INFLUENCE OF PHYSICAL FITNESS, AGE, EXPERIENCE,

AND WEEKLY TRAINING LOAD ON MATCH

PERFORMANCE IN ELITE AUSTRALIAN FOOTBALL
PAUL B. GASTIN,1 BRENDAN FAHRNER,1,2 DENNY MEYER,3 DEAN ROBINSON,4

AND

JILL L. COOK1,5

Center for Exercise and Sports Science, School of Exercise and Nutrition Sciences, Deakin University, Burwood, Victoria,
Australia; 2Richmond Football Club, Melbourne, Victoria, Australia; 3Faculty of Life and Social Sciences, Swinburne
University of Technology, Hawthorn, Victoria, Australia; 4Essendon Football Club, Melbourne, Victoria, Australia; and
5
Department of Physiotherapy, School of Primary Health Care, Monash University, Peninsula Campus, Victoria, Australia
ABSTRACT
Gastin, PB, Fahrner, B, Meyer, D, Robinson, D, and Cook, JL.
Influence of physical fitness, age, experience, and weekly
training load on match performance in elite Australian football.
J Strength Cond Res 27(5): 12721279, 2013Season long
competition schedules in football create unique challenges for
coaches in balancing the requirements of recovery, developing
and maintaining physical fitness, and adjusting the training load
before each match. The aim of this study was to investigate the
influence of player characteristics (physical fitness, age, and
playing experience) and weekly in-season training load on elite
match performance across an Australian football season.
Twenty-five players (age: 24.1 6 3.0 years; height: 188.3 6
7.3 cm; weight: 90.4 6 8.3 kg) from one elite team participated
in this study. Before the season, players age, experience,
height, and weight along with measures of aerobic (6-minute
run) and anaerobic (6 3 40 m repeated sprints) physical fitness were recorded. Individual player training load during the
season was measured using global positioning system technology for the main training session of the week. Player match
performance was calculated weekly from 33 individual playing
statistics. Multilevel modeling was used to investigate the relationship between weekly training load and match performance and to explore the influence of player characteristics
on this relationship. Playing experience (p , 0.01) and aerobic
fitness (p , 0.05) displayed positive relationships with performance, whereas player age (p , 0.01) showed a negative
relationship. Most players coped well with weekly variations in
training load; however, the relationship was moderated by the
results of the preseason repeated sprint test (p , 0.05). The
adverse effect on playing performance in selected players after
a more intense training session suggests that recovery from the
Address correspondence to Dr. Paul B. Gastin, paul.gastin@deakin.edu.au.
27(5)/12721279
Journal of Strength and Conditioning Research
2013 National Strength and Conditioning Association

1272

the

session may be delayed in players who exhibit a better anaerobic

fitness profile.

KEY WORDS periodization, aerobic fitness, repeated sprints,

GPS, playing statistics
INTRODUCTION

eason long competition schedules in team sports

such as in the football codes create unique challenges
for coaches in balancing the requirements of recovery, developing and maintaining physical fitness and
skill, and adjusting the training load before each match. The
prescribed training load is typically influenced by the periodized plan along with weekly variations in schedule and performance. Increasing levels of fatigue within a week or losing
fitness over the season both have potential detrimental effects
on performance. Although it is assumed that varying the training stimulus at regular intervals will prevent plateaus in training
responses (14) and that undulating nonlinear periodization
models are typically preferred in season (32), current understanding of the serial link between weekly variations in training
load and playing performance in team sports remains limited.
The literature suggests that player characteristics, such as
anthropometry, physical fitness, and playing experience, may
influence success in soccer (6,7,18) and Australian football
(21,24,27,31,34,35), although most studies have focussed on
the effect of these variables on team selection (21,31,34,35),
the likelihood of being drafted (27), and match running performance (6,7) rather than match performance per se. Australian football is a contact invasion sport that allows disposal of
the ball by hand and foot and is characterized by large running
distances and frequent high-intensity efforts (16). Athleticism,
speed, acceleration, and repeated sprint ability are considered
integral to success in the sport. Mooney et al. (24) have
recently demonstrated a relationship between the Yo-Yo intermittent recovery test (level 2) and the number of ball disposals
in elite Australian football, with the relationship being mediated by the level of high-intensity running within the match
and by playing experience. In contrast, in the same study,

