See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/260226858

Ball recovery patterns as a performance indicator in elite soccer

Article in Proceedings of the Institution of Mechanical Engineers Part P Journal of Sports Engineering and Technology · March 2014
DOI: 10.1177/1754337113493083

CITATIONS READS

143 11,606

5 authors, including:

Daniel Barreira Júlio Garganta

University of Porto University of Porto
90 PUBLICATIONS 1,682 CITATIONS 189 PUBLICATIONS 4,068 CITATIONS

SEE PROFILE SEE PROFILE

João Cláudio Machado

Federal University of Amazonas
91 PUBLICATIONS 843 CITATIONS

SEE PROFILE

All content following this page was uploaded by Júlio Garganta on 21 February 2014.

The user has requested enhancement of the downloaded file.

Original Article

Proc IMechE Part P:

J Sports Engineering and Technology
2014, Vol. 228(1) 61–72
Ball recovery patterns as a Ó IMechE 2013
Reprints and permissions:
performance indicator in elite soccer sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/1754337113493083
pip.sagepub.com

Daniel Barreira1,2, Júlio Garganta1,2, Pedro Guimarães1, João Machado1

and Maria T Anguera3,4

Abstract
In soccer, it seems relevant to understand the relationship between the ball recovering and the subsequent success or
failure of attacking play. However, few studies have considered the links between the type of ball recovery in different
pitch zones, the competition stages and the overall teams success. The present study aims to analyze the attacks (n =
1619) carried out by the semi-finalist teams in the 2010 FIFA World Cup in order to explore ball recovery patterns as a
performance indicator. SoccerEye observational instrument, SoccerEye recording software, Sequential Data Interchange
Standard-Generalized Sequential Querier (SDIS-GSEQ) and SPSS analytic software—one-way analysis of variance, two-
way analysis of variance and regressions—were applied. Direct ball recovery, in specific by interception and defensive
behavior followed by a pass, was the mostly frequent behavior, with the later inducing attacking play efficacy (p \ 0.017).
Differences were detected between the group and play-off stages with regard to the types of direct ball recoveries. The
ball was most often regained in defensive and mid-defensive central zones, evidencing differences to all other pitch zones
(p 4 0.001). Throw-ins were the only type of ball recovery that differentiated the semi-finalists, namely Germany and
Spain (p \ 0.009). It was found that recovering directly the ball possession in mid-defensive central zones increases
attacking efficacy. Consequently, coaches should consider this tactical determinant in order to organize the training pro-
cess. Specifically, it is fundamental to improve the collective defensive organization protecting central strip zones and
simultaneously performing high-pitched pressure to constrain the ball carrier.

Keywords
Match analysis, tactical determinants, competition stage, team patterns, FIFA World Cup, SoccerEye software

Date received: 20 March 2013; accepted: 15 May 2013

Introduction to attack efficacy whether in single matches, tourna-

ments, seasons or groups of seasons12 or even in a sin-
The overall tactics of soccer entail a permanent interre- gle move in one match.13
lationship between the patterns of attacking and defen- The success of attacking play among elite soccer
sive play. The variability in these patterns of play arises teams also depends on ball recovery patterns, that is,
from the ways as the players (and teams) manage the on the types of ball recovery and the zones where it
constraints that emerge from their cooperation/opposi- occurs.14 Thus, considering ball recovery as the purpose
tion behavior, and with situational variables such as of the defensive phase and the first phase of attack15 is
match status,1,2 game location,3 quality and identity of
the opposing teams,4 pitch size5 and competition stage.6
Notational research concerning elite soccer competi-
1
tions has mostly focused in the attacking play:7 on the 2
Faculty of Sport, University of Porto, Porto, Portugal
Centre of Research, Training, Innovation and Intervention in Sport,
one hand, with the measurement of scoring indicators University of Porto, Porto, Portugal
such as goals,8 efficacy of shots9 and the ratios of win- 3
Faculty of Psychology, University of Barcelona, Barcelona, Spain
ners to errors and goals to shots,3 and on the other 4
Institute for Research on the Brain, Cognition and Behaviour, University
hand, with the analysis of performance indicators such of Barcelona, Barcelona, Spain
as the type9 and number of passes,10 ball possessions11
Corresponding author:
and entries in the penalty area.2 Accordingly, the Daniel Barreira, Faculty of Sport, University of Porto, Rua Dr. Plácido
majority of published research has mostly studied the Costa, 100, 4200.450 Porto, Portugal.
determinants of ball possession and their relationship Email: [email protected]
62 Proc IMechE Part P: J Sports Engineering and Technology 228(1)

