Advanced performance analysis of the Illinois agility test based on the tangential velocity and...

This article was downloaded by: [RMIT University], [Miss Clara Usma]On: 03 July 2011, At: 21:50Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

Sports TechnologyPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/rtec20

Advanced performance analysis of the Illinois agilitytest based on the tangential velocity and turning radiusin wheelchair rugby athletesClara Cristina Usma-Alvarez a , Julian Jang Ching Chua a , Franz Konstantin Fuss a ,Aleksandar Subic a & Michael Burton aa School of Aerospace, Mechanical and Manufacturing Engineering, RMIT University,Melbourne, Australia

Available online: 30 Jun 2011

To cite this article: Clara Cristina Usma-Alvarez, Julian Jang Ching Chua, Franz Konstantin Fuss, Aleksandar Subic & MichaelBurton (2010): Advanced performance analysis of the Illinois agility test based on the tangential velocity and turning radius inwheelchair rugby athletes, Sports Technology, 3:3, 204-214

To link to this article: http://dx.doi.org/10.1080/19346182.2011.564284

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions

This article may be used for research, teaching and private study purposes. Any substantial or systematicreproduction, re-distribution, re-selling, loan, sub-licensing, systematic supply or distribution in any form toanyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contentswill be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses shouldbe independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims,proceedings, demand or costs or damages whatsoever or howsoever caused arising directly or indirectly inconnection with or arising out of the use of this material.

http://www.tandfonline.com/loi/rtec20

http://dx.doi.org/10.1080/19346182.2011.564284

http://www.tandfonline.com/page/terms-and-conditions

RESEARCH ARTICLE

Advanced performance analysis of the Illinois agility test based on thetangential velocity and turning radius in wheelchair rugby athletes

CLARA CRISTINA USMA-ALVAREZ, JULIAN JANG CHING CHUA,

FRANZ KONSTANTIN FUSS, ALEKSANDAR SUBIC, & MICHAEL BURTON

School of Aerospace, Mechanical and Manufacturing Engineering, RMIT University, Melbourne, Australia

(Received 30 November 2010; accepted 20 January 2011)

AbstractThe Illinois agility test is a timed task suitable for wheelchair sports with multidirectional movements and it is commonly usedto evaluate the agility performance of athletes. In addition to test times, tangential velocities and turning radii are parametersof high importance in the assessment of agility performance. The main objective of this study is to develop a method thatallows the evaluation of the tangential velocity and turning radius that are achieved during the Illinois agility test, which isusually used for training wheelchair rugby athletes. A group of nine Paralympic wheelchair rugby athletes performed threetrials of the Illinois agility test using their own custom made chairs. The wheelchairs were instrumented with a 6 DOF(degrees of freedom) sensor (accelerometer/gyrometer). Linear accelerations, angular velocities and test times were recordedduring the test. The investigation of obtained performance enabled identification of instantaneous radii and tangentialvelocities of the turning chair for each of the 11 turns of the Illinois test. To evaluate an athlete’s performance, hypothesesbased on optimisation strategy of performance for the test’s turning circuit were considered as follows: (1) athletes produce alonger rc in turns 3, 4, 5, 7, 8 and 9; (2) athletes produce a shorter rc (turning radius) in turns 2, 6 and 10; and (3) athletesproduce a longer rc in turns 1 and 11.

According to the performance parameters mentioned above, four performance quadrants could be identified for differentturn radii and velocities. The performance assessment of athletes presented here showed no significant difference in turningradius between expected longer and shorter radius turns in individual athletes. It was concluded that using turning radius as aperformance parameter is not appropriate. Velocity in the expected shorter radii proved to be generally larger than the one inthe expected longer radii with significant difference in four of the nine athletes. Significant difference was found between radiiof group A (low pointer athletes) and group B (mid pointer athletes), and larger radii were observed in group A (low pointerathletes). The differences in velocities between groups A and C, and between B and C were significant (P # 0.001 each). Themethod developed and proposed proved to be valid for assisting performance assessment of wheelchair sports athletes usingthe Illinois agility test for training purposes. The study concludes that, in addition to test times, the use of instrumentedchairs, a higher sample size and the application of the method presented are recommended for Illinois agility test assessment.

Keywords: Illinois agility test, wheelchair athletes’ performance, wheelchair rugby sport, tangential velocity, turning radius

Introduction

First introduced in 1977 in Canada, and granted full

medal status at the 2000 Summer Paralympic Games

in Sydney, wheelchair rugby is a growing competitive

sport with more than 32 countries actively participat-

ing, of which 23 compete at a professional level

(International Wheelchair Rugby Federation, 2009).

