Journal of Strength and Conditioning Research, 2004, 18(3), 486–490
q 2004 National Strength & Conditioning Association
ACTIVITY PROFILE OF INTERNATIONAL-LEVEL
SOCCER REFEREES DURING COMPETITIVE MATCHES
CARLO CASTAGNA,1 GRANT ABT,1,2 AND STEFANO D’OTTAVIO3
1
Research Department, teknosport.com, Ancona, Italy; 2St. Martin’s College, Lancaster, United Kingdom; 3Tor
Vergata University, Rome, Italy.
ABSTRACT. Castagna, C., G. Abt, and S. D’Ottavio. Activity pro-
file of international-level soccer referees during competitive
matches. J. Strength Cond. Res. 18(3):486–490.—The purpose of
this study was to investigate the activity profile of soccer ref-
erees competing at international level. Thirteen international-
level soccer referees (INRG, age 38 6 3 years, height 182.0 6
6.5 cm, and body mass 78.8 6 7.0 kg) were observed during
official international matches using a match analysis system
(Play Controller, Phromos, Italy) that enabled speeds and times
collection for 11 arbitrarily chosen match categories. For com-
parison, 13 Italian elite-level soccer referees (NRG, age 37 6 3
years, height 182.5 6 3.5 cm, and body mass 77.1 6 6.5 kg) were
examined during domestic official first division matches (Serie
A). Results showed that NRG covered more distance during
matches than the INRG (12,956 6 548 m and 11,218 6 1,056
m, respectively; p , 0.05). Running faster than 18 km·h21, NRG
covered more distance than INRG (2,378 6 423 m and 1,642 6
689 m, respectively; p , 0.05). No between-halves differences in
the distance covered at speed faster than 18 km·h21 were ob-
served in each group. The main finding of this study is that
international-level referees during international competitions
are less active than national-level officials directing domestic
matches. This unexpected result may suggest the need for spe-
cific training for the international referees, as recent reports
have shown that better positioning is associated with longer
space coverage during competitive matches. Across-halves, high-
intensity, space-coverage conservation should be regarded as a
peculiarity of the elite-level soccer referee.
KEY WORDS. match analysis, association football, fitness train-
ing
INTRODUCTION
he physiological stress imposed on outfield soc-
T cer players has been demonstrated to be role
and competitive-level dependent (2, 12, 19, 20–
22, 24).
Players belonging to first division teams have been
shown to cover more distance during a match compared
with their colleagues playing in lower divisions (12). Fur-
thermore, the lower space coverage observed in lower-di-
vision competitions have been associated with less dis-
tance spent at high intensity in the less skilled division
(3, 12). The above-mentioned findings have contributed to
the development of specific training drills and the design
of more game-related fitness tests (4).
Similar to players, referees are divided into categories
according to their skill level. Also much like players,
whose physical performance is competitive-level depen-
dent (3, 12), the exercise intensity imposed on officials
during actual match play is expected to vary according to
the level of competition officiated.
International soccer competitions and tournaments
are very competitive in nature as the participating teams
486
are the best of their nations and competitions are often
played on a knock-out basis (20). These highly competi-
tive soccer matches are officiated by international refer-
ees who have been physically tested and trained by Fed-
eration Internationale de Football Association (FIFA) and
Union Europenne de Football Association (UEFA) (11).
Given that, those referees that officiate international-lev-
el matches can be considered as an elite population of
referees selected from each FIFA-affiliated country’s best
officials.
Based on the above-mentioned considerations, the
present study was conducted with the aim of examining
the physical load imposed on international-level soccer
referees during official international-level competitions.
The work hypothesis was that according to what has pre-
viously been demonstrated for outfield soccer players (3,
12, 20), the competitive level of play may influence the
match activities of soccer referees.
The information gained regarding the activity profile
of international refereeing could then be used to develop
specific training drills or for the development of compe-
tition-dependent fitness standards.