TM

Journal of Strength and Conditioning Research

Copyright National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.

the

TM

Journal of Strength and Conditioning Research

similar relationships were not found with player impact scores
and coach performance ratings, which are considered more
appropriate measures of individual player performance than
disposals alone (17). Data were also limited to a small number
of players, matches, and the early rounds of the season.
As such, the purpose of this study was to investigate the
relationship between variations in weekly training load on
individual player match performance and to assess the
influence of player characteristics on this relationship. We
hypothesized that physical fitness and playing experience
would be positively related to individual player match
performance, whereas planned variations in the weekly
in-season periodized training load, designed to achieve both
short- and long-term success, would have little adverse effect.

METHODS

| www.nsca.com

played) and age (years) were recorded on January 1, 2008, as

listed in the Australian Football League (AFL) Record (23).
Measures of aerobic and anaerobic running fitness were
taken at the end of the preseason when at peak level (13).
Measures of individual player training load and match performance were recorded weekly throughout the season.
Subjects

Twenty-five players from one elite AFL team, all of whom

played at least one senior match during the 2008 season,
were recruited for this study. The Deakin University Human
Research Ethics Committee approved the study, and written
consent was obtained from all players before the commencement of this study.
Procedures

All data were collected as part of routine player assessment

during training and competition. Players were fully proThis prospective study investigated the influence of weekly
fessional and competing in the AFL, which is the highest
training load (primary independent variable) on individual
level of competition in the sport. A consistent pattern of
player weekly match performance (dependent variable). In
training was prescribed each week, and players were
addition, it evaluated both the direct and potential moderexpected to be appropriately prepared for all sessions,
ating influence of individual player characteristics (secondary
including attention to hydration, nutrition, sleep, and other
independent variables) such as age, playing experience, and
recovery practices. Aerobic fitness was measured in the field
measures of running fitness. Weekly variations in training
using a 6-minute continuous run for maximal distance. The
load were a result of the prescribed training program for the
total distance run was used as an indication of aerobic
team, designed to achieve short-term (i.e., weekly success)
power, based on the concept of velocity at V_ O2max and the
time limit to exhaustion (4), with 6 minutes equivalent to
and longer-term (i.e., success in the finals) performance
a work intensity of approximately 105% V_ O2max and time
objectives, and typical variations in training involvement of
spent at V_ O2max around 4 minutes (3). The well-maintained
individual players. Multilevel linear modeling (28) was used
grass oval had a running track measured to 400 m using
to account for repeated measures over 22 playing weeks,
a calibrated trundle wheel, with markers placed every 20
with individual weekly training load used as a level 1 prem to ensure that the total distance was accurately recorded.
dictor of match performance and player characteristics used
As such, the test had a maximal error of 610 m. The typical
as level 2 predictors. Playing experience (number of matches
error for this test in this playing
group, expressed as a coefficient
of variation, was 1.6% (n = 31).
A modified repeated sprint
test was used as a measure of
anaerobic high-intensity running fitness (30). Players completed a set of six 40-m sprints,
beginning every 15 seconds
through a system of timing
gates (Swift Technology, Lismore, NSW, Australia) set up
to allow out and back return
efforts. The work to rest ratio
Figure 1. Weekly training load and associated match performance throughout a regular season of elite Australian
was reduced compared to
football (N = 25 players). Training load in the main training session of the week was derived from GPS measures
other versions of this test in
and summarized as a single variable (speed exertion; expressed in arbitrary units [au]) in the manufacturers
an attempt to further increase
software (Team AMS; GPSports). The mean training load for the team ranged from 1,270 to 2,737 au across the
the anaerobic contribution and
season, whereas the range in individual player load ranged from 291 to 4032 au. Performance was derived from
33 individual player match statistics involving all aspects of play (i.e., offensive, defensive, stoppages) and
fatiguing nature of the test. The
weighted for importance by the coaching staff. The mean match performance (expressed in au) for the team
time for the first 40-m sprint
ranged from 88 to 139 au across the season, whereas the range in individual player performance ranged from 19
to 309 au.
was taken as a measure of the
players maximum speed. Total
Experimental Approach to the Problem

VOLUME 27 | NUMBER 5 | MAY 2013 |

1273

Copyright National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.