The main objective of the game is to carry a ball

(typically a volleyball) across the opposing team’s

goal line. Wheelchair rugby dynamics can be defined

as starting, sprinting, braking, turning (pivoting) and

blocking, which are the skills that are inherently

related to handling the wheelchair (International

WheelchairRugby FederationTechnical Commission,

2008a; Usma-Alvarez, Subic, Burton, & Fuss, 2010;

Vanlandewijck, Daly, & Theisen, 2001).

Current rugby wheelchairs are specifically

designed to meet the individual anthropometric

needs of each athlete due to high level of correlation

ISSN 1934-6182 print/ISSN 1934-6190 online q 2010 Taylor & Francis

DOI: 10.1080/19346182.2011.564284

Correspondence: C.C. Usma-Alvarez, School of Aerospace, Mechanical and Manufacturing, Engineering, RMIT University, PO Box 71, Bundoora,

Melbourne, VIC 3083, Australia. E-mail: [email protected]

Sports Technology, August 2010; 3(3): 204–214

Dow

nloa

ded

by [

RM

IT U

nive

rsity

], [

Mis

s C

lara

Usm

a] a

t 21:

50 0

3 Ju

ly 2

011

between athlete’s performance and wheelchair

design. A specific setting in terms of seat height has

significant effects on a wide range of variables,

including mechanical efficiency and hand range of

motion due to change in hand rim reach (Woude,

Veeger, Rozendal, & Sargeant, 1989). These vari-

ables can have a substantial effect upon an athlete’s

ability to achieve specific speeds or manoeuvrability

capacity levels.

In general, the Illinois agility test is a timed task

involving straight sprinting and multidirectional

movement around stationary obstacles and it is

performed across a wide range of team sports

(Sheppard & Young, 2006). Studies to date involving

able-bodied sports indicate that agility performance

does not appear to be strongly linked with straight-

speed segment components (Buttifant, Graham, &

Cross, 1999; Young, Hawken, & McDonald, 1996).

These studies concluded that training for change of

direction, speed and agility must involve highly

specific training that recognises the individual

demands of the sport (Sheppard & Young, 2006).

In the setting of wheelchair rugby, the Illinois

agility test is highly regarded for training purposes.

However, to our knowledge, investigations of

performance criteria relating to the Illinois agility

test for wheelchair sports are not available, in spite of

the Illinois agility test’s use and regarded importance

for athletes and coaches in the field.

Qualitative studies to date have shown that

tangential velocities and turning radii during the

Illinois agility test are variables of high importance

for wheelchair rugby athletes and coaches (Usma-

Alvarez et al., 2010). However necessary, time

measurement alone does not appear to be sufficient

for a comprehensive performance assessment of a test

that involves a high proportion of multidirectional

movement.

In addition, the general setting of an Illinois agility

test favours controlled acceleration rather than

top speed as it involves extensive acceleration activity

with multidirectional body changes (Sheppard &

Young, 2006). Hence, if an athlete’s skill was to be

assessed from the results of the Illinois agility tests, it

would be beneficial to consider instantaneous radius

and tangential velocity of the turning wheelchair-

athlete system, as these variables impact the test’s

elapsed time.

The study presented in this article deals with the

use of accelerometer data in wheelchair sports

performance to develop and validate a method by

which to evaluate the tangential velocity and turning

radius achieved during the Illinois agility test. It also

presents an example of how the method can be used

to evaluate the performance of different athletes

through the use of performance hypotheses.

The main objectives of the study involve: (1)

recording athletes’ linear and angular accelerations

during the Illinois test; (2) developing and validating

a method that calculates the tangential velocity and

radius of curvature of a turning wheelchair; and (3)

analysing the athlete’s performance with the aid of

the method developed and the use of performance

hypotheses.

The following sections of this article will give a

detailed explanation of the experimental method

developed for determining the instantaneous radius

and tangential velocity of the turning wheelchair. In

addition, data obtained from wheelchair rugby

Paralympic athletes will be presented and analysed

against set performance criteria. To evaluate athletes’

performance, hypotheses for the performance par-

ameters of turning radius and velocity were

considered in order to favour elapsed time on each

turning segment of the test circuit. Results of each

individual athlete’s performance as well as

group categories will be discussed.

Experimental investigation

Test participants

A group of athletes from the Australian wheelchair

rugby Paralympic team volunteered to participate in

this study. The group was comprised of nine male

participants aged between 22 and 42 years-old, each

of whom had a significant amount of experience in

participating in wheelchair rugby at an elite level.