METHODS
Experimental Approach to the Problem
A soccer match is controlled by a referee who has full
authority to enforce the laws of the game and is free to
move throughout the field using the most appropriate di-
rectional exercise modes in order to gain optimal posi-
tioning. The referee is assisted during the match by 2
assistant referees each moving on the touch-line in 1 of
the 2 halves of the field. A match lasts 2 equal periods of
45 minutes, and players are entitled to an interval at half
time not exceeding 15 minutes. Allowance for time lost is
made in either period at the discretion of the referee.
Given that soccer matches are controlled entirely by a
single referee (with minimal help from 2 assistant refer-
ees who control the sidelines), time motion analysis rep-
resents a good methodology to measure the match phys-
ical performance of elite soccer referees (8–10). Further-
more, referees (compared with soccer players) are not in-
volved in activities other than directional exercise modes,
and this promotes the convenience of recording distances
covered and times spent in different match categories as-
sessed by time-motion and match analysis (3, 8–10).
In the present study, in order to highlight the possible
peculiarities of the match performance of international-
level refereeing, 2 groups of elite-level referees were an-
alyzed. With this aim, the activity profile of a group con-
sisting of 13 international-level referees (UEFA) that of-
ficiated in Italy during competitive European Cup match-

TABLE 1. Physical characteristics of international- and na-
tional-level referees. There are no significant differences be-
tween groups for any of the physical characteristics.
International National
Characteristic (n 5 13) (n 5 13)
Age 38 6 3 37 6 3
Height 182.0 6 6.5 182.0 6 3.5
Body mass 78.8 6 7.0 77.1 6 6.5
Years of experience at the
higher competitive level .5 .5
TABLE 2. Country of affiliation of the international referee
group (IRG) group referees (n 5 13).
Country Frequency
Germany 3 ering the following categories: (a) standing; (b) walking
France 2 forward; (c) low-intensity run (less than 13 km·h21); (d)
Holland 3 medium-intensity run (from 13.1 to 18 km·h21); (e) high-
Scotland 2 intensity run (from 18.1 to 24 km·h21); (f) maximal speed
England 1 run (speed higher than 24 km·h21); (g) walking backward;
Sweden 1
Switzerland 1
Total 13
es were compared with a group of 13 Italian officials
belonging to Serie A. The Serie A is the highest level of
soccer played in Italy and is regarded by many as the best
domestic soccer competition in the world.
Subjects
The international-level referees group (IRG) consisted of
13 experienced elite-level referees enrolled in the UEFA
and thus eligible to officiate international competitions.
Each member of IRG had no less than 5 years’ experience
at the international level. All IRG referees were in good
health and able to pass the UEFA fitness standards (11).
Mean age, height, and weight of the referees are dis-
played in Table 1. Affiliation countries of the IRG group
referees are shown in Table 2.
The national-level referees group (NRG) consisted of
13 experienced referees enrolled in the Commisione Ar-
bitri Nazionali (CAN) and thus officiating in the Serie A
and B Italian championships. Each referee in the NRG
had no less than 15 years’ officiating experience, of which
5 years were spent at the first division level (Serie A and
B). All subjects were in good health and followed a su-
pervised physical training program aimed to develop aer-
obic fitness and short-sprint (10–50 m) and repeated
sprint ability performance. All NRG referees were in good
health and able to pass the UEFA fitness standards (11).
Written consent from all subjects was obtained before the
commencement of the study, following a brief but detailed
explanation as to the nature of the research. All of the
subjects were examined during first division Italian
championships (Serie A). Mean age, height, and weight
of the referees are displayed in Table 1.
In Italy no more than 36 referees are selected to offi-
ciate in the Serie A and B championships each year. Ev-
ery season those who exceed 45 years of age or are con-
sidered not to have the ability to cope with the demands
of the game are dismissed. Due to the rigorous selection
process and to the high level of play taking place in the
Serie A and B championships, the referees belonging to
ELITE REFEREEING AND MATCH ACTIVITIES PROFILE 487
the NRG may be considered representative of elite-level
officials.
