74 Faculty of Sport, University of Ljubljana, ISSN 1318-2269 Kinesiologia Slovenica, 9, 2, 74–85 (2003) 
Goran Vučković
1
*
 
COMPARATIVE MOVEMENT
Branko Dežman
1
 ANALYSIS OF WINNING AND LOSING
Franc Erčulj
1
 PLAYERS IN MEN’S ELITE SQUASH
Stanislav Kovačič
2
 
Janez Perš
2
  
PRIMERJALNA ANALIZA GIBANJA 
ZMAGOVALCEV IN PORAŽENCEV 
PRI VRHUNSKIH IGRALCIH 
SQUASHA
Abstract
The aim of this study was to identify statistically 
significant differences between the winners and 
losers of a game in terms of the distance covered 
and the velocity of movements in a squash game. 
For this purpose we used a sample of 8 elite squash 
players and video-recorded their movements in 6 
matches. The video-recordings were digitized and 
processed by the SAGIT/SQUASH tracking system. 
The one-way analysis of variance for independent 
samples was used to establish the differences be-
tween the winners and the losers of games in terms 
of velocity of movement and distance covered; Pear-
son’s correlation coefficient was used to determine 
the correlation between time of game, number of 
scored points and distance covered. Total distance 
covered in a game ranged from 254 m to 1.449 m. 
A statistically significant correlation exists between 
distance covered and time of game and number of 
points scored in a game. The differences between 
the winners and the losers of games in terms of aver-
age velocity and distance covered were statistically 
insignificant.
Key words: squash, movement analysis, differences, 
distance covered, velocity of movement
1
 Faculty of Sport, University of Ljubljana, Slovenia
2
  Faculty of Electrical Engineering, University of 
Ljubljana, Slovenia 
* Corresponding author: 
Faculty of Sport, University of Ljubljana
Gortanova 22, SI-1000 Ljubljana, Slovenia
Tel.: +386 1 5207766
Fax: +386 1 5207730
E-mail: goran.vuckovic@sp.uni-lj.si
Izvleček
Namen tega dela je bil izmeriti pot in hitrost gibanja 
v nizu na squash tekmah in ugotoviti, ali obstajajo 
statistično značilne razlike med zmagovalci in po-
raženci posameznih nizov v omenjenih kazalcih. 
V ta namen smo na vzorcu 8 vrhunskih igralcev 
squasha s S-VHS kamero posneli njihova gibanja 
na 6 tekmah. Posnetke iz te kamere smo nato pre-
nesli v digitalno obliko in jih obdelali s sledilnim 
sistemom SAGIT/SQUASH. Z enosmerno analizo 
variance za neodvisne vzorce smo testirali razlike 
med zmagovalci in poraženci nizov v hitrosti in poti 
gibanja, s Pearsonovim koeficientom korelacije pa 
smo ugotavljali povezanost med časom trajanja, 
številom doseženih točk in potjo v nizu. Celotna 
pot gibanja igralcev v posameznem nizu se je gibala 
med 254 m in 1449 metri. Pot je statistično značil-
no povezana s časom trajanja posameznega niza in 
številom doseženih točk igralcev v nizu. Razlike v 
povprečni hitrosti in poti gibanja med zmagovalci in 
poraženci nizov niso bile statistično značilne. 
Ključne besede: squash, analiza gibanja, razlike, 
pot gibanja, hitrost gibanja

