Faculty of Sport, University of Ljubljana, ISSN 1318-2269  51Kinesiologia Slovenica, 12, 2, 51–58 (2006)
Primož Pori*
Marko Šibila
ANALYSIS OF HIGH-INTENSITY LARGE-
SCALE MOVEMENTS IN TEAM HANDBALL
ANALIZA NAJINTENZIVNEJŠIH 
GIBANJ V ROKOMETU
Abstract
The intensity of large-scale movements in team
handball can be classified in four categories of velocity.
The primary aim of the article is to analyse the 4th
velocity class (large-scale movements above 5.2 m/s) 
during a handball match, with a particular focus on 
the frequency and distances of such events. The sample
group consisted of 36 male team handball players of 
different playing positions. The output data on large-
scale movements obtained from the SAGIT programme 
were processed using selected descriptive statistical 
methods. The differences between the mean values of
the results of the chosen loading variables were tested 
for the first and second halves of the match with the
use of a t-test method for pairs. During the analysed 
matches, players from the chosen sample on average 
covered 620 m of movements with a velocity above 5.2 
m/s. An analysis of the frequency of movements with the 
highest intensity showed that players from the sample 
averaged 80 events of the 4th VC per analysed match. 
The average distance of these movements (sprints) was
7.7 m. The results of the research show that players from
the sample were unable to maintain the same level of 
loading throughout the entire match. Reasons for the 
smaller number of highest intensity cyclic movements 
in the second half can mostly be found in the level of 
physical preparation of the players included in the 
sample, as well as in the tactical plan of the teams trying 
to maintain the achieved result.
Key words: team handball, loading, velocity classes, 
match analysis, computer vision
Faculty of Sport, University of Ljubljana, Slovenia
*Corresponding author:
Faculty of Sport, University of Ljubljana
Gortanova 22, 1000 Ljubljana, Slovenia
Tel.: +386 1 5207757
Fax: +386 1 5207730
E-mail: primoz.pori@fsp.uni-lj-si
Izvleček
V rokometu lahko intenzivnost cik ličnih gibanj 
opredelimo s štirimi hitrostnimi razredi. Hitrostni 
razred (HR) najvišje intenzivnosti je 4 HR, za katerega 
so značilna gibanja igralcev nad 5,2 m/s. Osnovni namen 
prispevka je bila podrobna analiza 4 HR med rokometno 
tekmo. Zanimala nas je pogostost pojavljanja ter dolžina 
tovrstnih gibanj, poleg tega pa smo želeli dobiti podatke 
tudi o količini najintenzivnejših gibanj v desetminutnih 
odsekih tekme. Vzorec merjencev je sestavljalo 36 
rokometašev na različnih igralnih mestih. Podatke smo 
zbirali na šestih modelnih tekmah s pomočjo sistema 
SAGIT in jih kasneje obdelali s osnovno opisno statistiko. 
Razlike med aritmetičnimi sredinami rezultatov 
spremenljivk obremenitev v prvem in drugem polčasu 
tekem smo testirali z t-testom parov. Izbrani igralci so v 
povprečju opravili 620 m gibanj v 4 HR. Na analiziranih 
tekmah smo zabeležili v povprečju 80 gibanj nad 5.2 
m/s, ki so bila v povprečju dolga 7.7 m. Rezultati analize 
so pokazali, da igralci niso bili sposobni ohranjati isto 
raven obremenitev skozi celotno tekmo. Razloge upada 
najintenzivnejših cikličnih gibanj v drugem polčasu smo 
iskali predvsem v ravni kondicijske priprave igralcev 
našega vzorca, kot tudi rezultatskemu taktiziranju ekip 
v drugem polčasu.
Ključne besede: rokomet, obremenitve, hitrostni raz-
redi, analiza tekem, računalniški vid
52 Analysis of movements in handball Kinesiologia Slovenica, 12, 2, 51–58 (2006) 
INTRODUCTION
The study of the intensity and loading of handball players during matches and in training is
currently very popular. Researches mainly examine the structure and dynamics of loading 
during matches. In general, researchers are interested in the types of loads which an athlete 
endures during the match, the frequency of their occurrence, as well as the relations between 
them (Al-Lail, 2000; Bon, 2001; Brandon, 1998; Pori, Kovačič, Bon, Dolenec, & Šibila, 2005; 
Šibila, Vuleta, & Pori, 2004). In handball, the intensity and volume of loads are very diverse. 