fundamental. In this regard, the influence of ball recov- recovery patterns in relation to pitch zones, competi-
ery patterns, considered as a performance indicator tion stage and observed team; and (c) to investigate the
selection or combination of action variables that aims relationship between the ball recovery patterns and fin-
to define some or all features of performance7 and pre- ishing attacking events.
dicts success,6 needs to be understood so as to be able
to implement adequate training regimes and to obtain
objective feedback.16 Moreover, it is important to scru-
Method
tinize ball recovery patterns according to the team and Design
competition stages, as well as the influence that the way
The flexibility and rigor of observational methodology
the ball is regained has on the final attacking events,
make it fully consistent with the characteristics of the
that is, considering the patterns of ball recovery as a
study.29 The observational design, in accordance with
performance indicator that possibly predicts the attack-
the specific taxonomy,30 was nomothetic (four teams),
ing success on elite soccer.
followed-up (continuous recording across matches, with
Indeed, research on ball recovery patterns has been
independent observation of each of the two opposing
applied to a lesser extent. Specifically, the influence of
teams) and multidimensional (three criteria included in
the zone in which the ball is recovered and its relation-
the observational instrument). This approach allowed
ship to attack efficacy has been determined.17–19
us to identify attacks in the observed matches.
Research in World Cups between 1982 and 1990 and in
European clubs and national teams showed that the
ball was mainly recovered in the central strip of the Sample
pitch due to the higher concentration of players in this
We recorded 1619 attacks from public TV broadcasts
zone. However, in World Cups between 1982 and 1990,
during 24 matches (six per semi-finalist team) played by
ball recoveries mostly occurred in the defensive sec-
Germany, The Netherlands, Spain and Uruguay—the
tor,20 whereas in European clubs and national teams,
2010 FIFA World Cup semi-finalists—also considered
the midfield sector was the predominant zone of ball
as successful teams.31 The 16-round stage was not
recovery.21 In the 1996 Euro Cup, it was observed that
attended. Matches were observed for the regular period
most of the goals scored resulted from ball recoveries
(i.e. 90 min, excluding extra time). The attacks in which
in the offensive sector,22 which confirmed that attack-
players left the camera’s recording field or in which a
ing success is associated with fast and frequent recovery
team had 10 or fewer players on the pitch were excluded
of the ball,23 in other words, in zones in which attack
from the analysis.
efficacy is greater.20 By contrast, an analysis of the
2002 World Cup concluded that goals in open play
occurred due to ball recoveries in the mid-defensive sec- Instruments
tor,24 which suggests that attack efficacy probably
Observational instrument. An updated version of the
appears when the attack starts at some distance from
SoccerEye observational instrument26,27 was used in
the opponent’s goal. In the 1994 World Cup, it was
this study and has been used in recent research.32,33
found that different playing styles were related to dif-
This tool follows an updated version of the
ferent patterns of ball recovery, with a fast defence/
Organizational Model of Soccer26,27 (Figure 1).
attack transition yielding benefits to attack efficiency.25
The present study focused exclusively on the offen-
Hence, the attacking game-pattern configuration is a
sive phase, in particular in the two corresponding types
consequence of ball recovery patterns.25
of ball recovering: (a) direct, that is, game flow was pre-
We noted, however, a lack of research considering
served, with no interruption, and a player performed at
the ball recovery patterns according to the level of team
least three consecutive ball touches, a positive pass, a
feat and the stage of the competition, while taking into
shot at the opponent’s goal34 or the goalkeeper con-
account the number of attacks performed and the rela-
trolled the ball35 and (b) indirect, that is, game flow
tionships between the type of ball recovery, the respec-
was broken up due to an opponent’s violation of the
tive pitch zone(s), and with the final attacking event.
laws of the game or because the ball leaves the pitch.
In light of this, the following tools were used:
Moreover, the final attacking events and the patterns
updated version of SoccerEye observational instru-
of pitch space position were considered (Table 1).
ment,26,27 SoccerEye recording software (v3.0, October
2012); and SDIS-GSEQ v5.128 and IBM SPSS v19.0
statistical analysis software to analyze the 1619 attacks Recording instrument. The SoccerEye recording software
recorded from the group and play-off (PO) stage (v3.0, October 2012) (Figure 2), written in Visual Basic
matches played by the semi-finalists in the 2010 FIFA Express 2010, is a soccer-specific tool that was designed
World Cup. to be used with the SoccerEye observational instru-
In summary, the aims of this study are (a) to charac- ment.32,33 This tool makes possible to observe and
terize the attacks performed and to look into differ- record the occurrence of perceivable behaviors in natu-
ences according to competition stage, observed teams ral context and data exporting in multiple formats that
and each type of ball recovery; (b) to characterize ball suits sequential data analysis with SDIS-GSEQ
Barreira et al. 63

Figure 1. Updated version of the organizational model of soccer.26,27

Table 1. Adapted version of SoccerEye observational instrument.32,33

Criterion Categories Description

Start of the offensive Direct ball recovery BRi Ball recovery by interception
phase/ball possession BRt Ball recovery by tackle
recovery (BR) BRgk Ball recovery by intervention of the
goalkeeper in the defensive phase
BRp Ball recovery by a defensive behavior followed by a pass
Indirect ball recovery BRst Start/restart of the offensive phase
BRv Ball recovery by opponent’s violation of the laws of the game
BRc Ball recovery by corner kick
BRgki Ball recovery by goal kick
BRdb Ball recovery by dropped ball
BRti Ball recovery by throw-in
End of the offensive With efficacy Fws Wide shot
phase (F) Fst Shot on target
Fso Shot stopped, with no maintenance of ball possession
Fgl Goal
With no efficacy Fed Loss of ball possession by error of the ball carrier/defender’s
intervention (exception to the goalkeeper)
Fgk Loss of ball possession by the intervention of the
opponent’s goalkeeper
Fo Throwing the ball out of the pitch
Fi Violation of the laws of the game
Patterns of pitch 1 Zone 1: defensive sector/left strip
space position 2 Zone 2: defensive sector/central strip
3 Zone 3: defensive sector/right strip
4 Zone 4: mid-defensive sector/left strip
5 Zone 5: mid-defensive sector/central strip
6 Zone 6: mid-defensive sector/right strip
7 Zone 7: mid-offensive sector/left strip
8 Zone 8: mid-offensive sector/central strip
9 Zone 9: mid-offensive sector/right strip
10 Zone 10: offensive sector/left strip
11 Zone 11: offensive sector/central strip
12 Zone 12: offensive sector/right strip

software. For each attack, the observer records the temporal and sequential structure, enabling to find and
match status, competition stage, match time, duration report the interactions concerning motor practices and
of the attack and any match events, regarding soccer dynamics of play.
64 Proc IMechE Part P: J Sports Engineering and Technology 228(1)

Figure 2. Soccer eye recording software (v3.0, October 2012).

Procedure there were 754 attacks (64.1 6 20.5), while across the
PO rounds, the four semi-finalists performed 865
Data quality. In line with the procedure described in recent
attacks (72.5 6 10.7). Spain—the winner of World Cup
research,36 data quality was ensured by assessing inter-
2010—performed the majority of attacks (n = 430, 72.3
observer reliability. Specifically, Cohen’s kappa index37
6 21.5), while Germany performed the fewest number
was calculated from the observations of the first half of
(n = 357, 59.2 6 21.1). The other two semi-finalists,
the 2010 World Cup final (The Netherlands vs Spain).
Dutch and Uruguayan national teams, completed 409
Application of SDIS-GSEQ software (v5.1)28 yielded val-
(70.1 6 12.0) and 423 (71.0 6 8.7) attacks, respectively.
ues of 0.92  0.98, well above the value of 50.75 estab-
The analysis of the number of attacks performed
lished as being indicative of high data quality.38
revealed no significant differences when comparing the
two stages of the competition (p \ 0.221) or the four
Statistical analysis. In addition to descriptive statistics, one- successful teams (p \ 0.509) (Figure 3).
and two-way analysis of variance (ANOVA) models were
used to analyze differences in the ball recovery according
to the stage of the competition, the four successful teams, Ball recovery patterns in the 2010 FIFA World Cup
the type of ball recovery and the respective pitch zones. Direct ball recovery prevailed. Direct ball recovery
Post hoc tests, namely Tukey’s honestly significant differ- revealed higher frequency than indirect ball recovery.
ence (HSD) test and the Sidak correction, were also used Specifically, ball recovery by interception (n = 380,
to check for specific differences. Multinomial and binary 16.0 6 9.7) and by a defensive behavior followed by a
logistic regression analyses were used to estimate the asso- pass (n = 412, 23.0 6 8.7) were the ball recovery pat-
ciation between the patterns of ball recovery and the final terns most frequently observed, while ball recovery by
attacking events. Statistical significance was set at 0.05. corner kick and dropped ball did not occur in any of
SPSS v19.0 was used for all analyses. the 24 matches (Figure 4).