Movement analysis in men’s elite squash 75 Kinesiologia Slovenica, 9, 2, 74–85 (2003)
INTRODUCTION
The typical characteristic of squash is high velocity of the ball on a relatively small court, which 
makes this game extremely dynamic. All this affects players’ movements and their external 
loading during the match. The term loading is defined as movement expressed either in exercise 
quantity indicators (Ušaj, 1996) or physical units and numerical marks (Dežman, 1999).
A thorough examination of athlete’s loading is highly important in view of the training activity.  
Such information provides an essential basis for adequate planning and dosing of loading dur-
ing trainings, which indirectly improves the efficacy of the training process. This is why many 
researchers examined players’ loading in various sports (basketball – Dežman, 1991; Mahorič, 
1994; football – Erdmann, 1992; O’Donoghue & Parker, 2002; handball – Bon, 2001; Pori, 2001). 
The methodology for tracking players during a match of squash was mostly dealt with by Hughes 
and colleagues (Hughes, 1998; Hughes, Franks & Nagelkerke, 1989).
Hughes and Franks (1994) established some statistically significant differences between the win-
ning and the losing players in terms of the distance covered during lateral and longitudinal move-
ments as well as the average velocity of movement in four groups of squash players of greater or 
lesser ability. They also attempted to characterise the competitive edge which helped the squash 
player Jahangir Khan maintain his lead at the top of the rankings for such a long time. The results 
of the research showed that the average acceleration of the above mentioned player during a rally 
was much higher (by up to 50%) than that of his opponents. The research of Hughes and Franks 
(1994) led to a surprising result: the average distance covered by the top-club squash players in 
a rally was 12 metres. According to Eubank and Messenger (2000) players made 2.866 steps in 
a match on average, i.e. 580 steps in a game. As much as 74.4% of movements consisted of the 
“flying phase”, which in the authors’ opinion was owing to a dynamic nature of squash.
None of the above researches provided any information on total distance covered in a game or 
a match. The movement, i.e. distance covered during a game or a match, is affected by various 
factors. Among them are the total time of game or match and in this framework also the “ac-
tive part of play” accounting for about 60% of the total time (Beaudin, Zapiec & Montgomery, 
1978; Mercier, Beillot, Gratas, Rochcongar, Lessard, Andre & Dassonville, 1987) and consisting 
of individual rallies. A rally starts at the moment the server throws the ball in the air and lasts 
until one of the players scores a point or makes an error which interrupts the play. The duration 
of rallies depends on the players’ ability, ranking, fitness, the importance of the match etc. The 
external loading of players also depends on their technical and tactical skills and knowledge. All 
this is reflected in players’ correct set-up and movements as well as the right choice of strikes. 
Better players usually stay longer on the T (basic) position, which enables them to make the right 
move, position themselves properly before striking and strike the ball optimally (McKenzie, 
1999). Thanks to this they may exert an indirect pressure on their opponents and force them to 
move faster and run greater distances. Based on the above we estimate that the velocity and the 
distance covered by the winning player in a game are lower than those of the losing player.
Therefore, the purpose of this study is to measure the average velocity of movement and the dis-
tance covered by squash players, and on this basis determine statistically significant differences 
between the winners and losers of a game in terms of the indicators described above.

76 Movement analysis in men’s elite squash Kinesiologia Slovenica, 9, 2, 74–85 (2003)
METHOD
Participants
The sample consists of eight top-ranked Slovene (4), Austrian (3) and Bavarian (1) squash players 
with experience in the major European competitions and professional tournaments. Three of the 
matches took place at the Slovenian National Championship finals (April 2001) and three at the 
Austrian International Championship finals (October 2001), featuring the players from Austria, 
Hungary, Slovenia and Bavaria.
Data collection method
Three matches were recorded at the 2001 Slovenian National Championship in Ljubljana and the 
other three at the Vienna international tournament of four countries in the same year. All matches 
consisted of 24 games and in all of them one of the players won and the other lost.
All matches were recorded with a fixed SVHS video camera (Ultrak CCD Color KC 7501 CP) 
with the frequency of capturing input pictures of 25 Hz. The camera was fastened to the ceiling 
in the centre of the squash court and its wide-angled lens covered the entire court. The camera 
did not interfere with the play and could not be hit by the ball.
The video-recordings were digitized using the miroVideo DC30+ video digitizer hardware (Ger-
many) with the resolution of 384x576 at 2 MB/sec data rate, while the processing was carried out 
at a resolution of 384x288 pixels.
Digital images were processed by the SAGIT/SQUASH tracking system (Perš, Vučković, Kovačič 
& Dežman, 2001). We were tracking both players’ movements in terms of space and time. The 
tracking algorithm uses the principle of matching the current image to a template made from 
empty court based on the following formula:
S
R
(i,j)  = | S
T
(R,i,j) – S
O
(R,i,j)| +
+ |S
T
(G,i,j) – S
O
(G,i,j)| +
+ |S
T
(B,i,j) – S
O
(B,i,j)|,
where S
R
 is the image of difference, i and j are indices of image elements (pixels) at x and y, S
T
 