Alongside cyclical movements, which include different types of walking and running, acyclic
activities also occur during a match, such as passes, shots and falls. During a handball game 
the loading occurs in intervals, which are a result of changes in the dynamics and types of 
loading. Cyclical movements are fundamental as they allow a player to move within the court 
in two dimensions (length and width). They include walking and running without a ball as
well as dribbling the ball while walking or running. 
The existing literature contains quite a lot of contributions concerning the volume of handball
players’ cyclical movements during matches. Authors who have used different methods of col-
lecting, sorting and analysing the related data state that during a match handball players travel 
a distance of between 4,000 and 6,000 m (Cardinale, 2000; Kotzamanidis, Chatzikotoluas, 
& Giannakos, 1999; Martin, 1990). The variability in the volume of distances run or walked
is caused by various factors such as the playing position (Cuesta, 1991), age category of the 
players (Pori et al., 2005), level or importance of the match (Cardinale, 2000) etc. It is difficult
to deduce the dynamics of loads during matches merely from the data about the volume of 
cyclical movements. As velocity is one of the main indicators of load intensity, which during 
the handball game manifests itself in different forms, the research of Bon (2001) also evaluated
the distances walked or run throughout a match according to the intensity of loading. The
intensity of cyclical movements has been described as involving four velocity classes (VC’s) 
(1st VC –velocity of up to 1.4 m/s; 2nd VC – velocity between 1.4 and 3.4 m/s; 3rd VC – velocity 
between 3.4 and 5.2 m/s; 4th – velocity above 5.2 m/s). Bon (2001) stated in the same research 
that players from the chosen sample spent on average at the analysed matches 7% of the playing 
time sprinting, 25% running fast, 31% running slowly and 37% of the playing time walking 
or standing still. Similar results were presented in the 2001 research by Pori on a sample of 
84 handball players from youth, junior and senior categories. The weakness of these analyses,
which were both carried out using the SAGIT system (Perš, Bon, Kovačič, Šibila, & Dežman, 
2002), lies in the fact that all the intensities of cyclical movements could only be presented in 
percentage shares of the time the players spent in a specific velocity class. For the purposes
of the research carried out by Pori in 2003, the module for showing the results in the SAGIT 
system was upgraded with a method that also allowed the presentation of data on the intensity 
of cyclical movements as the average volume of the distances which the players spent in a 
specific velocity class. Pori argued that the wing players from the chosen sample, whilst playing
zone defence 6_0 in the position of the first defence player, executed on average 1620 m, SD
= 100 m of movements in the 1st VC, 1450 m, SD = 80 m in the 2nd VC, 1170 m, SD = 80 m in 
the 3rd VC and 645 m, SD = 50 m in the 4th VC.
The present work aims to analyse in detail cyclical movements of the highest intensity in a
game of handball. The category with the highest intensity is the 4th velocity class (4VC), when 
Analysis of movements in handball 53Kinesiologia Slovenica, 12, 2, 51–58 (2006)
the players’ movements exceed the velocity of 5.2 m/s (Bon, 2001). The primary aim of the
article is to analyse the 4th VC during a handball match, with the frequency and distance of 
such events being of particular interest. In addition, data on the quantity of highest-intensity 
movements in 10-minute intervals of a match were studied.
METHOD
Participants
The sample group consisted of 36 male team handball players (average body height 189 cm,
SD = 2.4 cm; average body mass 86.2 kg, SD = 4.4 kg; average age 21.8 years, SD = 3.1 years) 
of different playing positions. Specifically, 12 were wing players, 18 back players and 6 pivots,
who at the time of the measurements were included in senior teams and had played in the first
or second Slovenian national handball league. Data on the work rate of large-scale movements 
were collected from six model matches and one team was monitored at each of the matches. 
In all of the matches certain environmental conditions were standardised: i) the playing 
time for all games was two times 30 minutes; ii) all the teams played a 6_0 zone defence; iii) 
selected players had to play the entire game; and iv) and a one-minute team time-out was not 
allowed.
Instrument
The sample of variables included those related to large-scale cyclical movements by defini-
tion. 