Results Competition stage influenced direct ball recovery

patterns. Indirect ball recovery patterns did not differ
Attacks performed by the successful teams in the
according to the competition stage of the 2010 FIFA
2010 FIFA World Cup World Cup. However, there were significant differences
A total of 1619 attacks (67.5 6 3.3) were registered in between the group and PO stages regarding the types
the 24 matches observed. During the group stage (GS), of direct ball recovery (Table 2). Interception (23.0 6
Barreira et al. 65

Figure 3. Frequency and mean number of attacks performed during the 2010 FIFA World Cup according to competition stage and
the semi-finalists.

Figure 4. Mean frequency for the types of ball recovery and their relationship to the competition stages (group and play-off stages)
of the 2010 FIFA World Cup.
BRi: ball recovery by interception; BRt: ball recovery by tackle; BRgk: ball recovery by intervention of the goalkeeper in the defensive phase; BRp: ball
recovery by a defensive behavior followed by a pass; BRst: start/restart of the offensive phase; BRv: ball recovery by opponent’s violation of the laws
of the game; BRc: ball recovery by corner kick; BRgki: ball recovery by goal kick; BRdb: ball recovery by dropped ball; BRti: ball recovery by throw-in.

8.7, p = 0.000) and tackle (10.6 6 4.5, p \ 0.006) (1.6 6 0.7) and throw-in (3.7 6 1.5) to recover the
were more frequent in PO matches than during the ball than did The Netherlands (BRi: 13.7 6 7.0; BRp:
group stage of the tournament (GS, BRi: 8.9 6 3.4; 20.0 6 12.7; BRv: 3.7 6 1.5; BRti: 8.3 6 2.1), Spain
BRt: 5.8 6 3.3) (Figure 4). By contrast, ball recovery (BRi: 15.5 6 5.7; BRp: 20.5 6 6.3; BRv: 4.2 6 1.7;
by defensive behavior followed by a pass was more fre- BRti: 10.2 6 2.1) and Uruguay (BRi: 16.0 6 9.7; BRp:
quently observed in group stage matches (GS: 23.3 6 18.2 6 9.7; BRv: 4.6 6 3.2; BRti: 7.3 6 3.8). The oppo-
10.2 vs PO: 13.0 6 5.8; p \ 0.006). In relation to ball site occurred regarding ball recovery by tackle;
recovery due to intervention of the goalkeeper, a simi- Germany (9.7 6 7.2) employed more often (p \
lar rate of occurrence was observed in both stages of 0.843) than did the teams from The Netherlands (7.5 6
the competition and for both indirect (BRgki: GS: 6.8 2.9), Spain (7.5 6 3.1) and Uruguay (8.2 6 4.6) (Figure
6 3.5, PO: 6.3 6 2.6) and direct (BRgk: GS: 7.0 6 4.2, 5). Overall, however, all semi-finalists in the 2010
PO: 5.4 6 2.6) ball recoveries (Figure 4). FIFA World Cup showed similar patterns of ball
recovery, with the exception of ball recovery by throw-
in (p \ 0.012), for which there was a significant differ-
Ball recovery by throw-in differentiated Germany and ence (p \ 0.009) between Germany (3.67 6 1.5) and
Spain. The results in Table 2 show that the German Spain (10.2 6 2.1) (Figure 5).
national team performed less (p \ 0.323) use of inter-
ceptions (12.5 6 7.5), defensive behavior followed by a Ball recovery occurred mainly in defensive and mid-defensive
pass (16.7 6 10.2), violation of the laws of the game central zones of the field. The defensive and mid-defensive
66 Proc IMechE Part P: J Sports Engineering and Technology 228(1)

Table 2. Types of ball recovery during the 2010 FIFA World Cup according to competition stage and the semi-finalist team.

WC2010 Group stage Play-off Germany The Netherlands Spain Uruguay

Direct BR BRia 22.2 6 15.0 8.9 6 3.4 23.0 6 8.7 12.5 6 7.5 13.7 6 7.0 15.5 6 5.7 16.0 6 9.7
BRta 8.0 6 4.7 5.8 6 3.3 10.6 6 4.5 9.7 6 7.2 7.5 6 2.9 7.5 6 3.1 8.2 6 4.6
BRgk 4.3 6 2.6 7.0 6 4.2 5.4 6 2.6 6.8 6 3.8 7.5 6 4.2 6.2 6 3.4 6.2 6 3.5
BRpa 15.5 6 10.2 23.3 6 10.2 13.0 6 5.8 16.7 6 10.2 20.0 6 12.7 20.5 6 6.3 18.2 6 9.7
Indirect BR BRst 1.5 6 1.2 1.1 6 0.5 1.6 6 1.0 1.5 6 0.8 1.2 6 0.8 1.2 6 0.4 1.3 6 0.8
BRv 1.8 6 0.8 3.8 6 3.5 5.5 6 2.8 1.6 6 0.7 3.7 6 1.5 4.2 6 1.7 4.6 6 3.2
BRcb – – – – – – –
BRgki 7.5 6 4.2 6.8 6 3.5 6.3 6 2.6 6.0 6 1.7 7.5 6 2.6 5.3 6 3.1 6.6 6 3.0
BRdbb – – – – – – –
BRtic 6.8 6 2.3 7.4 6 3.2 7.1 6 4.4 3.7 6 1.5d 8.3 6 2.1 10.2 6 2.1d 7.3 6 3.8

Values are mean 6 SD.

WC2010: FIFA World Cup 2010; BR: ball recovery; BRi: ball recovery by interception; BRt: ball recovery by tackle; BRgk: ball recovery by
intervention of the goalkeeper in the defensive phase; BRp: ball recovery by a defensive behavior followed by a pass; BRst: start/restart of the
offensive phase; BRv: ball recovery by opponent’s violation of the laws of the game; BRc: ball recovery by corner kick; BRgki: ball recovery by goal
kick; BRdb: ball recovery by dropped ball; BRti: ball recovery by throw-in; SD: standard deviation; ANOVA: analysis of variance.
a
Significant difference between the group and play-off stages of WC2010; p4 0.05 according to one-way ANOVA.
b
Type of ball recovery not performed during WC2010.
c
Significant difference when comparing the four successful teams in WC2010; p4 0.05 according to one-way ANOVA.
d
Significant difference between Germany and Spain (p \ 0.009) in WC2010; p4 0.05 according to Tukey’s HSD test.