Match Analysis
Match activities were monitored by means of a device
that used a technology similar to that reported by Ohashi
et al. (16, 17). Briefly, the referee was observed during
the whole match (90 minutes) by means of 2 cameras
(Play Controller, Phromos, Città di Castello, Perugia, It-
aly), with each attached to a tripod equipped with encod-
ers. The cameras were set approximately 10 m away from
1 of the sidelines of the football pitch. Signals of angles
obtained by the encoders were sequentially converted into
digital data and recorded on a floppy disc for postmatch
analysis.
Analysis of match activities was carried out consid-
(h) running backward; (i) running sideward; (j) high-in-
tensity activity (HIA, sum of the activities performed at
speed equal to or faster than 18.1 km·h21); and (k) unor-
thodox directional modes distance (backward running
plus sideward running, after Reilly and Bowen [18]).
Comparison between the 2 observed groups were also
made comparing average speeds (space covered/time
spent in each category, in meters per second) attained in
the above-mentioned match categories by each referee
(only for conventional running categories).
Validity and reliability of the method for measuring
the distances have been reported elsewhere (10).
Statistical Analyses
Data are presented as means and standard deviations of
the mean. Mean values for match activities were com-
pared using analysis of variance (ANOVA) for repeated
measures and factorial 2 3 2 ANOVA (group 3 half). Post
hoc analyses were performed using Tukey’s honestly sig-
nificant difference test. The alpha level of significance
was set at 5% (p # 0.05) a priori.
RESULTS
Entire Match
Covered distances, time spent standing still, and compar-
ison results between NRG and IRG are shown in Tables
3–5. Average speeds attained in the forward running cat-
egories are presented in Table 6.
First vs. Second Half
During the second half, NRG stood still longer (300 6 78
vs. 345 6 79 seconds, p , 0.05) and covered more distance
running at low intensity (2,523 6 259 vs. 2,617 6 303, p
, 0.05). Medium-intensity coverage decreased signifi-
cantly during the second half, switching from 1,829 6 194
to 1,684 6 169 m (p , 0.05). High-intensity run distance
did not show significant half to half variations (965 6 151
vs. 955 6 124 m for the first and second half, respectively;
p , 0.05). Also, maximal speed run and HIA distances
showed no significant changes switching from the first
half to the second half (206 6 140 vs. 253 6 138 m and
1,171 6 223 vs. 1,208 6 235, respectively; p . 0.05). No
half to half variations were also found in the walking
488 CASTAGNA, ABT, AND D’OTTAVIO

TABLE 3. Distances covered (meters) during an average match by the referees.†

Low-intensity run Medium-intensity run High-intensity run Max speed run High-intensity activity
Group (m) (m) (m) (m) (m)
NRG n 5 13 5148 6 549 3513 6 328 1920 6 258 458 6 249 2378 6 423
IRG n 5 13 4879 6 605 2562 6 471* 1277 6 398* 366 6 344 1642 6 689*
* Significantly different from NRG (p # 0.05).
† NRG 5 national referee group; IRG 5 international referee group.

TABLE 4. Distances covered (meters) and times spent standing still (seconds) during an average match.†
Standing still Walking forward Walking backward Backward run Sideward run
Group (s) (m) (m) (m) (m)
NRG (n 5 13) 645 6 143 800 6 142 47 6 20 896 6 519 172 6 217
IRG (n 5 13) 732 6 160 994 6 184* 42 6 23* 989 6 700 108 6 91
* Significantly different from NRG (p # 0.05).
† NRG 5 national referee group; IRG 5 international referee group.

TABLE 5. Match duration (seconds) and distance covered TABLE 7. Comparisons between halves (data revealed to be
(meters) during an entire match by the referees. significantly different between groups, p , 0.05).
