohnWiley & Sons, Inc., Hoboken, New

Jersey, ISBN 0-471-67580-6.

Meyland, J., Heilskov-Hansen, T,, Alkjær, T., Koblauch

Sigurd Mikkelsen, H., Wulff Svendsen, S., Frølund

Thomsen, J., Hansson, G.-Å., B. Simonsen, E., 2014.

Sex differences in muscular load among house

painters performing identical work tasks, European

Journal of Applied Physiology, 114, (p. 1901–1911).

Murrell, D., Dragunas, A., 2012. A Comparison of Two

Swimming Start Techniques from the Omega OSB11

Starting Block, Western Undergraduate Research

Journal: Health and Natural Sciences, Volume 3.

Nazário-de-Rezende, F., da Cunha Sousa, G., Haddad, E.

G., de Oliveira, V. S., da Silva Medeiros, R., de

Agostini, G. G., Marocolo, M., 2012.

Electromyographic study of the Deltoid,

PectoralisMajor and Triceps Brachii Muscles in

swimmers during bilateral contractions performed in

multi-joint exercise with different loads, Rev.

Bras.Med. Esporte., Vol. 18, No. 2.

Ozeki, K., Sakurai, S., Taguchi, M., Takise S., 2012.

Kicking the back plate of the starting block improves

start phase performance in competitive swimming,

Proceedings of the 30th Conference of the

International Society of Biomechanics in Sports, 4, (p.

373–376), Melbourne (Australia),

https://ojs.ub.unikonstanz.de/cpa/article/view/5336.

Slawson, S. E.; Chakravorti, N.; Conway, P. P.; Cossor, J.;

West, A. A., 2012. The Effect of Knee Angle on Force

Production, in Swimming Starts, using the OSB11

Block, Procedia Engineering, 34, (p. 801–806).

Vilas-Boas, J. P. et al., 2003. Biomechanical analysis of

ventral swimming starts: comparison of the grab start

with two track-start techniques, Biomechanics and

Medicine in Swimming IX. Proceedings of the IXth

Electromyographic Analysis of the Swim Start - Bilateral Comparison of the Front-weighted and Rear-weighted Track Start from the

OMEGA OSB11 Starting Block

319

International Symposiumon Biomechanics and

Medicine in Swimming (p. 249–253).

Vint, P. F., Hinrichs, R. N., Riewald, S. A.,Mason, R. K.,

McLean, S. P., 2009. Effects of handle and block

configuration on swim start performance, XXVII

international society of biomechanics in sports

conference, Limerick (Ireland), http://w4.ub.uni-

konstanz.de/cpa/article/view/3172/2976.

Welcher, R. L., Hinrichs, R. N. and George, T. R., 2008.

Front- or rear-weighted track start or grab start:

which is the best for female swimmers?, Sports

Biomechanics,7 (1), (p. 100–113).

Yasushi Ikuta, Yuji Matsuda, Yosuke Yamada, Noriyuki

Kida, Shingo Oda, Toshio Moritani, 2012.

Relationship between decreased swimming velocity

and muscle activity during 200-m front crawl,

European Journal of Applied Physiology, 112 (2012),

(p. 3417–3429).

WPPDSports 2015 - Special Session/Symposium on Weather, Position and Performance data in Outdoor Sports

320