is the current (colour) RGB image, S
O 
is colour image of the background, i.e. empty court. The 
symbols R, G and B stand for individual colour channels of the image – red, green and blue. The 
next step is binarisation (thresholding), where either 0 or 1 are attached to each pixel, depending 
on whether its value is higher or lower than the required threshold. The purpose of this procedure 
is to attach 0 to the pixels representing the background and 1 to those representing the players.
S(i.j) = 
{
 0, i f S
R
(i,j) ≤ threshold
1, if S
R
(i,j) > threshold.
Initially, the operator sets the threshold value so that the algorithm optimally distinguishes the 
players from the background; during the processing the algorithm automatically adjusts the 
threshold so that the areas representing the players on the binary image S remain of the same 
size throughout the processing.

Movement analysis in men’s elite squash 77 Kinesiologia Slovenica, 9, 2, 74–85 (2003)
Figure 1: The image of players 
Figure 2: The binarised image S
The following steps enabled the conversion into quantity data 
Step 1 – System calibration based on the court markings 
Step 2 – Determination of the processing point.
The result of the binarization is a new picture, showing only two values. The first value (in this 
case “zero”) determines the points representing the background (black points on the right side 

78 Movement analysis in men’s elite squash Kinesiologia Slovenica, 9, 2, 74–85 (2003)
of Figure 1), while the second value (in this case “one”) determines the points representing the 
moving objects – players (white points on the on the right side of Figure 1). In case there are two 
players on the picture, two clusters of points appear bearing the value “one”, representing the 
players. The position of each player, defined by the pair of coordinates (x
p
, y
p
), is determined as 
the centre of gravity of each of the white clusters, following from these equations:
                
1
         M     N                                     
1
          M   N
x
p
 = ————— 
∑ ∑
 S(i,j)*i and, y
p
 = —————
 
∑∑
 S(i,j)*j, where S(i,j) is the matrix
       
 
M    N
 
           i = 1 j = 1
                             
M    N
           
i = 1 j = 1 
       
∑∑
 S(i,j)                                       
∑∑
 S(i,j)
       i = 1 j = 1                                               i = 1 j = 1
of binarised pixel values of M x N dimensions.
Step 3 – Potential re-setting, in case the computer software “loses track” (manual redefining 
of the player’s position)
Step 4 – Reduction of measurement errors in the calculation of velocity and distance separately 
for each of the components, following from the two equations below (Perš, Bon, Kovačič, Šibila 
& Dežman, 2002):
                 
1
          
N
F
x'(t) = ————  
∑
  x(t + i) · G(i),
            2N
F
 + 1  
i = –N
F
                 
1
          N
F
y'(t) = ————  
∑
  y(t + i) · G(i),
            2N
F
 + 1  
i = –N
F
where x and y are original trajectory components, and x' and y' are smoothed trajectory compo-
nents. G(i) is the smoothing kernel, consisting of the coefficients of the Gauss function ranging 
from –3 sigma to 3 sigma. The kernel is normalized, therefore the sum of all its coefficients 
equals 1, and 2N
F 
+1 is the kernel width. Kernel width is directly related to smoothing intensity 
– the wider the kernel the smoother the trajectories. The results stated herein are based on the 
following kernel width: 2Nf+1 = 11 samples, which on a time scale equals 0.5 second. 
Step 5 – Final data processing
Calculation of distance and velocity curves in terms of time as well as calculation of distance 
covered by the player based on the following equations of velocity
1
:
v
x
(k) =
 ∆ x(k) 
=
 x(k) – x(k – 1),  
v
y
(k) =
 ∆ y(k) 
=
 y(k) – y(k – 1) 
and v(k) = √v
x
(k)
2
 + v
y
(k)
2
,           ——–    ————–——             ——–    ——————
           ∆ t(k)      t(k) – t(k – 1)               ∆ t(k)     t(k) – t(k – 1)                    
—————–
as well as the distance covered: , where ∆ T = t(k)-t(k-1) and is determined by the frequency of 
capturing input pictures, which in this case is 40 milliseconds (1/25 second). 
Step 6 – Numerical and graphical presentation of movements
The sample of variables included the distance covered, velocity of movement, time of a game and 
time of the active and passive parts of play in individual games and matches.
1
 The result of the used method of calculating velocity is in fact the average velocity in the interval starting 40 millisec-
onds before the indicated time and ending at the indicated time. Our study established that this delay was not signifi cant 
in view of the method of measuring and data processing.