Fitness, Training Load and Performance in Football

blades and sampled at a rate of 1
Hz. The SPI Elite GPS device
TABLE 1. Bivariate relationships between player (N = 25) variables of age, playing
has been shown to provide accuexperience, and measures of running fitness.
rate (#4%) and reliable data for
Playing
6-min run 40-m 6 3 40-m Fatigue
total distance traveled during
Age
experience distance sprint
sprint
index
team sport movement patterns
(10). In replicating these running
Age
1.00
circuits, we have demonstrated
Playing
0.90*
1.00
less than 2% errors in total disexperience
6-min run
20.27
20.27
1.00
tance, peak speed, and 50-m
40-m sprint
0.40
0.44
20.06
1.00
straight-line efforts, and an 8%
6 3 40-m sprints
0.35
0.41
20.30
0.71*
1.00
error in 50-m zigzag efforts
Fatigue index
20.22
20.27
20.49* 20.56*
0.00
1.00
(15). After each training session,
*Correlation is significant at the 0.01 level (2-tailed).
data were downloaded from the
Correlation is significant at the 0.05 level (2-tailed).
units using the associated Team
AMS software package (version
1.2.1.12; GPSports).
The speed exertion variable was chosen from the software
time and fatigue index were determined from the six 40-m
as the best single descriptive measure of training load because
sprints, disregarding recovery time between sprints. Fatigue
it takes account of both total distance and the manner
index was calculated using the 2 equations suggested by
in which this distance is accumulated. The use of an exertion
Zagatto et al. (36). The typical error for this test in this
index based on GPS time-motion data has previously been
playing group, expressed as a coefficient of variation, was
described in Australian football, although the algorithms used
1.8% for 40-m sprint time and 1.6% for total time for six
by Wisbey et al. (33) were developed by the authors from the
40-m sprints (n = 24).
raw data. In our study, the speed exertion variable (measured
Training load was measured using GPS technology (SPI Elite
in arbitrary units [au]) calculated within the manufacturers
Unit; GPSports, Fyshwick, ACT, Australia) during the main
software (Team AMS) was used, which is derived from linear
football training session each week (commencing at 10 AM, 3
algorithms based on the distance covered in 6 user-defined
days before the match). This session was typically used to vary
speed zones. These speed zones were individualized for each
the training load in the periodized training plan, with other
player according to their maximum speed, as measured from
sessions earlier and later in the week more consistent in load
the first sprint of the repeated sprint test, and ranged from
and designed to focus on recovery from the match and techni10 to 110% of a players maximum speed (1025%, 2545%,
cal and tactical aspects leading in to the match. The GPS unit
4565%, 6580%, 8095%, and 95110%). The 6 speed zones
was worn in a harness located between the players shoulder

TABLE 2. Descriptive statistics for the dependent variable (match performance) and level 1 and level 2 predictors
(independent variables), along with the output for significant predictors from the multilevel linear model analysis.*
Variable Name
Match performance (au)
Level 1 predictors
Week number
Training load (au)
Level 2 predictors
Age (y)
Playing experience (matches)
6-min run (m)
40-m sprint (s)
Six 40-m sprints (s)
Fatigue index (au)

Mean

SD

310

108.5

48.9

19

310
11.1
310 2051.0

6.3
503.4

25
24.1
25
84.1
25 1658.4
25
5.41
25
35.56
25
10.07

3.0
63.0
76.6
0.16
0.91
2.52

Minimum Maximum

Coefficient

Significance (p)

309

N/A

N/A

1
291

22
4032

0.74
(0.0056 +
(0.0082 3 six 40-m)

0.027
0.023

19
0
1480
5.10
33.71
6.22

30
213
1800
5.69
37.33
15.67

212.66
0.72
0.14

0.005
0.001
0.044
NS
NS
NS

*au = arbitrary units; six 40 m = total time for six 40-m repeated sprints; N/A: Not applicable as match performance was the
dependent variable; NS = not significant, therefore not included in the final model.
The constant term in the model in the prediction of match performance = 101.61 au.