Figure 5. Relationship between the types of ball recovery and successful teams in the 2010 FIFA World Cup.
BRi: ball recovery by interception; BRt: ball recovery by tackle; BRgk: ball recovery by intervention of the goalkeeper in the defensive phase; BRp: ball
recovery by a defensive behavior followed by a pass; BRst: start/restart of the offensive phase; BRv: ball recovery by opponent’s violation of the laws
of the game; BRc: ball recovery by corner kick; BRgki: ball recovery by goal kick; BRdb: ball recovery by dropped ball; BRti: ball recovery by throw-in.
Values are means.

sectors, both categorized as the defensive midfield 8: 4.1 6 0.6) was the zone of the pitch most widely used
(Table 1), were the pitch zones, in which the ball was for recovering the ball in the mid-offensive and offen-
most often recovered by the semi-finalists in the 2010 sive sectors, which confirms the tendency for using the
FIFA World Cup. A trend for recovering the ball in central strip observed for the defensive midfield.
central zones was observed (zone 2: 18.5 6 1.1; zone 5: However, zone 8 showed no significant differences with
18.9 6 1.6), with results showing significant differences respect to its corresponding lateral zones, that is, zone
between zones 2 and 5 and all the other pitch zones 7 (2.8 6 0.5, p \ 0.993) and zone 9 (2.9 6 0.3, p \
(p4 0.001) (Table 3); still, no differences were observed 0.998). On sporadic occasions, only the ball was recov-
between zones 2 and 5 (p = 1.000), corroborating the ered in offensive sector zones (zone 10: 0.2 6 0.1; zone
tendency for ball recovery to occur primarily near the 11: 0.1 6 0.1; zone 12: 0.1 6 0.1), which produced sig-
teams’ own goals. nificant less ball recovery situations than the overall
When considering each pitch sector separately pitch zones (p \ 0.005). This indicates that teams
(defensive, mid-defensive, mid-offensive and offensive), rarely recovered the ball near the opponent’s goal.
a slight tendency was detected for the ball to be simi-
larly recovered in the opposing lateral zones (p =
1.000) (e.g. zone 4 vs zone 6, both in the mid-defensive Patterns of ball recovery used by semi-finalists were similar and
sector). Similarly, the central mid-offensive zone (zone independent of the competition stage. Overall, there were
Barreira et al. 67

5.8 6 0.7c
2.0 6 1.0
24.0 6 3.1
4.3 6 1.1
8.0 6 2.6
20.7 6 4.7
3.7 6 2.5
1.7 6 1.3
2.7 6 1.6
2.0 6 0.9
0.0 6 0.3
0.0 6 0.2
0.0 6 0.2
no significant differences (p \ 0.053) between the

Play-off
stage
group and PO stages with regard to the zones in which
the ball was recovered. Although the mean number of
ball recoveries was lower in the PO stage (PO: 5.1 6

6.0 6 0.7c
1.0 6 1.0
15.0 6 3.1
1.3 6 1.1
9.3 6 2.6
21.6 6 4.7
10.3 6 2.5
4.0 6 1.3
6.7 6 1.6
3.0 6 0.9
0.0 6 0.3
0.0 6 0.2
0.3 6 0.2
0.4; GS: 6.2 6 0.4, Table 3), the results for zones 1 and
Group
stage

5 did not follow this trend. More ball recoveries were

observed during the PO matches (zone 1: 2.4 6 0.5;
Table 3. Zones of ball recovery (from 1 to 12, see Table 1) during the 2010 FIFA World Cup (WC2010) according to competition stage and each of the successful teams.

zone 5: 19.3 6 2.3) than during the group stage (zone

1.5 6 0.7
19.5 6 2.2
2.8 6 0.8
8.7 6 1.9
21.2 6 3.2
7.0 6 1.8
2.8 6 0.9
4.7 6 1.2
2.5 6 0.6
0.0 6 0.2
0.0 6 0.1
0.2 6 0.1
WC2010a
Uruguay

1: 1.9 6 0.5; zone 5: 18.5 6 2.3) (Table 3).

When considering each of the four semi-finalists sep-

–
arately, the results showed no influence of competition

5.1 6 0.7c
stage for any of the pitch zones in which the ball was
3.0 6 1.0
16.0 6 3.1
1.0 6 1.1
6.3 6 2.6
17.3 6 4.7
5.0 6 2.5
3.7 6 1.3
5.7 6 1.6
3.0 6 0.9
0.7 6 0.3
0.0 6 0.2
0.0 6 0.2
Play-off

recovered (p \ 0.968). Additionally, there were no dif-

stage

ferences between the German, Dutch, Spanish and

Ball recovery occurrence in relation to pitch zones per team in each stage of the competition: team 3 stage interaction, p4 0.05, using Sidak’s post-hoc correction.
Uruguayan national teams (p \ 0.386). The analysis
6.6 6 0.7c
0.7 6 1.0
20.7 6 3.1
3.7 6 1.1
10.7 6 2.6
18.3 6 4.7
12.3 6 2.5
2.3 6 1.3
4.7 6 1.6
5.0 6 0.9
0.0 6 0.3
1.0 6 0.2
0.3 6 0.2

therefore revealed no interaction (p \ 0.229) between

Group

competition stage and teams (p \ 0.630 for all com-

stage

parisons between teams).

4.0 6 0.64
1.8 6 0.7
18.3 6 2.2
2.3 6 0.8
8.5 6 1.9
17.8 6 3.3
8.7 6 1.8
3.0 6 0.9
5.2 6 1.2

0.3 6 0.2
0.5 6 0.1
0.2 6 0.1
a
WC2010

Ball recovery by tackle induced goal scoring. Analyzing the

Spain

association between the patterns of ball recovery and

the final attacking events, we found that recovering the
5.9 6 0.7c
2.7 6 1.0
21.7 6 3.1
2.3 6 1.1
4.0 6 2.6
23.3 6 4.7
6.0 6 2.5
4.3 6 1.3
2.3 6 1.6
3.3 6 1.0
0.3 6 0.3
0.0 6 0.2
0.3 6 0.2

ball by tackle induced goal scoring (p = 0.050) (Table

Play-off

4). Also, ball recovery by a defensive behavior followed

stage

Ball recovery occurrence in relation to pitch zones per stage of the competition; p4 0.05, using Sidak’s post-hoc correction.

by a pass increased the occurrence of shots on target (p

Six observations, corresponding to the sum of the group and play-off stages matches in WC2010 for each team per zone.

\ 0.017) (Table 4).

5.9 6 0.7c
3.0 6 1.0
18.0 6 3.1
4.0 6 1.1
10.3 6 2.6
13.3 6 4.7
9.7 6 2.5
3.7 6 1.3
5.0 6 1.6
3.0 6 1.0
0.3 6 0.3
0.0 6 0.2
0.0 6 0.2

In addition, the results showed that defensive beha-

Group

vior followed by a pass was positively associated with

stage

the efficacy of the attack (p \ 0.004) (Table 5). In

Netherlands

opposition, no significant association was found

2.8 6 0.7
19.8 6 2.2
3.2 6 0.8
7.2 6 1.8
18.3 6 3.3
7.8 6 1.8
4.0 6 0.9
3.7 6 1.2
3.2 6 0.6
0.3 6 0.2
0.0 6 0.1
0.2 6 0.1
a
WC2010

between indirect patterns of ball recovery and the final

attacking events with efficacy (p \ 0.403) (Table 5).
–
3.6 6 0.7c
2.0 6 1.0
11.7 6 3.1
1.7 6 1.1
3.7 6 2.6
15.7 6 4.7
3.3 6 2.5
2.0 6 1.3
2.3 6 1.6
1.0 6 0.9
0.0 6 0.3
0.0 6 0.2
0.0 6 0.2

Results are shown by stage and for the competition as a whole. Values are mean 6 SD.