Movement analysis in men’s elite squash 79 Kinesiologia Slovenica, 9, 2, 74–85 (2003)
Figure 3: Diagram of player’s movements in a game
Data analysis 
We used the SAGIT/SQUASH program to measure the time of individual rallies and the percent-
ages of active parts in various games. The testing of statistically significant differences between 
the winning and the losing sides in a game in terms of distance covered and average velocity 
was based on a one-way analysis of variance for independent samples. Testing of statistically 
significant differences was made on the premises of 5% risk, while Pearson’s correlation coef-
ficient was used to establish correlation between time, number of scored points and distance 
covered in a game.
RESULTS
The results in Table 1 show high variability in the time of individual games within the same match 
and in various matches. The longest game lasted for 1,113 seconds and the shortest for only 194 
seconds. Somewhat smaller differences were seen between the games in the same match in terms 
of percentage of active parts of play. Bigger differences are seen in percentages of the active part, 
if games of different matches are compared. The highest percentage of active part in individual 
game was 74.8% and the lowest 48.4%. Similar results may be observed in individual games and 
matches. T wo matches ended with a minimal number of games (3); in two of them the winner was 
decided by the fifth game, which is also the maximum number of played games in one match.

80 Movement analysis in men’s elite squash Kinesiologia Slovenica, 9, 2, 74–85 (2003)
Table 1: Time of individual matches and games, percentage showing the active part of play and results of 
all games and matches
1
st
 match 2
nd
 match 3
rd
 match 4
th
 match 5
th
 match 6
th
 match
Time (s)
1
st
 game 931 s 408 s 562 s 942 s 755 s 593 s
2
nd
 game 543 s 987 s 793 s 674 s 589 s 926 s
3
rd
 game 566 s 896 s 705 s 1095 s 533 s 863 s
4
th
 game 259 s 1113 s / / 526 s 194 s
5
th
 game / 502 s / / / 855 s 
Match 38 m 19 s 1 h 5 m 6 s 38 m 51 s 49 m 38 s 45 m 43 s 57 m 11 s
Active part Percentage
1
st
 game 58.8 67.2 74.8 60.5 56.6 64.3
2
nd
 game 54.5 65.7 62.4 60.6 57.8 61.8
3
rd
 game 50.2 61.8 62.7 58.3 49.3 57.4
4
th
 game 59.5 60.2 / / 48.4 59.7
5
th
 game / 57.1 / / / 57.7
Match (%) 55.8 62.4 66.6 58.9 537 60.2
Result
1
st
 game 8:10 1:9 9:0 9:4 4:9 9:1
2
nd
 game 9:1 7:9 9:2 9:4 9:2 0:9
3
rd
 game 9:6 9:1 9:1 9:6 9:3 6:9
4
th
 game 9:0 9:6 / / 9:1 9:0
5
th
 game / 9:5 / / / 9:6
Match 3:1 3:2 3:0 3:0 3:1 3:2
Table 2: Total distance covered, distance covered in the active part of play and percentage of distance 
covered in the active part in view of the total distance covered in individual games and matches
Matches 1
st
 match 2
nd
 match 3
rd
 match 4
th
 match 5
th
 match 6
th
 match
Players W1 L1 W02 L2 W3 L3 W4 L4 W5 L5 W6 L6
dc1 1297 1308 524 583 825 793 1207 1130 850 875 872 827
dc1-ap 925 923 391 434 716 681 876 807 583 614 659 641
%ap-dc1 71.3 70.6 74.6 74.4 86.8 85.9 72.6 71.4 68.6 70.2 75.6 77.5
dc2 762 747 1298 1378 972 1018 834 811 744 680 1303 1325
dc2-ap 508 504 1020 1042 741 780 641 583 548 495 966 993
%ap-dc2 66.7 67.5 78.6 75.6 76.2 76.6 76.9 71.9 73.7 72.8 74.1 74.9
dc3 776 690 1241 1132 905 946 1368 1350 606 603 1200 1187
dc3-ap 496 464 915 881 678 718 983 949 396 375 862 830
%ap-dc3 69.9 67.2 73.7 77.8 75 75.9 71.9 70.3 65.3 62.2 71.8 69.9
dc4 347 332 1449 1389 / / / / 637 591 292 254
dc4-ap 249 236 1063 1036 / / / / 432 378 193 179
%ap-dc4 71.8 71.1 73.4 74.6 67.8 64 66.1 70.5
dc5 / / 655 619 / / / / / / 1183 1090
dc5-ap / / 455 452 / / / / / / 836 754
%ap-dc5 / / 69.5 73 / / / / / / 70.7 69.2
To t a l 3182 3077 5167 5101 2702 2757 3418 3291 2837 2749 4850 4683
Xa (game) 796 769 1033 1020 900 919 1139 1097 709 687 970 937
Legend: 
W1, W2, W3, W4, W5, W6 – winning side of game in a match 
L1, L2, L3, L4, L5, L6 – losing side of game in a match
DC – istance covered in a game (m)
DC-AP – distance covered in the active part of play
%AP-DC – distance covered in the active part of play in view of total distance covered in a game, in percentage (%)
Xa (game) – average distance covered in games.