Table I presents test participant’s characteristics,

including disability level, sport classification as well

as their typical weekly training hours.

The performance data obtained from the nine

athletes (27 experiments) was used to develop and

validate the method to obtain the tangential velocity

and radius of curvature previously outlined in the

study’s objectives.

Per regulation, wheelchair rugby athletes are

placed within one of seven sport classifications,

which range from 0.5 to 3.5 points. The higher

number is intended for those athletes who have

higher physical function on court (International


2008b). Due to differences in athletes’ performance

levels and for the purpose of demonstrating a possible

performance analysis tool with the aid of the method

developed, the athletes and experimental results were

further categorised in line with their sports classifi-

cation into three groups (these are also outlined in

Table I): group A–low pointers: Athletes with sports

class categories ranging from 0.5–1.5; group B–mid

pointers: Athletes with sports class categories ranging

from 1.5–2.5; and group C–high pointers: Athletes

with sports class categories ranging from 2.5–3.5.

Performance analysis of the Illinois agility test in wheelchair rugby 205

Dow

nloa

ded

by [

RM

IT U

nive

rsity

], [

Mis

s C

lara

Usm

a] a

t 21:

50 0

3 Ju

ly 2

011

These groupings have been developed based on the

approach that was adopted by Mason, van der

Woude, and Goosey-Tolfrey (2006), and with the

aim of presenting the analysis of the performance

data obtained, highlighting effects of sports classifi-

cations and demonstrating analysis of significance

between participants as a performance analysis

example using the method developed in this research.

The categorisation of athletes ensures that the

results are matched against athletes of similar

characteristics (upper body range of mobility).

Current performance values for the main categories

of wheelchair rugby athletes will be reported as an

example of method’s scope.

The athletes were fully informed of the research

procedures and potential risks and written consent

was obtained from all research participants. Ethics

approval was obtained from the RMIT University

Human Research Ethics Committee prior to the

experimental study.

Instrumentation

Because of the strong correlation between wheelchair

configuration and performance, all athletes com-

pleted testing using their own custom built match

chairs. The study of experienced athletes’ perform-

ances using their own customised equipment is

appropriate and highly regarded, as the athlete’s

wheelchair is considered an extension of the athlete’s

body for game rules compliance (International


2008a).Hence, all conclusions on performance for the

purpose of this study will be based on wheelchair þ

athlete system. All test wheelchairs were specifically

designed to meet each athlete’s individual anthropo-

metric needs and the mechanical characteristics of

all chairs complied with section four of the

International Wheelchair Rugby Federation IWRF

rules (International Wheelchair Rugby Federation

Technical Commission, 2008a).

The wheelchairs were instrumented with a 6 DOF

(degrees of freedom) sensor (accelerometer þ

gyrometer) at 100Hz sample rate (minimaxX,

Catapult Innovations Pty Ltd, Scoresby, Australia)

which measured the linear accelerations and angular

velocities along/about the three axes of an orthogonal

coordinate system. The three acceleration readings of

the sensor were calibrated using a drop test, based on

gravitational acceleration (9.81m/s2) at small vel-

ocities. The three angular velocities were calibrated

with a turntable at known rpm values.

Each wheelchair was instrumented by securely

placing the sensor in the upright position (x-axis

directed to the left side of the chair’s forward position)

on the pick bar of the chair. The sensor’s positive y-axis

was positioned on the chair’s sagittal plane in line with

Table

I.Participants’data,wheelchairmass

and

dvalue.

Group

Participant

Disabilitylevel

andclassification

Age(year)

Mass

(kg)

Size(m

)Traininghours/w

eek

Elite

experience

(years)

Chairmass

(kg)

dvalue(m

)

A(low

pointer)

A1

C5–1.5

32

71.28

1.82

16

117.77

0.69

A2

C6–0.5

39

73.77

1.83

16

12

17.15

0.56

A3

C6–1.5

36

64.00

1.88

15

316.27

0.59

B(m

idpointer)

B1

C6/7–2.0

24

87

1.87

10

217

0.54

B2

C7–2.0

34

71

1.87

15–20

10

17

0.57

B3

C6/7–2.0

22

84

1.97

20

418

0.55

C(highpointer)

C1

Spinalneu

rologicaldisorder–3.0

42

94.15

1.70

10

15

18.79

0.55

C2

C6/7

Inc–3.0

23

70.09

1.85

18–20

218.94

0.53

C3

C7Inc–2.5

40

95.00

1.90

10

16

17.23

0.57

C. C. Usma-Alvarez et al.206

Dow

nloa

ded

by [

RM

IT U

nive

rsity

], [

Mis

s C

lara

Usm

a] a

t 21:

50 0

3 Ju

ly 2

011

the chair’s forward position. The distance between the

sensor’s centre and themidpoint of the line connecting

the hubs of the wheels (referred as d in Figure 2 and

Table I) was recorded for each athlete.