Discussion
Play-off
stage

The main purpose of this investigation was to examine

the patterns of ball recovery and attacking play accord-
6.1 6 0.7c
3.0 6 1.0
21.3 6 3.1
4.7 6 1.1
8.3 6 2.6
20.7 6 4.7
7.7 6 2.5
1.0 6 1.3
3.3 6 1.6
2.7 6 0.9
0.3 6 0.3
0.0 6 0.2
0.0 6 0.2

ing to competition stages, successful teams and pitch

Group
stage

zones during the 2010 FIFA World Cup. Indeed, this

study provided an overview of the patterns of ball
recovery as a performance indicator with potential to
2.5 6 0.7
16.5 6 2.2
3.2 6 0.8
6.0 6 1.8
18.2 6 3.3
5.5 6 1.8
1.5 6 0.9
2.8 6 1.1
1.8 6 0.6
0.2 6 0.2
0.0 6 0.1
0.0 6 0.1
a
WC2010
Germany

predict efficacy of the subsequent attack in elite soccer.

WC2010: FIFA World Cup 2010; SD: standard deviation.

In the 2010 FIFA World Cup, a mean number of

–

67.5 attacks per team were performed, as compared

5.1 6 0.4b

with 54.0 attacks in the 2008 Euro Cup held in

2.4 6 0.5
18.3 6 1.6
2.3 6 0.6
5.5 6 1.3
19.3 6 2.3
4.5 6 1.2
2.9 6 0.7
3.3 6 0.8
2.3 6 0.5
0.3 6 0.2
0.0 6 0.1
0.2 6 0.1

Switzerland and Austria.32 In an analysis of attacks

Play-off
stage

that finished close to goal 26.3 attacks per match per

team in the 2002 FIFA World Cup in Korea and Japan
6.2 6 0.4b

were observed;39 similar results were found in the 1998

1.9 6 0.5
18.8 6 1.6
3.4 6 0.6
9.7 6 1.3
18.5 6 2.3
10.0 6 1.2
2.8 6 0.7
4.9 6 0.8
3.4 6 0.5
0.2 6 0.2
0.3 6 0.1
0.1 6 0.1

FIFA World Cup40 and in UEFA Euro 2000.41 Using

Group
stage

identical methods to the current study, investigation42

added that in the PO rounds of the 2010 World Cup,
2.2 6 0.3
18.5 6 1.1
2.9 6 0.4
7.6 6 0.9
18.9 6 1.6
7.3 6 0.9
2.8 6 0.5
4.1 6 0.6
2.9 6 0.3
0.2 6 0.1
0.1 6 0.1
0.1 6 0.1
WC 2010

South American teams performed fewer attacks per

match (n = 59) than did European ones (n = 83). It is
–

worth noting, however, that some European teams,

such as France, The Netherlands and Spain, have
Pitch
zone

10
11
12

nowadays blended the two styles of play into a more

1
2
3
4
5
6
7
8
9

b
c
a
68 Proc IMechE Part P: J Sports Engineering and Technology 228(1)

Table 4. Association between patterns of ball recovery and the final attacking events.

Patterns of BR Final attacking events Odds ratio 95% CI p

BRi Shot wide 1.068 0.624–1.827 0.811

Shot on target 1.577 0.673–3.694 0.294
Shot stopped, with no maintenance of ball possession 0.657 0.163–2.647 0.555
Goal 1.752 0.602–5.099 0.303
BRt Shot wide 1.376 0.744–2.543 0.309
Shot on target 1.036 0.320–3.347 0.953
Shot stopped, with no maintenance of ball possession 0.863 0.172–4.319 0.858
Goala 3.021 1.001–9.119 0.050
BRgk Shot wide 0.956 0.445–2.055 0.908
Shot on target 1.020 0.276–3.760 0.977
Shot stopped, with no maintenance of ball possession 0.566 0.068–4.748 0.600
Goal 2.266 0.630–8.152 0.211
BRp Shot wide 1.524 0.932–2.490 0.093
Shot on targetb 2.566 1.186–5.555 0.017
Shot stopped, with no maintenance of ball possession 1.283 0.410–4.014 0.668
Goal 1.925 0.678–5.460 0.218

BR: ball recovery; BRi: ball recovery by interception; BRt: ball recovery by tackle; BRgk: ball recovery by intervention of the goalkeeper in the
defensive phase; BRp: ball recovery by a defensive behavior followed by a pass; CI: confidence interval.
a
Significant association between ball recovery by tackle and goal scoring event; p4 0.05, according to multinomial logistic regression.
b
Significant association between ball recovery by a defensive behavior followed by a pass and shot on target event; p4 0.05 according to multinomial
logistic regression.

Table 5. Association between patterns of ball recovery and the within the midfield and attacking areas, reducing the
efficacy of the attack. chance of losing possession, that is, both styles were
blended into a patient passing strategy named as pos-
Patterns of Attack with efficacy session football style.45 Furthermore, a 2002 FIFA
ball recovery World Cup study found that Europeans performed
Odds ratio 95% CI p
significantly more dribbling sequences in the midfield
BRia 1 – – third and more possession techniques in offensive
BRt 1.455 0.983–2.154 0.061 areas than South Americans, who carried out direct
BRgk 0.897 0.498–1.616 0.718 attacks in midfield areas and possession techniques in
BRpb 2.788 1.055–4.100 0.004 advanced offensive areas.46 Although, probably due to
IndBR 0.839 0.555–1.266 0.403
Europe has become the most successful venue for
BRi: ball recovery by interception; BRt: ball recovery by tackle; BRgk: professional soccer players,47 the European style in
ball recovery by intervention of the goalkeeper in the defensive phase; general became hang back less on defense and attack
BRp: ball recovery by a defensive behavior followed by a pass; IndBR:
more on offense,43 with both styles—European and
indirect ball recovery; CI: confidence interval.
a
Reference category. South American—revealing nowadays effective simila-
b
Significant association between ball recovery by tackle and attack with rities. Also, comparing both halves of the 2010
efficacy; p4 0.05 according to binomial logistic regression. World Cup matches, a study33 found that teams per-
formed more attacks during the first half (485 vs 471 in
attacking or modern European style, with both being the second half), although this difference was not
equally efficient in World Cup tournaments.43 significant.
In other words, top South American and European In the current study, no significant differences were
national teams, according to FIFA ranking, were char- found when comparing the two stages of the 2010
acterized by performing different styles of play, that is, World Cup and the semi-finalist teams with regard to
long ball possessions with predominance to occurrence the number of attacks performed per match and in the
of short passes, with greater relevance to goal scoring whole competition. This suggests that successful teams
in South American teams and, contrarily, shorter dura- maintain an attacking rate independent of their playing
tion possessions with direct long passes and counter- style and competition stage, which probably is likely to
attack patterns of play to European teams.43 However, be a feature of success, even though it does not predict
playing styles found in the analysis of France and the ultimate winner of the competition. Contrarily, in
Brazil do not corroborate this idea, with both teams the 2006 FIFA World Cup, it was found that competi-
performing multiple sequences of possession, hence tion stage variable predicted the winner across the
identical and elaborate patterns of play.44 These find- group stage matches; however, performance of winners
ings were confirmed by 1990 World Cup analysis was worse than the losers during PO rounds.6 This is
results, in which Europeans and South Americans com- justified by the characteristics of PO stage, namely, the
pleted an extent number of passes, runs and dribbles knock-out system and the lower distance in FIFA
Barreira et al. 69