Movement analysis in men’s elite squash 81 Kinesiologia Slovenica, 9, 2, 74–85 (2003)
The results in Table 2 also show large differences in the distance covered by players in individual 
games of the same match. Some minor differences are shown in percentage of distance covered 
by the players in the active part of play. Moreover, distances covered by the winners of individual 
games or matches are quite similar to those of the losers.
Table 3: Players’ average velocity of movement in total time and in the active part of play in individual 
games and matches
Matches 1
st
 match 2
nd
 match 3
rd
 match 4
th
 match 5
th
 match 6
th
 match
Players W1 L1 W2 L2 W3 L3 W4 L4 W5 L5 W6 L6
v1 1.39 1.4 1.29 1.43 1.47 1.42 1.29 1.2 1.13 1.16 1.47 1.39
v1-AP 1.69 1.7 1.44 1.62 1.72 1.64 1.55 1.43 1.38 1.46 1.73 1.68
v2 1.37 1.37 1.31 1.4 1.23 1.29 1.25 1.21 1.26 1.16 1.41 1.43
v2-AP 1.71 1.76 1.58 1.62 1.51 1.58 1.58 1.44 1.63 1.46 1.7 1.74
v3 1.22 1.34 1.39 1.27 1.29 1.35 1.25 1.24 1.14 1.14 1.39 1.37
v3-AP 1.64 1.62 1.67 1.61 1.55 1.63 1.55 1.5 1.54 1.45 1.71 1.78
v4 1.34 1.28 1.3 1.25 / / / / 1.22 1.13 1.48 1.29
v4-AP 1.62 1.54 1.6 1.57 / / / / 1.72 1.5 1.63 1.52
v5 / / 1.3 1.23 / / / / / / 1.33 1.27
v5-AP / / 1.6 1.59 / / / / / / 1.7 1.5
v (xa) 1.33 1.34 1.32 1.32 1.33 1.35 1.26 1.22 1.19 1.15 1.42 1.35
v AP(xa) 1.67 1.66 1.59 1.6 1.59 1.62 1.56 1.46 1.57 1.47 1.69 1.64
Legend: 
v1, v2, v3, v4, v5 – player’s average velocity of movement in a game (m/s)
v1-AP – player’s average velocity of movement in the active part of play (m/s) 
v (xa) – player’s average velocity of movement in the match (m/s)
v AP(xa) – player’s average velocity of movement in the active part of play in a match (m/s)
 