Testing protocol – Illinois agility test. The experimental

tests were carried out during a single day session at an

indoor basketball court in exactly the same

conditions as a regular Illinois agility test setting to

ensure that the athletes familiarised quickly with the

task. In this specific setting of the Illinois agility test

the athletes were required to sprint across a

predetermined path where 11 turns around

different gates took place (as shown in Figure 1).

The length of the course was 10m and the width or

distance between the start and finish points was 5m.

Four cones were used to mark the start, finish and the

two turning points. Another five cones were placed at

the centre of the course and spaced 1.5m apart,

starting 2m from the finish line as shown in Figure 1.

Testing was performed by lining up the athlete on

the starting line. Once an athlete was ready to start,

the sensor was switched on and the athlete was

requested to remain still for 5 s in order to normalise

acceleration data (reading of zero acceleration). On

Figure 1. Illinois agility test setting provided by the athletes’ coach.

Figure 2. Top view of a rugby wheelchair: d ¼ distance between centre of accelerometer and midpoint of the line connecting the hubs of the

wheels (C), IC ¼ instant centre of the turning wheelchair, rc ¼ turning radius of the wheelchair (from IC to C), ra ¼ distance between IC

and centre of accelerometer, u ¼ angle between ra and rc, ax and ay ¼ components of the instantaneous resultant of the horizontal

acceleration, a r and a t ¼ radial and tangential acceleration.


Dow

nloa

ded

by [

RM

IT U

nive

rsity

], [

Mis

s C

lara

Usm

a] a

t 21:

50 0

3 Ju

ly 2

011

the “go” command a stopwatch was initiated and the

athlete pushed around the course in the direction

indicated without knocking the cones over (Figure 1).

Athletes were requested to perform the drill in same

conditions as a regular training session. The overall

trial time was recorded once the athlete reached the

finish line and accelerometer data were recorded for

further 5 s with the chair completely stopped. Three

different data sets were taken for each athlete with

2min rest in between sets. The measurement

protocol was the same for all participants and

reference video footage was recorded for each trial.

Data processing

To synchronise stopwatch and accelerometer data,

the slight offset of the data at zero acceleration (due

to normalising activity prior to test) was corrected

such that ay (forward/backward acceleration) over-

lapped the zero acceleration line before evident

acceleration activity (start of stopwatch) was

initiated. Video data was also taken on same “go”

command as the stopwatch and was used as reference

to further analyse acceleration data. This visual data

helped in the identification of sections and turns of

the Illinois agility test setting in the acceleration

dataset as peaks in the angular acceleration curve.

Original and filtered data were analysed and

percentages of real numbers retrieved were com-

pared. Highest percentages of real numbers were

achieved by using original (not filtered) data.

Therefore filtering was avoided for obtaining

experimental results. Further data processing

included conversion of angular velocity data about

vertical axis from deg/s to rad/s.

The three trials for each athlete were averaged and

results analysed according to the method developed

for determining the instantaneous radius and

tangential velocity as explained in the following

section.

Method for determining the instantaneous radius and

tangential velocity of a turning wheelchair

The instant centre of turning (IC, Figure 2) is located

on a line connecting the two hubs of the two wheels.

From Figure 3, the instantaneous turning radius (rc)

corresponds to the distance between IC and the

centre of the wheelchair (C). The tangential velocity

(v) of the wheelchair is determined from its angular

velocity (vz) and the radial acceleration (a r) of the

accelerometer:

ar ¼ axcosuþ aysinu ð1Þ

sinu ¼d

ra

ð2Þ

ra ¼ar

v2z

ð3Þ

where d is the distance between centre of accel-

erometer and C; ra is the distance between IC and

centre of accelerometer; v is the angle between ra and

rc; ax and ay are the components of the instantaneous

resultant of the horizontal acceleration (Figure 2).

The unknowns of Eqs (1–3) are a r, v and ra.