1274

the

TM

Journal of Strength and Conditioning Research

Copyright National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.

the

TM

Journal of Strength and Conditioning Research

| www.nsca.com

Victoria, Australia: http://

prowess.com.au/). Champion
Data provide 99% accuracy
for match statistics (25).
These statistics were weighted
for importance using a confidential formula developed by
the coaching staff and agreed
to at the beginning of the season. The impact score was measured in arbitrary units.
Statistical Analyses

Preliminary analysis of the

data, including confirmation
of statistical assumptions, and
assessment of bivariate relationships between variables
were undertaken using Statistical Package for the Social
Sciences for Windows (version
were linearly weighted from 1 to 6, similar to methods used
17.0, SPSS, Inc., Chicago, IL, USA). Pearsons correlation
to quantify load using heart rate technology (1,12), so that
coefficients (r) are discussed as very large (r $ 60.7), large
faster speed zones had a higher weighting. Therefore, a player
(r $ 60.5), moderate (60.5 . r $ 60.3), and small (60.3 .
who covered a greater proportion of their total distance in
r $ 60.1) (19). All data are presented as mean 6 SD and
higher speed zones would have a higher speed exertion comrange, with statistical significance set at p , 0.05.
pared with a player who covered a similar distance yet spent
To address the influence of training load and player
less time in these zones.
characteristics on match performance, all variables were
subsequently analyzed using a multilevel linear model
Performance Measures
(HLM version 6, Scientific Software International, Inc.,
Individual match performance was measured using an
Lincolnwood, IL, USA) (28). This analysis allowed for
impact rating system similar to those previously described
repeated measures of training load and performance (level 1
in Australian football (17,24,29). The impact score quanpredictors) to be modeled against each other while accounttified individual playing performance from 33 match
ing for differences in player characteristics (level 2 predictors)
statistics that incorporated all aspects of play (i.e., offenof player age, playing experience, aerobic, and anaerobic runsive, defensive, and stoppages) as collected by 2 AFLning fitness. As such, a predictive equation for match perforapproved companies (Champion Data, Victoria, Australia:
mance was produced showing the impact each variable had
http://www.championdata.com.au/; ProWess Sports,
on match performance after all other independent variables
had been accounted for. The
independent variables were all
centered before being added to
the model. Level 1 predictors
were player centered (i.e., each
players mean was subtracted
initially for all individual data
points), and level 2 predictors
were grand mean centered
(i.e., the group mean was subtracted initially for all individual
data points). This is an imporFigure 3. Raw data for paired samples of weekly training load and match performance for the player with (A)
tant procedure because it
the fastest repeated sprint time (37.3 seconds) and (B) the slowest repeated sprint time (37.3 seconds); N = 11
reduces the impact of multiand 18 paired samples, respectively. Linear regression (solid line) with 90% confidence intervals (dotted lines).
collinearity (e.g., the correlation
Correlation coefficients (r) were moderate in both sets of data, with r2 values explaining 15% and 9% of the
variance, respectively.
between age and experience
was very high).
Figure 2. Illustrated example of the multilevel linear model describing the influence of training load and match
performance. Data is shown in relation to the lightest (291 arbitrary units [au]), mean (2051 au), and heaviest (4032 au)
individual training loads recorded during the season. Mean data for the squad (N = 25) is shown along with modeled
data corresponding to the fastest and slowest times recorded for the repeated sprint test, which significantly (p =
0.023) moderated the relationship between training load and match performance. Match performance decreases for
the fastest player (repeated sprint = 33.7 seconds) as weekly training load increases, whereas for the slowest player
(repeated sprint = 37.3 seconds), match performance increases with increasing training load.

VOLUME 27 | NUMBER 5 | MAY 2013 |

1275

Copyright National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.

Fitness, Training Load and Performance in Football

Figure 4. Relationship between training load and match performance

for individual players (N = 19) who had at least 10 paired samples of
data (16.2 6 3.2 weeks). In this subset, 5 players showed positive
performance improvements of greater than 20% with increasing training
load, whereas 7 players showed negative performance responses of
greater than 20%. The length of each regression line represents
the minimum and maximum training load for each player. Correlation
coefficients (r) ranged from 20.60 to 0.48, with r2 values ranging from
0.001 to 0.23. au = arbitrary units.