Movements in the 4th VC (velocity above 5.2 m/s) were presented as distances the players 
travelled during an entire match, in each half and in 10-minute intervals of the match. In 
addition, data on the frequency of occurrences and the length of movements in the 4th VC 
were obtained.
Data collection method
The collection of data on players’ large-scale cyclical movements in handball matches was
based on a computer-aided automatic tracking method using the SAGIT (Ljubljana, Slovenia) 
system (Perš, et al., 2002). In the field of biomechanical motion an analysis the accuracy of
the motion acquisition system is one of the most important factors when such systems are 
compared, and the accuracy is measured in millimetres (Richards, 1999). The SAGIT system
has some errors, which have been analysed in several stages (Perš, 2004). 
Data analysis
The output data on large-scale cyclical movements obtained from the SAGIT programme
were processed using selected methods of descriptive statistics. The differences between the
arithmetical mean values of the results of the chosen loading variables were tested for the first
and second halves of a match with the use of a t-test method for pairs. Statistically significant
differences were accepted with a 5% statistical significance threshold (two-tailed testing).
54 Analysis of movements in handball Kinesiologia Slovenica, 12, 2, 51–58 (2006) 
RESULTS
Figure 1 shows the average volume of distances run within the fourth velocity class. Players 
from the sample executed on average during the analysed matches 620 m, SD = 70 m of 
movements with a velocity above 5.2 m/s (above 18.7 km/h). The average distances are greater
in the first half of the matches. The average volume of movements during the first half was
350 m, SD = 58 m, compared to 265 m, SD = 41 m during the second half. The t-test results
of pairs confirmed statistically significant differences (p < 0.001). The average decrease in the 
volume of movements in the 4th VC from the first to the second half was 85 m.
Figure 1: Average volume of large-scale movements in the 4th velocity class (VC) 
Legend: 1H – 1st half; 2H - 2nd half; EG – entire game; m – metres; * - statistically significant difference
An analysis of the volume of movements in the 4th VC during 10-minute intervals revealed 
different average distances in particular periods of a match (see Figure 2). During the first ten
minutes (145 m, SD = 35 m) and in the second ten minutes (119 m, SD = 23 m) of a match, the 
distances in the 4th VC were greater than during the last part of the match. In the last analysed 
interval of the match (from the 50th minute until the end of the match) the average volume 
of movements in the 4th VC (78 m, SD = 18 m) nearly halved relative to the first 10-minute
interval of a match.
Figure 2: Average volume of large-scale movements in the 4th VC for 10-minute intervals of 
a match
Legend: min – minutes; m – metres; 0-10 - (interval from the start to the 10th minute of a match)…, 50-60 – (inter-
val from the 50th to the 60th minute of a match)
Analysis of movements in handball 55Kinesiologia Slovenica, 12, 2, 51–58 (2006)
Figure 3 shows the results of the frequency of movements with the highest intensity for players 
from the sample. In the analysed matches, on average 80, SD = 14 of such events were noted. 
Like before, the frequency of movements was higher during the first half of a match (42, SD
= 8), compared to the second half (38, SD = 6). This difference was statistically significant (p 
< 0.002).
Figure 3: Average frequency of large-scale movements in the 4th VC 
Legend: 1H – 1st half; 2H - 2nd half; EG – entire game; f – frequency of movements in the 4th VC; * -statistically 
significant difference
The average distance of movements (sprints) with a velocity above 5.2 m/s was 7.7 m, SD = 2.3
m (see Figure 4). During the first half of a match, the average distance of sprints by players
from the sample was 8.1 m, SD = 2 m and was significantly higher (p < 0.001) than during the
second half (6.9, SD = 2.4). The difference in the average distance of sprints from the first to 
second half was 1.2 m.
Figure 4: Average distance of large-scale movements in the 4th VC 
Legend: 1H – 1st half; 2H - 2nd half; EG – entire game; m – metres; * - statistically significant difference
DISCUSSION
During a handball match players are exposed to cyclical loads of both high and low intensity. 
The share of phases with high and low intensity varies from one match to another (Šibila,Vuleta,
& Pori, 2004). The present work analyses high-intensity cyclical movements in handball; they
56 Analysis of movements in handball Kinesiologia Slovenica, 12, 2, 51–58 (2006) 
are described by the movements of players with a velocity of above 5.2 m/s. These movements
are placed in the fourth velocity class (VC) (Bon, 2001).