Substituting Eq. (3) in Eq. (2) and the latter in

Eq. (1) yields:

ax

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi12

v2zd

ar

� �2s

2 ayv2

zd

ar2 ar ¼ 0 ð4Þ

Solving the 2nd order Eq. (4) for a r yields two

Figure 3. Angular velocity, turning radius and tangential velocity plot for Illinois agility test analysis.


Dow

nloa

ded

by [

RM

IT U

nive

rsity

], [

Mis

s C

lara

Usm

a] a

t 21:

50 0

3 Ju

ly 2

011

solutions:

ar1 ¼

ffiffiffiffiffiffiffiffiffiffiffiffiffiB þ C

2

rð5aÞ

and

ar2 ¼

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiB 2 C

2

rð5bÞ

where:

B ¼ a2x 2 2d ayv

2z ð6Þ

and:

C ¼

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiB2 2 4d 2v4

z a2x þ a2

y

� �rð7Þ

Eq. (5b) delivers a r , dwhich implies thatv in Eq. (2)

is no longer real. C in Eq. (7) can, in rare cases, be

imaginary, if 4d 2v4z a2

x þ a2y

� �. B2. The angle v is

calculated from Eq. (2), and rc and v from

rc ¼d

tan uð8Þ

v ¼ vzrc

j j ð9Þ

When moving along a straight path, v z ¼ 0 and rc¼ 1. In order to obtain the data v z, rc, v at the turns

only, all data with rc . 2m were removed (Figure 3).

Subsequently, the average rc and v were calculated for

each turn.

Validation

In order to achieve the second objective of this study

and prove the validity of the developed method, it was

necessary tohave a known radius andvelocity value as a

target for expected performance in order to eliminate

ambiguity of an unknown target value (i.e. skill based

validation). A rugby wheelchair, instrumented as

shown in Figure 2 was moved twice around the

circumference of a circle (7208), and the time taken to

do this was recorded with a stopwatch. Thewheelchair

was moved at different velocities around circles of

diameters of 2.16m, 1.36m and 0.72m, resulting in

average linear speeds of 0.5–2m/s. Additionally, the

accelerometer was placed on the turntable of a

polishing machine that was rotated at 50, 55, 60, 65

and 70 rpm with the accelerometer unit of the sensor

placed at a distance of 100mm from the rotation

centre. From the data that was recorded, rc was

calculated and compared to the original radius. The

radius calculated was on average 1.015 times larger

than the original one, resulting in an error of 1.5%.

The coefficient of determination (r2) of the linear

regression through origin was 0.9958.

Data analysis

A typical angular velocity plot of the Illinois agility

test drill is shown in Figure 3. From this figure, the

number of turns performed during the test and their

direction can be identified as the plot peaks (black

curve). The turning radius (rc) of the wheelchair (red

curve) as well as the linear velocity (v) (green curve)

achieved on each turn are also presented in this

graph. As mentioned above, when v z ¼ 0, rc ¼ 1.

Therefore, rc data larger than 2m were removed in

order to confine the data to turning radii of # 2m.

From Figure 3 is also noticeable that, as expected,

the velocity data seem to be larger when vz

approached zero. As point of reference zero vz is

indicated in blue in Figure 3. Calculation of the mean

of v and rc of each peak provided results for analysis of

each turn of the Illinois agility test for each of the

participants.

Average time for each participant [A (1–3),

B (1–3), C (1–3)] as well as category average are

presented in Table II. Descriptive statistics were used

to analyse each participant’s data as well as category

subgroups (A, B, C) effects. Statistical significance

was denoted by P , 0.05.

Demonstration of performance analysis using the method

developed

In order to analyse the performance of test

participants, a predefined set of performance

parameters were required. In order to favour elapsed

time and to relate the analysis of the test to

wheelchair rugby game dynamics, the Illinois agility

test results for all nine athletes were analysed as to the

following optimisation of performance:

(1) Velocity v: the larger v the better; large v results

in fast movements during turn.

(2) Turns 3, 4, 5, 7, 8 and 9: the longer rc the better;

longer rc enables direct path, closer to obstacles

gates or cones.

(3) Turns 2, 6 and 10: the shorter rc the better; short

rc reduces the gap between wheelchair and

obstacles, gates or cones.

Table II. Test time average in seconds for each athlete and

classification group (e.g. Athlete A1 ¼ 28.72 s, classification group

A ¼ 30.42 s).