ranking between the opposing teams. The overall con- was concluded that the interruption of an attack by
clusion to be drawn from our results is that the number means of a defensive foul enhances rather than disrupts
of attacks is independent of the competition stage and the probability of conceding a goal.58
the considered teams. In the present study of the 2010 FIFA World Cup,
Also, our analysis showed a predominance of direct indirect ball recovery patterns were not influenced by
ball recovery during the 2010 FIFA World Cup (direct the change in competition stage. However, ball recov-
ball recovery: 77.3% vs indirect ball recovery: 22.3%; ery by interceptions and tackles was more common in
Table 3), corroborating the 2002 World Cup findings.48 PO matches than during the group stage (Figure 4).
Thus, open play predominated over set-plays, and con- The single exception to this was the case of ball recov-
sequently, the flow of the game was generally preserved. ery by a defensive behavior followed by a pass, which
Moreover, categorizing the types of ball recovery into was more frequent during group stage matches, while
(a) interception or tackle, (b) set-plays and (c) error by ball recovery by intervention of the goalkeeper was
the opponent was found that 54% of ball recoveries similarly present in both stages of the competition
occurred due to a wrong pass or poor ball control by (Figure 4). In general, our findings confirm studies59,60
the opponent,49 while other study found that 33% of that show relevant differences concerning direct or indi-
attacks ending with a shot were preceded by an error rect ball possession recovering19 and confirm the influ-
on the part of the opponent.34 ence of competition stages in teams’ performance.6
Regarding indirect ball recoveries (i.e. set-plays), it Overall, the four successful teams in the 2010 World
was found that goalkeeper intervention (BRgki: Cup showed similar patterns of ball recovery, excepting
7.5 6 4.2) and throw-ins (BRti: 6.8 6 2.3) were the most for regaining of possession by throw-in, which was
commonly used behaviors (Table 2). The least frequent achieved significantly more often by Spain than by
kinds of indirect ball recovery were the dropped ball Germany (Figure 5).
and the corner kick, neither of which occurred during The results regarding ball recovery according to
the 24 matches. Specifically, we observed an average of pitch zones were similar to those reported to 2008 Euro
4.3 goalkeeper interventions per match, which is more Cup.32 Thus, teams tended to regain possession directly
than that reported in a study of the 1998 World Cup in the central strip of the first and second quarters of
(only one ball possession starting with the goal- the pitch (defensive midfield) and less often in mid-
keeper).50 However, on English Premier League 2007– offensive pitch zones (third quarter). The attacking
2008, goalkeepers intervened more times per match quarter was not used to recover the ball. Our results
(13.4) due to attack distribution tasks.51 Thus, results corroborate the findings of the study of Norwegian
confirmed that ball possession by goalkeepers has men’s professional league during one competitive sea-
increased over the years, as shown in outcomes of the son,18 showing only 2% of the attacks starting on the
analysis of the 1953, 1974, 1998 and 2002 World final third of the pitch and a balance between the first
Cups,52 and that goalkeeper role also varies according and middle thirds (53% and 45%, respectively). In the
to the level of competition.53 In this sense, due to goal- same line, the analysis of the 708 attacks considered
keeper’s dependence on what and where attackers per- from the Spanish La Liga 2009–2010, excluding posses-
form,54 teams’ attacking style of play needs to be sions starting with a set-play, showed that 34%, 43%,
considered to understand the goalkeeper’s intervention 22% and 1% started from defensive, mid-defensive,
during the matches. In this sense, the literature11,55,56 mid-offensive and offensive pitch zones, respectively.17
shows that the number of passes per attack and the Furthermore, in a study among top-level European
maintenance of ball for longer durations—considered teams, it was found that attacks started on the midfield
as indicators of success—have increased in the last central zone were the most widely used to regain pos-
years, probably enhancing the goalkeeper participation. session,21 probably because these attacks can be done
Our findings also corroborate the results of the while maintaining an ideal defensive formation.34
19908 and 200257 World Cup studies, with both sets of During the 2010 FIFA World Cup, semi-finalist
authors stating that in elite soccer approximately one- teams mostly used the direct ball recovery, with tackle
third of attacks start from a set-play. Additionally, it and pass after defensive behavior inducing attacking
was found that a similar number of goals were scored play efficacy. Regarding the dynamic nature of the soc-
from direct or indirect set-plays,8 concluding that open cer game, after direct ball recovery and inversely after
play showed greater efficacy accounting for most of the ball lost—state-transition moments27—the opposing
goals scored in the 1998 World Cup,58 in which goals teams compete to get advantage in time and space,
recorded from set-plays (i.e. free kicks, corner kicks and developing individual and collective behaviors to
long throw-ins) represented 25% of all goals scored. improve their own organization levels and to benefit
A study of the 1998 World Cup50 concluded that the from the opponent’s unbalance. Transition moments
most common way of regaining possession was control- evidence chaotic behaviors, with no organization; nev-
ling a free ball (29%), followed by a restart (19%), ertheless, the literature uncovers the importance of
tackle (11%), throw-in (11%) and an interception guiding the training process to improve the attitude-
(10%), with the respective relationship to the goals changing moments of performance and, consequently,
scored being 38%, 18%, 18%, 0% and 18%. Also, it the efficacy of the subsequent attacking play. Thus,
70 Proc IMechE Part P: J Sports Engineering and Technology 228(1)