RESULTS
Data are presented for 25 players (age: 24.1 6 3.0 years;
height: 188.3 6 7.3 cm; weight: 90.4 6 8.3 kg) across the 22
weeks (i.e., matches) of an AFL season. A total of 310 paired
training load and match performance samples were analyzed.
The weekly relationship between training load and match
performance throughout the season, including the variability
in the data, is illustrated in Figure 1. Bivariate analysis
indicated that age-playing experience and 40-m sprint-40-m
repeated sprints were highly correlated (Table 1). Sprint and
repeated sprint performance significantly declined with age
and playing experience. Level of fatigue in the repeated sprint
test decreased with improving aerobic fitness and increased
with improving player 40 m speed.
Descriptive statistics for preseason player characteristics
and in-season training load and match performance
measures along with the output of the multilevel linear
model for significant predictors are provided in Table 2. The
coefficients from the modeling can be used to create a linear
algorithm to predict performance:

Match performance au 101:61212:66 Age224:12


0:72 Experience284:12 0:14 6min run




21658 0:74 Week211:1 0:0056 0:0082



3 Six 40m sprints235:56 Training load


22051 :
Overall training load and week number had very little
impact on match performance with only 3.2% of the

1276

the

variability (i.e., r2) in the data being explained by these 2

variables. Individual player characteristics, however, had
a significant impact accounting for 45.3% of the total variability in the match performance data. Players age (negative
impact), playing experience (positive impact), and preseason
aerobic fitness (positive impact) were observed to have significant coefficients when predicting match performance.
Week number was also significant in the model, indicating
improvements in match performance across the season. Variables from the repeated sprint test had no significant direct
influence on the prediction of match performance, although
the coefficient of training load was influenced by the speed at
which the players completed the six 40-m sprints (Table 2,
Figure 2). That is, as training load (i.e., speed exertion)
increased above the mean (2051 au), corresponding match
performance in players with good repeated sprint ability
decreased. Conversely, players with poor repeated sprint
ability coped well and showed improved performance.
Examples showing raw data for the 2 players who demonstrated the fastest and slowest repeated sprint times illustrate
this from a practical perspective (Figure 3).
Considerable variability existed in how individual players
responded to differences in training load and also in their
level of playing performance (Figure 4). Approximately
equal numbers of players improved, declined, or showed
little change in match performance when training load
increased. Absolute correlation coefficients (r = 0.23 6
0.16; range = 0.020.60) and r2 values (0.075 6 0.093;
range = 0.000.23) indicate that in many players the
relationship between training load and match performance
is small.

DISCUSSION
The aim of this study was to investigate the relationship
between weekly in-season training load and associated elite
match performance across an Australian football season.
The influence of player characteristics (physical fitness, age,
and playing experience) was also evaluated, both directly
and as moderating variables on this relationship. Playing
experience and aerobic fitness were found to have a significant positive effect on match performance, whereas player
age had a significant negative effect. The relationship
between training load and match performance was influenced by a players total time for six 40-m sprints, such that
variations in training load were tolerated in different ways
depending on a players anaerobic profile (i.e., repeated
sprint ability).
The team overall coped well with weekly variations in
training load with less than 10% variability in match
performance associated with the lightest and heaviest main
training session for the week (see the mean data in Figure 2).
The recommendations in the periodization literature that
undulating nonlinear loading models are typically preferred
in season (32) and that varying the training stimulus prevents

TM

Journal of Strength and Conditioning Research

Copyright National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.