Total distance Match duration International-
Group (m) (s) National-level level
referees referees
NRG (n 5 13) 12,956 6 548 5674 6 71
IRG (n 5 13) 11,218 6 1056* 5561 6 74* Walking forward 1 half 388 6 87 486 6 78
Medium-intensity run 1 half 1829 6 194 1310 6 215
* Significantly different from NRG (p # 0.05).
High-intensity run 1 half 965 6 151 648 6 188
† NRG 5 national referee group; IRG 5 international referee
High-intensity activity 1 half 1171 6 223 803 6 315
group.
Total distance 1 half 6549 6 363 5672 6 494
Walking forward 2 half 412 6 59 508 6 111
Medium-intensity run 2 half 1684 6 169 1252 6 271
backward category (25 6 21 m vs. 22 6 22 m; p . 0.05). High-intensity run 2 half 955 6 124 629 6 223
A significant decrease in the amount of distance covered High-intensity activity 2 half 1207 6 235 839 6 392
running backward was observed during the second half Total distance 2 half 6407 6 246 5546 6 581
as the coverage changed from 507 6 278 to 390 6 255 m Medium-intensity 1 half 4.23 6 0.02 4.20 6 0.02
(p , 0.05). Similarly, the referees belonging to this group speed
covered less distance running sideward during the second Medium-intensity 2 half 4.23 6 0.02 4.21 6 0.03
half (98 6 98 m vs. 74 6 72 m; p , 0.05). No variation speed
in total distance was observed in this group between
halves (6,548 6 363 and 6,407 6 246 m for the first and
second half, respectively; p . 0.05). First- and second-half played in Table 7. Only significant differences between
duration were the same (2,826 6 40 and 2,848 6 53 sec- the selected variables are shown.
onds, respectively; p . 0.05). Average speeds attained in
the running forward categories did not show any change DISCUSSION
between halves (p . 0.05). The main finding of this study is that international-level
IRG stood still the same amount of time across halves refereeing is less demanding than national-level referee-
(351 6 84 vs. 379 6 100 seconds; p . 0.05). Running at ing. This was unexpected as international-level soccer
maximal speed, the IRG covered significantly more dis- matches are reported to be very demanding from the
tance during the second half (156 6 155 vs. 210 6 189 physical point of view (20). However, the distances cov-
for the first and second half, respectively; p , 0.05). No ered by the international-level group are similar to that
significant differences across halves were observed in this reported for elite-level soccer referees who were not sub-
group of highly competitive referees in the other match mitted to specific high-volume training (10). This may
activity categories considered in this study. Also, average corroborate the effect of a well-designed and supervised
speeds attained in the running forward categories did not training program in promoting match intensity. However,
show any change between halves (p . 0.05). from this study design it is not possible to make precise
Comparison between the 2 referees’ groups are dis- inferences in this regard.

TABLE 6. Average speeds (meters per second) attained during an average match by the referees in the running categories.†
High-intensity
Group Low-intensity run Medium-intensity run High-intensity run Max speed run activity
NRG (n 5 13) 2.36 6 0.05 4.23 6 0.01 5.62 6 0.04 7.41 6 0.22 5.89 6 0.16
IRG (n 5 13) 2.29 6 0.09* 4.20 6 0.02* 5.61 6 0.05 7.40 6 0.33 5.90 6 0.22
* Significantly different from NRG (p # 0.05).
† NRG 5 national referee group; IRG 5 international referee group.

The results of the present study have revealed that
international-level soccer referees covered 11.218 6 1.056
km over the course of a 90-minute match, which is very
close to that reported by D’Ottavio and Castagna (10) for
elite-level soccer officials. However, national-level soccer
referees covered significantly more distance during the
entire match than their international-level counterparts
(12.956 6 0.548 vs. 11.218 6 1.056 km; p , 0.05). Al-
though Serie A matches lasted longer than international
matches (p . 0.05), this cannot solely explain the greater
match coverage observed in national-level soccer referees.