The players from the sample ran, during the analysed matches, in the 4th VC on average 620 ± 
70 m. These results are on average slightly lower than those found by Pori in 2003. The wing
players included in that research ran during a zone defence 6_0 on average 645 m, SD = 50 
m, and 558 m, SD = 114 m during the 3_2_1 zone defence. However, the research from 2003 
only included wing players in the sample of measured subjects, whereas the sample in the 
present research also includes outside players and pivots. Namely, studies show that the specific
nature of the wing players’ model of the game (i.e. starting playing position in attack) results 
in a larger total volume of movements during matches compared to other playing positions 
(Cuesta, 1991; Pori, 2001; Šibila, Vuleta, & Pori, 2004).
Also analysed was the average frequency of occurrences of movements with the highest inten-
sity, as well as their average distance. The results show that players from the sample reached
the 4th VC on average 80 times, SD = 14 times, with an average distance of sprints during the 
entire match at 7.7 m, SD = 2.3 m. Similar results can also be found in the study by Martin 
from 1990. Considering that a handball match lasts 60 minutes, this consequently means that 
the analysed players executed a movement of the highest intensity on average every 45 s, with 
the average distance of such movement being between 7 and 8 m.
High loadings during specific periods of a match also require large effort and result in a 
decrease of loading in the second half compared to the first half of a match (Cuotts, Reaburn,
& Abt, 2003; Deutch, Maw, Jenkins, & Reaburn, 1998; Krunstrup, Mohr, & Bangsbo, 2002). The
authors of several researches in the fields of rugby and football found that fatigue manifests
itself in a decrease of both the volume and intensity of cyclical movements in the second half. 
Whilst studying the movements of side referees, Krustrup et al., (2002) found that movements 
of the highest intensity in second half decreased by 33% and the overall volume of movements 
by 30%. He also presumed that the decrease of intensity of a player’s game on the field leads
to a decrease in the abovementioned loading variables.
Similar to previous studies, the present research found a statistically significant decrease in
loading in the second half of a match for all chosen variables of large-scale movements of the 
highest intensity. An analysis of 10-minute intervals of a match allowed a closer look at the 
dynamics of the loading intensity for specific periods of the match. The volume of large-scale
movements in the 4th VC was decreasing from the beginning to the end of the match, with 
the exception of the first ten minutes of the second half. When comparing the first and last
ten minutes of a match, it can be seen that movements of highest intensity fell by nearly a 
half, specifically by 67 m. As was previously estimated, the average distance of sprints was 8
m, meaning that players from the sample executed approximately 8 sprints less in the last ten 
minutes of a match than in the first ten minutes.
The results of the research show that players from the sample were unable to maintain the
same level of loading throughout an entire match. The reasons for this can mainly be found
in the level of physical preparation of the players from the chosen sample; in the opinion of 
the authors it was insufficient to allow for such a model of a game which required playing
throughout the entire match at a higher level. However, it needs to be considered that the 
experiment did not allow any substitutions to be made during a match. It rarely happens in a 
modern game of handball that one player plays throughout the entire match. Five years ago 
Analysis of movements in handball 57Kinesiologia Slovenica, 12, 2, 51–58 (2006)
the European Handball Federation (EHF) decided to enlarge handball teams, which can now 
include seven players on the field and seven substitutions.
Clubs and coaches quickly adapted to the changed rules and now more often include in a team
two players of high quality for each playing position. In addition to positive competitiveness 
among the players in a specific playing position, this change of rules (allowing more substitu-
tions) also has a positive impact on maintaining the high competition pace in all playing 
positions and phases of the handball game.
Conclusion
The ongoing development of the game of handball, resulting from a modern game model, fa-
cilitates the need to improve and optimise training. The modern approach to handball training
should be based on training procedures, methods and forms which closely imitate the loading 
and effort of players during a match. The results of the research show that players from the
sample performed a movement of the highest intensity on average every 45 s, with the distance 
of such movement being between 7 and 8 m. These movements represented approximately 10%
of the total distance covered during a match. The high loads seen in individual intervals of a
match, which required great effort, also resulted in smaller loads in the second half. Reasons
for the smaller number of highest intensity cyclical movements in the second half can mostly 
be found in the level of physical preparation of the players included in the sample, as well as 
in the tactical plan of the teams trying to maintain the achieved result.