the

TM

Journal of Strength and Conditioning Research

plateaus in training responses seem justified. The hypothesis
in this investigation that weekly variations in training load
would have little effect on match performance is supported
at a team level; however, this may not be the case at an
individual level.
There seem to be groups of players that either respond
positively, negatively, or fairly neutrally to increases in
weekly in-season training load. Although it is likely that
a complex mix of factors influence playing performance,
modeling of the available data indicates that performance on
the repeated sprint test moderates the relationship between
training load and match performance suggesting that players
at the extremes in repeated sprint ability respond differently
(i.e., slower players are able to tolerate higher training loads
better than their faster counterparts). This may be because of
individual differences in the degree of muscle damage
associated with a heavy training session along with the
speed of recovery. Muscle damage is influenced by a variety
of factors including exercise intensity, the number and
velocity of contractions during exercise, work performed,
exercised muscle length, and individual differences in fibertype composition and muscle architecture (8,26). Players
with enhanced anaerobic running profiles may reach faster
speeds in training (reflected in larger speed exertions and
calculated training loads) and experience greater eccentric
loading while decelerating (22). The recovery from fast-velocity eccentric exercise has been shown to be considerably
longer compared with slow-velocity exercise, despite muscle
soreness being similar in severity (26). For the players who
did not perform well on the repeated sprint test, improved
match performances associated with higher training loads
may also be reflective of increases in high-intensity running
fitness as the season progressed.
Variability in playing performance is clearly evident within
the data, particularly at an individual level (Figures 1, 3, 4).
This variability suggests that the specific demands of
Australian football, similar to other team sports, vary each
week because of its unpredictable nature (14). Individual
performance is not only dependant on physical fitness, training load, and recovery but also team-based factors such as
the influence of weather conditions, the effects of travel,
ground size and location, the strength of the opposition,
and tactics within the game (9,14,16,20,29).
The opposing influence of player age and playing
experience on match performance provides an insight that
has not been previously demonstrated. For example, Young
et al. (34) found that from a squad of 34 players who were
tested during preseason at one elite AFL club, the 22 players
who started the season playing in the senior side were significantly older and more experienced than the nonstarters.
In this situation, age and playing experience are confounding
variables because playing experience can only ever increase
with advancing age. The analysis approach employed in this
investigation allowed each of these variables to be
controlled. In doing so, it was found that playing experience

| www.nsca.com

was positively related to match performance, whereas age

was negatively related. This suggests that at one end of the
continuum, older players may compensate for declining
physical abilities with greater experience [e.g., enhanced tactical and competitive skills (24)], whereas younger players
may compensate for a lack of experience with greater athleticism. Mean age in this study (24.1 years) was similar to
the starting group in the Young et al. (34) study (24.0 years),
whereas playing experience was 84 and 90 matches, respectively. These likely represent typical means across the competition; however, from a list management perspective, our
findings suggest that exposing young players to more elite
matches early in their career may help to prolong a higher
performance output for a club over a longer period of time.
The data available to assess fitness characteristics and
individual match performance in football is limited, although
proxies for performance such as selection (21,31,34,35) and
running parameters within a match (6,7) have been used. This
study indicated that aerobic fitness, measured as a continuous
run for distance over 6 minutes, significantly influenced match
performance in our sample of 25 players and 310 matches.
Aerobic endurance training has been shown to significantly
improve performance in elite junior soccer by increasing running performance, work intensity, and the number of involvements with the ball during a match (18). Mooney et al. (24)
found a positive relationship between the Yo-Yo intermittent
recovery test (level 2) and ball disposals in elite Australian
football, yet failed to demonstrate similar relationships with
coach performance ratings and player impact scores similar
to those used in this study. The limited sample size (9 players)
and number of individual performances (21 matches) may help
explain these inconsistencies, although performance on the
Yo-Yo intermittent recovery test would be expected to correlate well with match performance, given it is a test designed to
match the running demands of field team sports and is influenced by a variety of physiological determinants (2).
An unexpected finding in this study was the lack of
influence of maximal speed and repeated sprint ability on
match performance. The repeated sprint protocol was modified to place considerable stress on anaerobic metabolism and
limit recovery (i.e., a 40-m sprint completed every 15 seconds),
although still meeting recommendations for repeated sprint
testing in field team sports (30). Given the frequency of sprinting in elite Australian football and the perceived importance of
high-intensity running distance during a match (11,33), it was
thought that repeated sprint ability would influence players
match performance. For example, high-intensity running distance is known to be a more discriminating variable compared
with total running distance when describing differences
between elite and sub-elite players in Australian football (5).
While further investigation is warranted, one possible explanation for our results relates to the relative homogeneity of
the data, such that the similarity in performance tests across
the playing group may mask potential differences that could
otherwise be evident in a more heterogeneous group.
VOLUME 27 | NUMBER 5 | MAY 2013 |

1277

Copyright National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.

Fitness, Training Load and Performance in Football

This study highlights the influence and the complex
interaction of player characteristics and program variables
that must be considered when training an elite Australian
football squad for competition. It reinforces the importance
of having a training program individualized to account for
differences in player characteristics within the team environment. Factors such as training load, player age, playing
experience, and physical fitness qualities may all impact
upon individual player match performance.