As regards winners and losers of individual games their average velocity of movement is very 
similar. The highest average velocity in the active part of play was 1.78 m/s and the lowest 1.38 
m/s. 
Table 4: Results of the variance analysis in terms of distance covered and average velocity of movement 
of winners and losers of all games
Variable Player M SD F p
DC
Winners 922.79 328.15
0.046 0.831
Losers 902.42 329.86
DC-AP
Winners 672.17 250.39
0.049 0.826
Losers 656.21 250.35
v
Winners 1.32 0.094
1.451 0.235
Losers 1.29 0.097
v-AP
Winners 1.62 0.097
1.896 0.175
Losers 1.58 0.099
Legend: 
DC – distance covered in a game (m)
DC-AP – distance covered in the active part of play (m)
v – average velocity of movement (m/s) 
v-AP – average velocity of movement in the active part of play (m/s)

82 Movement analysis in men’s elite squash Kinesiologia Slovenica, 9, 2, 74–85 (2003)
There were no statistically significant differences between the winning and the losing players in 
terms of distance covered and average velocity of movement in individual games. 
DISCUSSION
A comparison of distances covered by the winners and those by the losers of individual games 
shows that the winners in seventeen games covered a greater distance than the losers as well. 
As regards the active part of play the winners covered greater distances in 18 games of total 24. 
Even though these differences are not statistically significant, the results are surprising, as the 
winning side was expected to cover smaller distances. Such results may stem from the fact that 
winners make more services and winning returns than losers. Immediately after servicing the 
player reaches his/her basic position in two or three steps. Meanwhile, the receiving side is already 
in a position enabling it to make a good return and thus it need not move a lot. The second reason 
may lie in different tactics used by the players. Some players with more defensive way of playing 
will run down more skilful player’s shot until he/she makes an error.  A comparison between the 
winners and losers in terms of distance covered also shows an equalised play in all of the many 
games, ending in very low scores.
There was a statistically significant correlation between the time of individual games and the 
distance covered by the players in a game (r = .979, p < .001) as well as between the results of 
games (r = .563, p < .001) and the above stated distance. Therefore, distance covered in a game 
depends mostly on the time of game and the number of points scored by the players in a game. In 
approximately 13 m inutes 1.0 0 0 met res of dista nce a re covered. I n view of t he fact t hat individua l 
highest-level games last for at least 25 minutes, it may be concluded that top squash players may 
cover up to two kilometres in a game. The greatest distance covered, which was recorded in this 
study, was that in the fourth game of the second match (1,449 m). That game was the longest of 
all. The shortest distance covered was recorded in the fourth game of the sixth match which was 
also the shortest game, where the losing side’s total distance covered was only 254 m. Distance 
covered is only smaller in the active part of play; in just two games it was greater than 1,000 
m for both players. Distance covered in the active part of play, expressed as percentage of total 
distance covered, varies between 62.2% and 86.8%. Such a high correlation confirms that the 
percentage of player’s total distance covered in the active part of play substantially depends on 
the percentage of active part of play (r = .891, p < .001) and the player’s velocity of movement in 
the same game (r = .425, p < .001). It is not possible to compare our results of distances covered 
to those of other authors, as according to foreign professional literature there is no other study 
dealing with players’ distances covered in individual game or match of squash. Therefore, these 
results are important information and serve as a basis for adequate planning and implementing 
of training processes.
The velocity at which the body moves is the key criterion in measuring of intensity of internal 
loading and, together with the velocity at which individual parts of the body move, plays an 
important role in a quality play (Ackland, Bloomfield & Elliott, 1994; Sharp, 1982). Our study 
examined the average velocity of movement in total time of game and particularly in the active 
parts of game, which was a better criterion for thorough analysis of play than that of highest 
velocity of movement. We assumed that the latter would be lower with the winners, as they are 
often at a priority position. The reasons for the above are better execution of various strokes, better 
control over the basic position and consequently less intensity of movement. Obviously, average 