1 2 3 Classification average

A 28.72 31.95 30.59 30.42

B 29.58 27.25 27.39 28.07

C 29.40 28.06 29.04 28.83


Dow

nloa

ded

by [

RM

IT U

nive

rsity

], [

Mis

s C

lara

Usm

a] a

t 21:

50 0

3 Ju

ly 2

011

(4) Turns 1 and 11: no clear trend: longer rc allows

the velocity to be maintained, whereas shorter rc

results in less time spent performing the turn at a

smaller velocity. The decision between these two

options is guided by energy principles: it is

preferable to maintain the kinetic energy rather

than decelerating before and accelerating after

the turn.

The hypotheses based on optimisation strategy of

performance, which were tested in this study, were as

follows:

(1) athletes produce a longer rc in turns 3, 4, 5, 7, 8

and 9;

(2) athletes produce a shorter rc in turns 2, 6 and 10;

and

(3) athletes produce a longer rc in turns 1 and 11.

According these hypotheses, and following the test

circuit shown in Figure 4a (turn identification

number in the test circuit), four performance

quadrants can be identified for different turn radii

and velocities as shown in Figure 4b.

Is important to note that a hypothesis of expected

performance is necessary prior to analysis of results

and is subjected to athlete or coach demands for

performance targets. For instance, a desired per-

formance can be targeted to suit a specific game

strategy or to train specific athlete’s skills. This

hypothesis is to be confirmed or rejected with

application of the method proposed. Assessment

(acceptance or rejection) and discussion of hypoth-

eses for the complete research sample will be

presented on the following sections.

Figure 5 presents data for participant B3 as an

example of the results analysed for each of the

athlete’s data. The turning radius rc and tangential

velocity v is plotted against the turn number for three

runs performed by each of the athletes. In addition,

averaged trials for each athlete were used to assess

change of direction performance for each athlete

according to values of tangential velocity and

instantaneous radius for each turn. This information

is presented for participant B3 in Figure 6, where the

tangential velocity v (m/s) ( y axis) against the

instantaneous radius rc (m) (x axis) are plotted for

each of the Illinois turns (numbered in Figures 1 and

3) to show the location of the turn numbers on the

Figure 4a. Turn identification number in test turning circuit.

Figure 4b. Performance quadrants for radii and tangential velocity

performance by turn number identified in Figure 4a.


Dow

nloa

ded

by [

RM

IT U

nive

rsity

], [

Mis

s C

lara

Usm

a] a

t 21:

50 0

3 Ju

ly 2

011

quadrant plot (Figure 4) for assessment of the stated

hypotheses.

Results

The results showed no difference in turning radius

between expected longer (turns 3, 4, 5, 7, 8 and 9)

and shorter radius (turns 2, 6 and 10) turns in

individual athletes (hypotheses (1) and (2) respect-

ively). Only a single athlete exhibited a statistically

significant difference. Hence, hypotheses (1) and (2)

and longer/shorter radius as a performance par-

ameter were rejected.

The velocity found in the turns with the expected

shorter radius (2, 6 and 10) is generally larger than

the one in the expected longer radius (turns 3, 4, 5, 7,

8 and 9) with a statistically significant difference

(P , 0.05) in four of the nine athletes.

Turns 1 and 11 showed a tendency for higher radii

for all athletes with significant velocity values during

the turn by classification group. Hence, hypothesis

(3) of longer radius to maintain velocity as a

performance parameter was accepted.

If turns 1, 2, 6 and 11 are classified as . 908 then

turns , 908 should have a larger turning radius. This

was significantly true for one athlete, and insignif-

icant for all others.

When analysing experimental results as classifi-

cation groups, tangential velocities and turning radii

achieved by each category of athletes on each of the

test turns were outlined as data points and are

presented on Figures 7, 8 and 9. There was

significant difference between radii of group A and

B (P # 0.02) and a larger radius was observed in

group A (low pointer athletes). The obtained average

radii for groups A, B and C were: A: 0.98 ^ 0.13m,

B: 0.94 ^ 0.11m and C: 0.95 ^ 0.10m.

Investigation of differences in velocities between

groups A and C as well as between B and C were

significant (P # 0.001 each). Smaller velocities were

found in the low and mid pointer groups. The

obtained velocity for group A, B and C were: A:

2.28 ^ 0.30m/s, B: 2.32 ^ 0.19m/s and C:

2.41 ^ 0.22m/s.

Figure 10 presents the differences in velocity and

radius for all athletes. Figure 10a shows more

velocity significances between outlined clusters of

Figure 5. rc and v against turn number for three runs performed by

participant B3.

Figure 6. rc (m) against v (m/s) plot results for participant B3.

Figure 7. Group A: low pointer skill range–n ¼ 3 participants with

sports class categories ranging from 0.5–1.5.