PRACTICAL APPLICATIONS
Varying the weekly training load in field team sports, as is
recommended in the literature, would seem justified from
a team perspective; however, individuals within the team
may respond differently depending on personal characteristics. Our data suggest that players with good anaerobic
running profiles, as evidenced by performance on a repeated
sprint test, appear less equipped to cope with a heavy
training load leading into a match than their slower teammates. Care must therefore be taken with the extent of the
load, the need to expose all players to the same load, and the
recovery time and practices provided before a match.
Routine monitoring of players in how they are coping with
training and competition loads along with detailed profiling
of players over time will assist in guiding player management
and individual training prescription.
The opposing influence of player age and playing
experience on match performance has implications for
player list management in elite Australian football.
Providing opportunities for young players to be exposed
to elite competition earlier in their career may ensure
a longer period of sustained performance in the years when
optimal physical capacity and playing experience overlap
(e.g., players in their mid 20s who are well experienced,
have a good training history, and remain physically capable
of performing and coping with a highly demanding game).
The significant relationship between 6-minute run and
match performance suggests that even at the elite level, small
gains in aerobic fitness may result in performance gains. The
importance of running fitness cannot be underestimated in
Australian football, a sport that demands mean total distances
of approximately 13 km, high-intensity running distances
(.14.4 km$h21) nearing 4 km, and 28 efforts above 23 km$h21
per match (11). As has been demonstrated in soccer (18),
coaches should be encouraged to use high-intensity aerobic
interval training to improve aerobic power, match running performance, and involvement in the play.

REFERENCES
1. Alexiou, H and Coutts, AJ. A comparison of methods used for
quantifying internal training load in women soccer players. Int J
Sports Physiol Perform 3: 320330, 2008.
2. Bangsbo, J, Iaia, FM, and Krustrup, P. The Yo-Yo intermittent
recovery test: A useful tool for evaluation of physical performance in
intermittent sports. Sports Med 38: 3751, 2008.

1278

the

3. Billat, LV. Interval training for performance: A scientific and

empirical practice. Sports Med 31: 1331, 2001.
4. Billat, LV and Koralsztein, JP. Significance of the velocity at VO2max
and time to exhaustion at this velocity. Sports Med 22: 90108, 1996.
5. Brewer, C, Dawson, B, Heasman, J, Stewart, G, and Cormack, S.
Movement pattern comparisons in elite (AFL) and sub-elite (WAFL)
Australian football games using GPS. J Sci Med Sport 13: 618623,
2010.
6. Castagna, C, Impellizzeri, F, Cecchini, E, Rampinini, E, and Alvarez, JC.
Effects of intermittent-endurance fitness on match performance in young
male soccer players. J Strength Cond Res 23: 19541959, 2009.
7. Castagna, C, Manzi, V, Impellizzeri, F, Weston, M, and Barbero
Alvarez, JC. Relationship between endurance field tests and match
performance in young soccer players. J Strength Cond Res 24:
32273233, 2010.
8. Chapman, D, Newton, M, Sacco, P, and Nosaka, K. Greater muscle
damage induced by fast versus slow velocity eccentric exercise. Int
J Sports Med 27: 591598, 2006.
9. Clarke, SR. Home advantage in the Australian football league.
J Sports Sci 23: 375385, 2005.
10. Coutts, AJ and Duffield, R. Validity and reliability of GPS devices for
measuring movement demands of team sports. J Sci Med Sport 13:
133135, 2010.
11. Coutts, AJ, Quinn, J, Hocking, J, Castagna, C, and Rampinini, E.
Match running performance in elite Australian Rules Football. J Sci
Med Sport 13: 543548, 2010.
12. Edwards, S and Reed, S. The Heart Rate Monitor Book for Outdoor and
Indoor Cyclists: A Heart Zone Training Program. Boulder, CO:
VeloPress, 2001.
13. Gabbett, TJ and Domrow, N. Relationships between training load,
injury, and fitness in sub-elite collision sport athletes. J Sports Sci 25:
15071519, 2007.
14. Gamble, P. Periodization of training for team sports athletes. Strength
Cond J 28: 5666, 2006.
15. Gastin, P and Williams, K. Accuracy of 1Hz versus 5Hz GPS devices
to measure movement patterns in team sport activities. J Sci Med
Sport 13: e32, 2010.
16. Gray, AJ and Jenkins, DG. Match analysis and the physiological
demands of Australian football. Sports Med 40: 347360, 2010.
17. Heasman, J, Dawson, B, Berry, J, and Stewart, G. Development and
validation of a player impact ranking system in Australian football.
Int J Perf Anal Sport 8: 156171, 2008.
18. Helgerud, J, Engen, LC, Wislff, U, and Hoff, J. Aerobic endurance
training improves soccer performance. Med Sci Sports Exerc 33:
19251931, 2001.
19. Hopkins, WG, Marshall, SW, Batterham, AM, and Hanin, J.
Progressive statistics for studies in sports medicine and exercise
science. Med Sci Sports Exerc 41: 313, 2009.
20. Kelly, VG and Coutts, AJP. Planning and monitoring training loads
during the competition phase in team sports. Strength Cond J 29:
3237, 2007.
21. Keogh, J. The use of physical fitness scores and anthropometric data
to predict selection in an elite under 18 Australian rules football
team. J Sci Med Sport 2: 125133, 1999.
22. Lakomy, J and Haydon, DT. The effects of enforced, rapid
deceleration on performance in a multiple sprint test. J Strength Cond
Res 18: 579583, 2004.
23. Lovett, M. AFL Record Season Guide. Melbourne: Geoff Slattery
Publishing, 2008.
24. Mooney, MG, Hunter, JR, OBrien, BJ, Berry, JT, and Young, WB.
Reliability and validity of a novel intermittent peak running speed
test for Australian football. J Strength Cond Res 25: 973979, 2011.
25. OShaughnessy, DM. Possession versus position: Strategic
evaluation in AFL. J Sports Sci Med 5: 533540, 2006.