Movement analysis in men’s elite squash 83 Kinesiologia Slovenica, 9, 2, 74–85 (2003)
velocity of winners throughout a game and in individual active parts is slightly higher than that 
of the losers, but the differences are statistically insignificant. This may be ascribed to equality 
of the sampled players’ skills and to the fact that the winning and the losing sides used quite 
similar tactics. The velocity ranges from 1.13 m/s to 1.48 m/s throughout the game. The relevant 
figures for the active part of play are statistically higher (from 1.38 m/s to 1.78 m/s). Such results 
were expected, as the players’ velocity of movement in the passive part of play is low, because 
the players do not move much and rather concentrate on the continuation of the game.
I n t hei r st udy Hughes a nd Fra n k s (1994) measu re d sl ight ly h igher va lues showi ng average velocit y 
of movement i n t he act ive pa r t of play. T he h igher average velocit y of movement of t he top -ra n k i ng 
players was 1.98 ms
-1
 a nd could have be en t he result of players’ bet ter te ch n ica l-t act ica l sk i l ls. Top 
players’ strokes are more accurate, which forces the opponent to run greater distances and allows 
them less time to react. Their tactics involves playing all over the court, which is why the players 
are forced to continually move. Beside that best players are probably capable of developing higher 
velocity of movement. Statistically significant differences in the average velocity of movement of 
winners and losers may have arisen from differences in the data collection methodology, as the 
above mentioned authors based their study only on the last 10 seconds of each rally.
We have also established that the times of individual games in the same match as well as in dif-
ferent matches vary greatly. This could be due to player’s tactical decision to intentionally let 
the opponent win a game (for various reasons) and then show a different face in the next game. 
Many times short game times are due to players’ tiredness or lack of concentration, as they are 
not capable of enduring a high-level play. The results may be compared to the times recorded at 
four matches played by the national teams of England and France at the European Squash Team 
Championship finals (May 2002). On average, the four matches lasted for 92 minutes and an 
individual game for 24 minutes, which is longer than in our study (European Team Champion-
ship, 2002). Bearing in mind that the matches played an important role in our study and the 
players’ condition was accordingly high, such big differences could be ascribed to the technical 
and tactical skills of the players from the abovementioned countries.
In spite of a probably poorer technical and tactical knowledge of the sampled players, the percent-
age of the active part of play (59.5% in all matches) was quite similar to that established in the 
studies by Beaudin, Zapiec and Montgomery (1978) as well as Mercier et al., (1987). A slightly 
higher percentage of the active part of play (68%) was established by Hughes and Robertson 
(1998) in a sample of top-ranking squash players. Our study recorded similar and even higher 
percentages of active part of play in a game, and not so high percentages in a match.
One previous squash research (Hughes and Franks, 1994) led us to believe that the losers of 
individual games would have statistically higher velocities and greater distances covered than 
the winners, owing to winners’ better technical and tactical skills and knowledge. The results 
actually reflect the fact that the winners of games did not always win individual rallies of a game. 
Therefore, it would be reasonable to examine the discussed indicators of the players’ loading dur-
ing a rally and distinguish between the winners and the losers on the basis of shorter, completed 
units (individual rallies). The study of intensity of movement and the differences between the 
winning and the losing side should also count acceleration and deceleration among major and 
typical features of movements in squash. Presently, the SAGIT/SQUASH tracking system does 
not enable us to a na lyse t hese i nd icators of movement i ntensit y; t herefore, we bel ieve t hat it would 
be reasonable to develop a system for this purpose.