Dow

nloa

ded

by [

RM

IT U

nive

rsity

], [

Mis

s C

lara

Usm

a] a

t 21:

50 0

3 Ju

ly 2

011

athletes with a smaller standard deviation.

Figure 10b shows less significance in turning radius

between outlined clusters of athletes with a larger

standard deviation.

Stopwatch time (Table II) does not correlate

strongly with mean turning velocity. Still, the general

trend is: the shorter the stopwatch time, the faster the

linear velocity during turns. However, the coeffi-

cients of determination (r2) and correlation (r) are

0.276 and 0.526 respectively, indicating that only

27.6% of the stopwatch time is due to the linear

velocity during turns. Figure 11 presents the

stopwatch time against linear velocity v of all 27

experiments.

Discussion of results

The results of current study demonstrate that, in

addition to test times, tangential velocities and

instantaneous radii can also be studied to achieve a

better performance assessment of athletes during a

turning task test.

Hypotheses assessment for the available research

sample showed that only hypothesis (3) could be

accepted as performance parameter.

The assessment of the performance parameters

hypothesised, namely the turning radius and velocity,

showed that only the performance parameter of

velocity seems to be relevant for the assessment of the

available sample, since it was found to have more

significance than the parameter of turning radius.

Smaller velocities were found in both in the low

and mid pointer groups. This confirmed the

importance of faster velocity as a performance

parameter, as group C was known to be the higher

performance group. As evident from Figure 10,

velocity was a better performance parameter than

radius.

The significant difference between radii of groups

A and B (P ¼ 0.02) and the larger radius observed in

group A (low pointer athletes) is understandable as

category groups have a known variance in perform-

ance ability. In general, group A is expected to have

the most reduced limb ability out of the three groups.

In addition, the wheelchairs used by group A athletes

are usually longer and less manoeuvrable in terms of

turning activity than those used by athletes in the

other groups.

According to the hypotheses outlined, larger radii

in low pointers (group A) would be understandable if

the velocity was higher. However, it was evident that

radius was not an appropriate performance par-

ameter in this case.

In regards to test times; the fact that the stopwatch

time did not correlate strongly with the mean velocity

is understandable considering that the test time was

also dependent on the speed achieved in the straight

segments of the test (before and after turns 1 and 11)

whereas the velocity presented in the study was

determined only at the turns.

Low velocities were observed for participants B2

and C2 on turns other than 1, 2 and 11. Yet, these

participants recorded two of the fastest times

according to the stopwatch data. Further examin-

ation of visual data for the two participants displayed

a more controlled turning technique during the

continuing turning section of the test. Their

technique correlates with their rc curves, as high

accuracy on turning keeps smaller radii and thus,

generally constant angular velocity was observed.

Consequently, the overall test time helps to

Figure 8. Group B: mid pointer skill range–n ¼ 3 participants with

sports class categories ranging between 1.5–2.5.

Figure 9. Group C: high pointer skill range–n ¼ 3 participants

with sports class categories ranging between 2.5–3.5.


Dow

nloa

ded

by [

RM

IT U

nive

rsity

], [

Mis

s C

lara

Usm

a] a

t 21:

50 0

3 Ju

ly 2

011

determine which athletes are faster at turns and

slower at straight segments, and vice versa.

If agility performance is not strongly linked with

straight-speed segment components as outlined by

Sheppard and Young (2006), then the current

assessment of performance by stopwatch times only

is not accurate. Observation of the behaviour of

athletes performing the current agility test setting as a

familiar procedure, leads to a questioning of the

validity of the current agility test. Current test

performance seem to link agility to velocity (faster to

complete test ¼ more agile). Test assessment does

not seem to consider a small or large radius

depending on obstacle placement, hence the weak

correlation with stopwatch data. The value of the

current Illinois agility test for assessing an athlete’s

performance is questionable if athletes can compen-

sate less agility on multidirectional movement by

speeding up through straight segments.

Conclusions

The ability to achieve rapid and repeated whole-body

movement with change of direction is one of the

determinants of sports performance (Sheppard &

Young, 2006). Hence, detailed data related to

instantaneous turning behaviour is important in

sports performance analysis and training. This

research demonstrated that additional to tests

times, tangential velocities and instantaneous radii

are variables that can be investigated when qualifying

the agility level of a specific athlete. The trial time

alone is not a sufficient assessment for this type to test

as is closely related to straight line paths rather than

to change of direction activity.