according to our results, it seems essential to keep up References

the collective defensive balance in mid-defensive zones, 1. Lago C. The influence of match location, quality of
increasing pressure in opponent players in the central opposition, and match status on possession strategies in
strip, to warrant resourceful ball recovering and attack- professional association football. J Sport Sci 2009; 27:
ing efficacy. 1463–1469.
2. Ruiz-Ruiz C, Fradua L, Fernandez-Garcia A, et al.
Analysis of entries into the penalty area as a performance
Conclusion indicator in soccer. Eur J Sport Sci 2011; 1–8. DOI:
This study shows that the type and the zone of ball 10.1080/17461391.2011.606834
recovering seem to affect attacking efficacy in elite soc- 3. Lago C and Lago-Ballesteros J. Game location and team
quality effects on performance profiles in professional
cer. Also, it was emphasized that the competition stage
soccer. J Sport Sci Med 2011; 10: 465–471.
and the team should be taken into consideration. 4. Taylor J, Mellalieu S, James N, et al. The influence of
Consequently, pertinent data for coaches were provided match location, quality of opposition, and match status
to better organize the training process and improve per- on technical performance in professional association
formance. In the 2010 World Cup, Spain was the team football. J Sport Sci 2008; 26: 885–895.
that carried out the highest number of attacks, while no 5. Fradua L, Zubillaga A, Caro O, et al. Designing small-
significant differences were detected between the four sided games for training tactical aspects in soccer: extra-
successful teams during the tournament. polating pitch sizes from full-size professional matches.
Overall, direct ball recovery was more widely used J Sport Sci 2012; 1–9. DOI: 10.1080/02640414.2012.
than indirect ball recovery. Patterns of indirect ball 746722
recovery did not seem to be influenced by changing of 6. Lago C. Are winners different from losers? Performance
competition stage. Interceptions and tackles were the and chance in the FIFA World Cup Germany 2006. Int J
Perform Anal Sport 2007; 7: 36–47.
most frequent defensive behaviors in PO matches, while
7. Hughes M and Bartlett R. The use of performance indi-
ball recovery by a defensive behavior followed by a
cators in performance analysis. J Sport Sci 2002; 20:
pass was the most common during the group stage. 739–754.
The most successful teams showed similar patterns 8. Jinshan X, Xiakone C, Yakamaka K, et al. Analysis of
of ball recovery, with the exception of throw-ins that the goals in the 14th World Cup. In: Reilly T, Clarys J
differentiated Germany from Spain. Ball recovery and Stibbe A (eds) Science and football II. Eindhoven: E
behaviors were not regularly performed in the defensive & FN Spon, 1993, pp.203–205.
lateral wings or in the offensive sector of the pitch, indi- 9. Ali A, Williams C, Hulse M, et al. Reliability and
cating that successful teams normally recover the ball validity of two tests of soccer skill. J Sport Sci 2007; 25:
in the central strip of the defensive and mid-defensive 1461–1470.
sectors. Patterns of direct ball recovery increase attack- 10. Hughes M and Franks I. Analysis of passing sequences,
ing play efficacy, namely through ball recover by tackle shots and goals in soccer. J Sport Sci 2005; 23: 509–514.
11. Lago C, Lago-Ballesteros J, Dellal A, et al. Game-related
and by a defensive behavior followed by a pass related
statistics that discriminated winning, drawing and losing
to goal scoring and shots on target, respectively. The
teams from the Spanish soccer league. J Sport Sci Med
zones and the way the ball is recovered in elite soccer 2010; 9: 288–293.
are important tactical determinants that coaches should 12. Jones P, James N and Mellalieu S. Possession as a perfor-
consider when preparing the training process. It seems mance indicator in soccer. Int J Perform Anal Sport 2004;
fundamental to train specifically the collective defensive 4: 98–102.
organization and simultaneously the state-transition 13. Brillinger D. A potential function approach to the flow
moments, rather than the set-plays. of play in Soccer. J Quant Anal Sports 2007; 3: 1–19
(article 3).
14. Garganta J, Marques A and Maia J. Modelac xão táctica
Declaration of conflicting interests do jogo de Futebol: estudo da organizac xão da fase ofen-
The authors declare that there is no conflict of interest. siva em equipas de alto rendimento. In: Garganta J,
Suarez A and Lago C (eds) A Investigacxão em Futebol:
Estudos Ibéricos. Porto: Faculdade de Ciências do
Funding Desporto e de Educac xão Fı́sica, 2002, pp.51–66.
The lead author was supported by Fundac xão 15. Bayer C. L’Enseignement des jeux sportifs collectifs. Paris:
Vigot, 1986.
Portuguesa para a Ciência e a Tecnologia (grant
16. Franks I and Miller G. Training coaches to observe and
SFRH/BD/48558/2008). This study was supported by
remember. J Sport Sci 1991; 9: 285–297.
the Spanish government project Observación de la 17. Lago-Ballesteros J, Lago C and Rey E. The effect of play-
interacción en deporte y actividad fı́sica: Avances técni- ing tactics and situational variables on achieving score-
cos y metodológicos en registros automatizados box possessions in a professional soccer team. J Sport Sci
cualitativos-cuantitativos (Secretarı́a de Estado de 2012; 30: 1455–1461.
Investigación, Desarrollo e Innovación del Ministerio 18. Tenga A, Holme I, Ronglan LT, et al. Effect of playing
de Educación y Ciencia) (grant DEP2012-32124). tactics on achieving score-box possessions in a random
Barreira et al. 71