Differences in match duration have been caused by a
more generous allowance of compensation for time lost by
the Serie A referees. The results of the present study con-
flict with those reported previously by Asami et al. (1)
comparing elite Japanese and international soccer refer-
ees officiating in domestic and international tournament
matches, respectively. They found no significant differ-
ence in the amount of distance covered during the match
between the 2 groups of elite-level officials (1). However,
the Japanese scientists made use of hand notational anal-
ysis and employed only 4 movement categories.
Available soccer refereeing–related literature reports
a wide range of match coverage examples. The first pub-
lished report available in the scientific literature (1) dis-
closed an average coverage of approximately 10 km in
elite soccer players. Harley et al. (13) reported a 7.5 km
coverage for county-level English soccer referees. Analyz-
ing the match coverage of premier league soccer referees,
Catteral et al. (8) found a match coverage of 9.5 km. Sim-
ilar average coverage (9,408 6 838 m) was reported by
Johnston and McNaughton (14) for Australian elite soccer
referees. More recently, Krustrup and Bangsbo (15) and
D’Ottavio and Castagna (10) reported distances ranging
from 10 to 11.5 km per match for Italian elite-level and
Danish top-level soccer referees, respectively.
Although the reported data are difficult to compare as
being obtained with different technologies, it seems that
elite-level referees cover no less than 9.5 km during a
competitive match. Coverage differences may be due to
different refereeing strategies, team aggressiveness, and
the referee’s fitness level. Refereeing strategies regarded
as distance from play have proved to be inversely related
to judgment mistakes (23). Verheijen et al. (23) suggested
judgments should be made when walking and pointed to
20–25 m as the optimal distance from infringements.
As Harley et al. (13) reported that match total cover-
age would probably promote better positioning, the pre-
sent study suggests that international-level referees were
able to officiate the games while being more distant from
the performing action than their national-level counter-
parts. Verheijen et al. (23) reported that referees could
have an advantage in being not that close to the perform-
ing action in order to detect infringements during the
game. However, this point of view was not shared by the
Italian Refereeing Association when this study was per-
formed. Aside from this possible refereeing philosophy
disparity, other variables could have determined match
coverage, 1 of them being the fitness level and training
status of the 2 populations. Recently Castagna et al. (5–
7) and Krustrup and Bangsbo (15) have shown that aer-
obic fitness assessed either directly or indirectly has a
strong relationship with the total match coverage and
with key match activities performed by the elite referee
during competition. Furthermore, Krustrup and Bangsbo
ELITE REFEREEING AND MATCH ACTIVITIES PROFILE 489
(15) reported that properly designed and planned inter-
mittent endurance training impacts favorably with match
physical performance of top-level soccer referees.
In this regard the national-level referees in the pre-
sent study were submitted to supervised specific training
similar to that suggested by Krustrup and Bangsbo (15),
whereas the international-level referees were not in-
volved in specific high-intensity interval training. These
training differences may have determined the higher
amount of distance covered during the match and partic-
ularly in the HIA category (15) for the national-level ref-
erees.
Differences in match coverage between international
and national referees may be explained by lower distanc-
es covered at speed comprised in the 13–18 km·h21 range
by the IRG. However, international-level referees showed
significantly less distance coverage in HIA. In fact, in this
match category national-level referees covered approxi-
mately 31% more distance than their international-level
counterparts. These results seem to show that compared
with national-level officials, the international-level ref-
erees tended to cover less distance during the match and
to limit their activity performed at high intensity. Anal-
ysis of average speeds attained in match categories re-
vealed significantly slower running speeds in the low- and
medium-intensity categories in the international-level
referees, but no differences were detected among those
activities performed at speeds above 18 km·h21.