REFERENCES
Al-Lail, A. (2000). A motion analysis of the work-rate and heart rate of the elite Kuwaiti handball players. 
Retrieved April 18, 2004, from http://www.sportscoach-sci.com
Bon, M. (2001). Kvantificirano vrednotenje obremenitev in spremljanje srčne frekvence igralcev rokometa 
med tekmo [Quantified evaluation of loading and monitoring of heart rate of handball players in a
match]. Unpublished doctoral dissertation, University of Ljubljana, Slovenia. 
Brandon, R. (1998). Nogomet – kakšne so energijske potrebe tega športa maksimalnih obremenitev s 
prekinitvami [Football – what are the energetic needs of this sport with maximum loading with breaks]. 
Vrhunski dosežek, 1, 7-9.
Cardinale, M. (2000). Handball performance: Physiological considerations and practical approach for the 
training metabolic aspects. Retrieved April 17, 2003, from http://www.sportscoach-sci.com
Coutts, A., Reaburn, P. & Abt, G. (2003). Heart rate, blood lactate concentration and estimated energy 
expenditure in a semi-professional rugby league team during a match: A case study. Journal of Sport 
Sciences, 21(3), 97-103.
Cuesta, G. (1991). Balonmano [Handball]. Madrid: Spanish Handball Federation.
Deutsch, M. U., Maw, G. J., Jenkins, D., & Reaburn, P. (1998). Heart rate, blood lactate and kinematical 
data of elite colts (under –19) rugby union players during competition. Journal of Sports Sciences, 16 
(2), 516-570.
Kotzamanidis, C., Chatzikotoluas, K., & Giannakos, A. (1999). Optimisation of the training plan of the 
handball game. Handball, 12(1), 64-71.
58 Analysis of movements in handball Kinesiologia Slovenica, 12, 2, 51–58 (2006) 
Krunstrup, P., Mohr, M., & Bangsbo, J. (2002). Activity profile and physiological demands of top-class
soccer refereeing in relation to training status. Journal of Sport Sciences, 20, 861-871.
Martin, D. (1990). Kursbusch 2: Trainingslehre [Manual 2: Trainings book]. Wiesbaden: Sportwiesen-
schaft.
Perš, J. (2004). Analiza človeškega gibanja na različnih nivojih podrobnosti z metodami računalniškega 
vida [Computer vision-based analysis of human motion at different level of detail]. Unpublished doc-
toral dissertation, University of Ljubljana, Slovenia.
Perš, J., Bon, M., & Kovačič S., (2001). Errors and mistakes in automated player tracking. In B. Likar 
(Ed.), 6th Computer vision winter workshop: Computer vision (pp. 25-36), Bled, Slovenia: Slovenian 
Pattern Recognition Society.
Perš, J., Bon, M., Kovačič, S., Šibila, M., & Dežman, B. (2002). Observations and analysis of large-scale 
human motion. Human Movement Science, 21, 295-311.
Pori, P. (2001). Analiza cikličnih obremenitev med rokometno tekmo pri igralcih, ki igrajo na različnih 
igralnih mestih v napadu [Analysis of large-scale cyclical movements of team handball players in differ-
ent playing positions in attack]. Unpublished master’s thesis, University of Ljubljana, Slovenia.
Pori, P. (2003). Analiza obremenitev in napora krilnih igralcev v rokometu napadu [Analysis of the 
loading and effort of wing players in team handball]. Unpublished doctoral dissertation, University of
Ljubljana, Slovenia.
Pori, P., Kovačič, S., Bon, M., Dolenec, M., & Šibila, M. (2005). Various age category-related differences
in the volume and intensity of large-scale cyclic movements of male players in team handball. Acta 
Universitatis Palckianae Olomucensis, Gymnica, 45(2), 199-126. 
Richards, J. G. (1999). The measurement of the human motion: A comparison of commercially available
systems. Human Movement Science, 18(5), 589-602.
Šibila, M., Vuleta, D., & Pori, P. (2004). Position-related differences in volume and intensity of large-
scale cyclic movements of male players in handball. Kinesiology, 36(1), 58-68.