84 Movement analysis in men’s elite squash Kinesiologia Slovenica, 9, 2, 74–85 (2003)
REFERENCES
Ackland, T. R., Bloomfield, J., & Elliott, B. C. (1994). Applied Anatomy and Biomechanics in Sport. Vic-
toria: Blackwell scientific publication.
Beaudin, P., Zapiec, C., & Montgomery, D. (1978). Heart Rate Response and Lactic Acid Concentration in 
Squash Players. Research Quarterly, 49(4), 406-412.
Bon, M. (2001). Kvantificirano vrednotenje obremenitev in spremljanje frekvence srca igralcev rokometa 
med tekmo [Quantified Evaluation of Loading and Monitoring Heart Rate of Handball Players in a Match]. 
Unpublished doctoral dissertation, Ljubljana: Faculty of Sport.
Dežman, B. (1991). Obremenitve sodnikov na košarkarski tekmi [Loading of Referees in a Basketball 
Match]. Šport, 39 (4), 11-13.
Dežman, B. (1999). Zunanje in notranje obremenitve igralcev med košarkarsko tekmo [Internal and External 
Loading of Players in a Basketball Match]. Research project (draft). Ljubljana: Faculty of Sport.
Erdmann, W. S. (1992). Gathering of Kinematic Data of Sport Event by Televising the Whole Pitch and 
Track. In Rodano R., Ferrigno G., Santambrogio G., (Eds), Proceedings of 10
th 
ISBS Symposium (pp. 
159–162). Roma.
Eubank, C., & Messenger, N. (2000). Dynamic Moves and Stepping Patterns T ypical to the Game of Squash. 
Journal of Sports Sciences, 18 (7), 471-472.
Hughes, M., Franks, I.M., & Nagelkerke, P. (1989). A Video-system for the Quantitative Motion Analysis 
of Athletes in Competitive Sport. Journal of Movement Studies 17, 217-227.
Hughes, M., & Franks, I. M. (1994). Dynamic Patterns of Movement of Squash Players of Different Stand-
ards in Winning and Losing Rallies. Ergonomics, 37 (1), 23-29.
Hughes, M. (1998). The Application of Notational Analysis to Racket Sports. In Lees A., Maynard I., Hughes 
M., & Reilly T. (Eds.), Science and Racket Sports II (pp. 211-220). London. 
Hughes, M., & Robertson, C. (1998). Using Computerised Notational Analysis to Create a Template for Elite 
Squash and its Subsequent Use in Designing Hand Notation Systems for Player Development. In Lees A., 
Maynard I., Hughes M., & Reilly T. (Eds.), Science and Racket Sports II (pp. 227-234). London.
Mahorič, T. (1994). Zunanje in notranje obremenitve beka na košarkarski tekmi [Internal and External Load-
ing of Guard in a Basketball Match]. Unpublished bachelor’s thesis. Ljubljana:  Faculty of Sport.
McKenzie, J. (1999). Tactical Tips. The Squash Player, 27(6), 9.
European Team Championships (2002). Retrieved from November 2, 2002, from the World Wide Web: http:
//www.squashplayer.co.uk/events/events.htm
Mercier, M., Beillot, J., Gratas, A., Rochcongar, P., Lessard, Y., Andre, A.M., & Dassonville, J. (1987). 
Adaptation to Work Load in Squash Players: Laboratory Tests and On-court Recordings. The Journal of 
Sports Medicine and Physical Fitness, 27, 98-104.
O'Donoghue, P.G., & Parker, D. (2002). Time-motion Analysis of FA Premier League Soccer Competition. 
Journal of Sports Sciences. 26 (1), 26.
Perš, J., Vučković, G., Kovačič, S., & Dežman, B. (2001). A Low-cost Real-time Tracker of Live Sport 
Events. In Lončarić S.,& Babić, H. (Eds.), ISPA 2001: Proceedings of the 2
nd
 international symposium on 
image and signal processing and analysis in conjunction with 23
nd
 int'l conference on information technol-
ogy interfaces (pp. 362-365). Pula, Croatia. 

Movement analysis in men’s elite squash 85 Kinesiologia Slovenica, 9, 2, 74–85 (2003)
Perš, J., Bon, M., Kovačič, S., Šibila, M., & Dežman, B. (2002). Observation 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 Cyclic Loading of Players at Various Positions in Attack in a Handball 
Match]. Unpublished master’s thesis, Ljubljana:  Faculty of Sport.
Sharp, C. (1982). Well, Just How do Y ou Measure Up to the Tests of Competitive Fitness? The Squash Player 
International, November, 26-28.
Ušaj, A. (1996). Kratek pregled osnov športnega treniranja [Short Overview of Sport Training Basics]. 
Ljubljana: Fakulteta za šport, Inštitut za šport.