Intragroup analysis has revealed that national-level
referees stood still longer during the second half of the
game, but nevertheless they were able to total the same
amount of distance covered as during the first period of
play. Although totalling less distance, international-level
soccer referees spent the same amount of time standing
still across halves, and as for the national-level counter-
part they did not show total coverage decrements during
the second half. The observed constancy of total coverage
observed in these 2 populations of referees is somehow
unexpected as previous authors have reported a signifi-
cant second-half decrement in match coverage in elite-
level soccer referees (11). This finding also contrasts with
what has been previously reported for elite-level soccer
players (2, 19, 22). Recently, Krustrup and Bangsbo (15)
reported in top-level referees a significant decrement in
both total coverage and HIA during the second half. Prob-
ably the high skill level and/or training status possessed
by both populations of officials in the present study has
exacerbated the documented sparing behavior previously
reported by the available international literature (10). In
fact, differing from previous studies, both groups of ref-
erees were able to maintain space coverage across halves
without negatively affecting those activities performed at
high intensity (15). In this regard international-level soc-
cer referees not only conserved the global amount of dis-
tance but were even able to increase the performance in
the maximal speed category. Such a behavior detected in
this selected population of referees may be considered as
a sign of a specific ability acquired by elite officials
through years of experience at high competitive levels
and probably supported by a well-developed training sta-
tus.
Practical Applications
All of the referees who were observed for match analysis
in this study were able to pass the UEFA fitness test
490 CASTAGNA, ABT, AND D’OTTAVIO
standards. UEFA tests consist of 2 3 50 m sprint, 2 3
200 m sprint, and a 12-minute time-trial run (11). Re-
cently it has been reported that the 12-minute run test
for distance positively correlates with key match activi-
ties performed by the elite referee during highly compet-
itive matches (6). In light of the findings of the present
study and of the training status of the referees partici-
pating in the present study, it seems that international
soccer referees should be submitted to a more specific
training regimen. Particular attention should be paid to
intermittent high-intensity training (15). Given the con-
servation of match categories coverage across halves ob-
served in the present study, it may be of great relevance
in the development of prospective soccer referees to pur-
sue the achievement of this unique tactical pattern. This
could be obtained both by promoting fitness and devel-
oping appropriate refereeing strategies that allow the ref-
eree to maintain physical performance in the second half.
In this regard a match coverage of 11.5 km and conser-
vation of the total distance and HIA during the second
half may be considered as the attainment of superior
skills by the referee.
REFERENCES
1. ASAMI, T., H. TOGARI, AND J. OHASHI. Analysis of movement
patterns of referees during soccer matches. In: Science and
Football. T. Reilly, A. Lees, K. Davids, and W.J. Murphy, ed.
London: E. & F.N. Spon, 1988. pp. 341–345.
2. BANGSBO, J., L. NøRREGAARD, AND F. THORSøE. Activity profile
of competition soccer. Can. J. Sport Sci. 16:110–116. 1991.
3. BANGSBO, J. The physiology of soccer with special reference to
intense intermittent exercise. Doctoral thesis, Krogh Institute
University of Copenhagen, 1994.
4. BANGSBO, J. Fitness Training in Football: A Scientific Ap-
proach. Copenhagen, Denmark: HO1Storm, 1993.
5. CASTAGNA, C., AND S. D’OTTAVIO. Effect Of maximal aerobic
power on match performance in elite soccer referees. J.
Strength Cond. Res. 15:420–425. 2001.
6. CASTAGNA, C., G. ABT, AND S. D’OTTAVIO. The relationship be-
tween fitness tests and match performance in elite Italian soc-
cer referees. J. Strength Cond. Res. 16:231–235. 2002.
7. CASTAGNA, C., G. ABT, AND S. D’OTTAVIO. The relationship be-
tween selected blood lactate thresholds and match performance
in elite soccer referees. J. Strength Cond. Res. 16:623–627.
2002.