The presented method is adequate for analysing an

athlete’s turning performance and technique during a

turning drill. Turning accuracy, test techniques,

achieved velocities whilst turning and test times can

be analysed with the aim of assessing or improving

athletes’ skills and game strategy.

The developed method was shown to be valid for

assisting the performance assessment of wheelchair

Figure 10. Linear velocity v against radius rc,; a: the difference in velocity between the clusters of athletes marked with a grey ellipse is

significant at p , 0.01, b: the difference in radii between the clusters of athletes marked with a grey ellipse is significant at p , 0.01.

Figure 11. Stopwatch time against average linear velocity v of all 27

experiments. Dashed line ¼ linear regression.


Dow

nloa

ded

by [

RM

IT U

nive

rsity

], [

Mis

s C

lara

Usm

a] a

t 21:

50 0

3 Ju

ly 2

011

sports athletes using the Illinois agility test for

training purposes.

The evaluation presented in this paper has been

performed and presented from a sports engineering

perspective based on the investigation of change of

direction speed and radius of turning of the wheel-

chair. In the context of sports science, Sheppard and

Young (2006) have offered the most recent definition

of the term “agility”: “a rapid whole-body movement

with change of velocity or direction in response to a

stimulus”. This study does not include factors such

as an athlete’s reaction component or decision

making. Hence, the researchers aim is not to assess

the athlete’s agility but to provide a method that

could serve as a performance assessment tool.

Desired performance parameters should be analysed

and set according to match strategy and an athlete’s

performance target for confirmation of the

hypothesis.

Averaged trials were used to analyse the athlete’s

performance as mode of demonstration with avail-

able data samples. When assessing an athlete’s

performance using this or any other method, a

greater data sample of trials per athlete is

recommended to improve the accuracy of an athlete’s

performance results, as well as to analyse the

consistency of a participant’s performance.

We suggest improving the performance assessment

of the Illinois agility test by using instrumented

wheelchairs and applying the method described in

this paper. Although, only the linear (tangential)

velocity seems to be a performance parameter, and

turning radius was not found to be an accurate

measure of performance in the sample analysed,

applicability of the turning radius could be a potential

performance parameter if a fixed path as mobility

target was enforced in the protocol and valid for

training purposes.

References

Buttifant, D., Graham, K., & Cross, K. (1999). Agility and speed

measurement in soccer players are two different performance

parameters. In Fourth world congress of science and football.

Sydney: University of Technology, February 22–26.

International Wheelchair Rugby Federation Technical Commission

(2008a). International Wheelchair Rugby Federation rules 2008.

Retrieved October 27, 2008, from http://www.iwrf.com/rules.

htm

International Wheelchair Rugby Federation Technical Commission

(2008b). IWRF classification manual (3rd ed.). RetrievedApril 30,

2009, from http://www.iwrf.com/classification.htm

International Wheelchair Rugby Federation (2009). International

Wheelchair Rugby Federation homepage. Retrieved May 12, 2009,

from http://www.iwrf.com

Mason, B. S., van der Woude, L. H. V., & Goosey-Tolfrey, V. L.

(2009). Influence of glove type on mobility performance for

wheelchair rugby players.American Journal of Physical Medicine &

Rehabilitation, 88 (7), 559–570.

Sheppard, J. M., & Young, W. B. (2006). Agility literature review:

Classifications, training and testing. Journal of Sports Sciences,

24(9), 919–932.

Usma-Alvarez, C. C., Subic, A., Burton,M., & Fuss, F. K. (2010).

Identification of design requirements for rugby wheelchairs using the

QFD method. Procedia Engineering, 2 (2), 2749–2755.

Vanlandewijck, Y., Daly, D. J., & Theisen, D. M. (2001).

Wheelchair propulsion biomechanics: Implications for wheel-

chair sports. Sports Medicine, 31(5), 339–367.

Woude, L. V. d., Veeger, D., Rozendal, R., & Sargeant, T. (1989).

Seat height in handrim wheelchair propulsion. Journal of

Rehabilitation Research and Development, 26, 31–50.

Young, W. B., Hawken, M., & McDonald, L. (1996).

Relationship between speed, agility, and strength qualities in

Australian Rules football. Strength and Conditioning Coach, 4(4),

3–6.


Dow

nloa

ded

by [

RM

IT U

nive

rsity

], [

Mis

s C

lara

Usm

a] a

t 21:

50 0

3 Ju

ly 2

011

Advanced performance analysis of the Illinois agility test based on the tangential velocity and...

Documents

Transcript of Advanced performance analysis of the Illinois agility test based on the tangential velocity and...