TM

Journal of Strength and Conditioning Research

Copyright National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.

the

TM

Journal of Strength and Conditioning Research

| www.nsca.com

26. Paddon-Jones, D, Keech, A, Lonergan, A, and Abernethy, P. Differential

expression of muscle damage in humans following acute fast and slow
velocity eccentric exercise. J Sci Med Sport 8: 255263, 2005.

32. Wathen, D, Baechle, TR, and Earle, RW. Training variation:

Periodization, in: Essentials of Strength Training and Conditioning.
Champaign, IL: Human Kinetics, 2000. pp. 513527.

27. Pyne, DB, Gardner, AS, Sheehan, K, and Hopkins, WG. Fitness testing
and career progression in AFL football. J Sci Med Sport 8: 321332, 2005.

33. Wisbey, B, Montgomery, PG, Pyne, DB, and Rattray, B. Quantifying

movement demands of AFL football using GPS tracking. J Sci Med
Sport 13: 531536, 2010.

28. Raudenbush, S, Bryk, A, Cheong, YF, Congdon, R, and du Toit, M.

HLM 6: Hierarchical Linear and Nonlinear Modeling. Lincolnwood,
Illinois: Scientific Software International, Inc., 2004.
29. Richmond, LK, Dawson, B, Stewart, G, Cormack, S, Hillman, DR,
and Eastwood, PR. The effect of interstate travel on the sleep
patterns and performance of elite Australian Rules footballers. J Sci
Med Sport 10: 252258, 2007.

34. Young, WB, Newton, RU, Doyle, TL, Chapman, D, Cormack, S,

Stewart, C, and Dawson, B. Physiological and anthropometric
characteristics of starters and non-starters and playing positions
in elite Australian Rules football: A case study. J Sci Med Sport 8:
333345, 2005.

30. Spencer, M, Bishop, D, Dawson, B, and Goodman, C. Physiological

and metabolic responses of repeated-sprint activities: Specific to
field-based team sports. Sports Med 35: 10251044, 2005.

35. Young, WB and Pryor, L. Relationship between pre-season

anthropometric and fitness measures and indicators of playing
performance in elite junior Australian Rules football. J Sci Med Sport
10: 110118, 2007.

31. Veale, JP, Pearce, AJ, Koehn, S, and Carlson, JS. Performance and
anthropometric characteristics of prospective elite junior
Australian footballers: A case study in one junior team. J Sci Med
Sport 11: 227230, 2008.

36. Zagatto, AM, Beck, WR, and Gobatto, CA. Validity of the
running anaerobic sprint test for assessing anaerobic power and
predicting short-distance performances. J Strength Cond Res 23:
18201827, 2009.

VOLUME 27 | NUMBER 5 | MAY 2013 |

1279

Copyright National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.