8. CATTERALL, C., T. REILLY, G. ATKINSON, AND A. COLDWELLS.
Analysis of work rate and heart rates of association football
referees. Br. J. Sports Med. 27:153–156. 1993.
9. D’OTTAVIO, S., AND C. CASTAGNA. Physiological load imposed
on elite soccer referees during actual match play. J. Sports
Med. Phys. Fitness. 41:27–32. 2001.
10. D’OTTAVIO, S., AND C. CASTAGNA. Analysis of match activities
in elite soccer referees during actual match play. J. Strength
Cond. Res. 15:167–171. 2001.
11. EISSMANN, H.J., AND M. D’HOOGHE. Sports medical examina-
tions. In: The 23rd Man: Sports Medical Advice for Football
Referees. H.J. Eissmann, ed. Leipzig, Germany: FIFA, Gersöne-
Druck, 1996. pp. 7–19.
12. EKBLOM, B. Applied physiology of soccer. Sports Med. 3:50–60.
1986.
13. HARLEY, R.A., K. TOZER, AND J. DOUST. An analysis of move-
ment patterns and physiological strain in relation to optimal
positioning of association football referees. J. Sports Sci. 17:
813. 1999.
14. JOHNSTON, L., AND L. MCNAUGHTON. The physiological re-
quirements of soccer refereeing. Aust. J. Sci. Med. Sport. 26:
67–72. 1994.
15. KRUSTRUP, P., AND J. BANGSBO. Physiological demands of top-
class soccer refereeing in relation to physical capacity: Effect
of intense intermittent exercise training. J. Sports Sci. 19:881–
891. 2001.
16. OHASHI, J., H. TOGARI, M. ISOKAWA, AND S. SUZUKI. Measuring
movement speeds and distances covered during soccer match
play In: Science and Football. T. Reilly, A. Lees, K. Davids, and
W.J. Murphy, eds. London: E. & F.N. Spon, 1988. pp. 329–333.
17. OHASHI, J., O. MIYAGI, H. NAGAHAMA, T. OGUSHI AND K.
OHASHI. Application of an analysis system evaluating inter-
mittent activity during a soccer match. In: Science and Football
IV. W. Spinks, T. Reilly, and A. Murphy, eds. London: Routled-
ge, 2002. pp. 132–136.
18. REILLY, T., AND T. BOWEN. Exertional costs of changes in di-
rectional modes of running. Percept. Motor Skills. 58:149–150.
1984.
19. REILLY, T. Motion analysis and physiological demands. In: Sci-
ence and Soccer. T. Reilly, ed. London: E. & F.N. Spon, 1996.
pp. 65–81.
20. RIENZI, E., B. DRUST, T. REILLY, J.E.L. CARTER, AND A. MAR-
TIN. Investigation of the anthropometric and work-rate profiles
of elite South American international soccer players. J. Sports
Med. Phys. Fitness. 40:162–169. 2000.
21. THOMAS, V., AND T. REILLY. Application of motion analysis to
assess performance in competitive football. Ergonomics. 19:
530–535. 1976.
22. VAN GOOL, D., D. VAN GERVEN, AND J. BOUTMANS. The phys-
iological load imposed on soccer players during real matchplay.
In: Science and Football. T. Reilly, A. Lees, K. Davids, and W.J.
Murphy, eds. London: E. & F.N. Spon, 1988. pp. 51–59.
23. VERHEIJEN, R., R. OUDEJANS, P.J. BEEK, AND F.C. BAKKER.
Factors affecting decision-making of soccer referees. In: Fourth
World Congress of Science and Football. W. Spinks, eds. Syd-
ney: University of Technology Sydney, 1999. p. 28.
24. WITHERS, R.T., Z. MARICIC, S. WASILEWSKI, AND L. KELLY.
Match analysis of Australian professional soccer players. J.
Hum. Mov. Stud. 8:159–161. 1982.
Address correspondence to Prof. Carlo Castagna,
carlocastagna@teknosport.com.