series of team possessions from Norwegian professional Drust B and Dawson B (eds) Science and football VII: the
soccer matches. J Sport Sci 2010; 28: 245–255. proceedings of the seventh world congress on science and
19. Pollard R and Reep C. Measuring the effectiveness of football. London and New York: Routledge, 2013,
playing strategies at soccer. Statistician 1997; 46: 541–550. pp.193–198.
20. Castelo J. Futebol: a organizac xão do jogo. Como entender 34. Garganta J. Modelacxão táctica do jogo de Futebol. Estudo
a organizac xão dinâmica de uma equipa de futebol e a partir da organizacxão da fase ofensiva em equipas de alto rendi-
desta compreensão como melhorar o rendimento e a mento [Tactical modelling of Soccer game]. Porto: Facul-
direccxão dos jogadores e da equipa. Lisboa: Edicxão do dade de Ciências do Desporto e de Educac xão Fı́sica,
autor, 1996. Universidade do Porto, 1997.
21. Garganta J, Maia J and Basto F. Analysis of goal-scoring 35. Castellano J. Observación y análisis de la acción de juego
patterns among top-level European soccer teams. J Sport en el fútbol. Vitoria-Gasteiz: Departamento de Teorı́a e
Sci 1995; 13: 513–514. Historia de la Educación, Universidad del Paı́s Vasco,
22. Reina E, Abad P and Losa J. La velocidad en el juego de 2000, p.680 + 7 anexos.
ataque: análisis táctico de los goles de la Eurocopa 96 de 36. Perea A, Castellano J, Hernández-Mendo A, et al. Pautas
Fútbol. Fútbol: Cuadernos Técnicos 1997; 8: 36–43. para el análisis de la calidad del dato en la observación de
23. Gréhaigne J. La organización del juego en el fútbol. Barce- los deportes colectivos: una aplicación en el fútbol. I Con-
lona: INDE, 2001. gresso virtual de investigación en la actividad fı´sica y el
24. Carling C, Williams M and Reilly T. Handbook of soccer deporte. Vitoria-Gasteiz, 2005.
match analysis. London: Routledge, 2005. 37. Cohen J. A coefficient of agreement for nominal scales.
25. Gréhaigne J, Marchall D and Duprat E. Regaining pos- Educ Psychol Meas 1960; 20: 37–46.
session of the ball in the defensive area in soccer. In: 38. Bakeman R and Gottman JM. Observing interaction: an
Spinks W, Reilly T and Murphy A (eds) Science and foot- introduction to sequential analysis. 2nd ed. Cambridge:
ball IV. London: Routledge, 2002, pp.112–120. Cambridge University Press, 1997.
26. Barreira D. Transic xão defesa-ataque em Futebol. Análise 39. De Baranda P, Ortega E and Palao J. Analysis of goal-
Sequencial de padrões de jogo relativos ao Campeonato keepers’ defence in the World Cup in Korea and Japan in
Portugueˆs 2004/05. Tese de Licenciatura. Porto: 2002. Eur J Sport Sci 2008; 8: 127–134.
Faculdade de Desporto da Universidade do Porto, 2006, 40. Ruiz D and Sáinz P. Análisis del portero de fútbol en el
p.231. mundial de Francia-98. Training Fútbol: Revista técnica
27. Barreira D and Garganta J. Padrão sequencial da tran- profesional, 2000, pp. 24–41 November, 2000.
xão defesa-ataque em jogos de Futebol do Campeonato
sic 41. De Baranda P and Ortega E. Estudio comparativo de las
Português 2004/2005. In: 1º Congresso Internacional de acciones realizadas por los porteros de fútbol partici-
Jogos Desportivos (eds F Tavares, A Gracxa and J Gar- pantes en el Mundial de Francia 98 versus Eurocopa
ganta), 2007. Porto: Faculdade de Desporto da Universi- del 2000. Lecturas: Educación Fı´sica y Deportes 2002; 8.
dade do Porto (in CD-ROM) Porto 12–14th July 2007. Lecturas: Educación Fisica y Deportes. Revista Digital,
28. Bakeman R and Quera V. Analyzing interaction: sequen- 24 de Agosto. http://www.efdeportes.com/efd49/francia.
tial analysis with SDIS and GSEQ. Cambridge: Cam- htm. (accessed on August 10, 2012)
bridge University Press, 1995. 42. Barreira D, Garganta J, Guimarães P, et al. Attacking
29. Gutiérrez-Santiago A, Prieto I, Camerino O, et al. game-patterns differences between South American and
Sequences of errors in the judo throw Morote Seoi Nage European national Soccer teams in the World Cup 2010.
and their relationship to the learning process. Proc In: 7th world congress on science & football 2011 & 9th
IMechE, Part P: J Sports Engineering and Technology annual conference of Japanese Society of Science and Foot-
2011; 227: 57–63. ball 2011, 2011, p.259. Nagoya, Japan: Japanese Society
30. Anguera MT, Blanco A, Hernández-Mendo A, et al. Dis- of Science and Football. Nagoya, Japan May 26–30 2011.
eños observacionales: ajuste y aplicación en psicologı́a 43. Skirka N. Finding meaning in the World Cup’s results.
del deporte. Cuadernos de Psicologı´a del Deporte 2011; Soccer J 2010; 55: 63–64.
11: 63–76. 44. Pollard R, Reep C and Hartley S. The quantitative com-
31. Hughes M, Robertson K and Nicholson A. Comparison parison of playing styles in soccer. In: Reilly T, Lees A,
of patterns of play of successful and unsuccessful teams Davids K, et al. (eds) Science and football. Liverpool: E
in the 1986 World Cup for soccer. In: Reilly T, Lees A, & FN Spon, 1988, pp.309–315.
Davids K, et al. (eds) Science and football. Liverpool: E 45. Yamanaka K, Hughes M and Lott M. An analysis of
& FN Spon, 1988, pp.363–367. playing patterns in the 1990 World Cup for association
32. Barreira D, Garganta J and Anguera MT. In search of football. In: Reilly T, Clarys J and Stibbe A (eds) Science
nexus between attacking game-patterns, match status and and football II. Eindhoven: E & FN Spon, 1993,
type of ball recovery in European Soccer Championship. pp.206–214.
In: Research methods and performance analysis 5th inter- 46. Brown S and Hughes M. The attacking playing patterns
national Christmas sport scientific conference: qualitative in offensive areas of European, South American, African
and quantitative research in sport science (eds M Hughes, and Asian teams in the 2002 World Cup for association
H Dancs, K Nagyváradi, et al.), Szombathely, Hungary, football. In: O’Donoghue P and Hughes M (eds) Perfor-
2011, pp.226–237 Published by University of West-Hun- mance analysis of sport VI. Cardiff: Centre for Perfor-
gary, Institute of Sport Science, Szombathely 12nd–14th mance Analysis, 2004, pp.99–102.
December 2010. 47. Goldblatt D. The ball is round: a global history of soccer.
33. Barreira D, Garganta J, Pinto T, et al. Do attacking game New York: Riverhead Books, 2006.
patterns differ between first and second halves of soccer 48. Silva A, Sánchez-Bañuelos F, Garganta J, et al. Patrones de
matches in the 2010 FIFA World Cup? In: Nunome H, juego en el fútbol de alto rendimiento. Análisis secuencial del
72 Proc IMechE Part P: J Sports Engineering and Technology 228(1)

proceso ofensivo en el campeonato del mundo Corea-Japón 55. James N, Jones P and Mellalieu S. Possession as a perfor-
2002. Cultura, Ciencia y Deporte 2005; 1: 65–72. mance indicator in soccer as a function of successful and
49. Joaquim T. Caracterı´sticas da posse de bola e do processo unsuccessful teams. J Sport Sci 2004; 22: 507–508.
ofensivo nos Escalões de Formac xão em Futebol. Porto: 56. Hook C and Hughes M. Patterns of play leading to shots
Faculdade de Desporto da Universidade do Porto, 2009. in Euro 2000. In: CPA (ed.) Passcom. Cardiff: UWIC,
50. Kirkendall D, Dowd WW and DiCicco T. Patterns of 2001, pp.295–302.
successful attacks: a comparison of men’s and women’s 57. Yiannakos A and Armatas V. Evaluation of the goal
games in World Cups. Int J Soccer Sci 2002; 1: 29–36. scoring patterns in European Championship in Portugal
51. Oberstone J. Comparing English premier league goal- 2004. Int J Perform Anal Sport 2006; 6: 178–188.
keepers: identifying the pitch actions that differentiate 58. Grant A, Williams M, Reilly T, et al. Analysis of the suc-
the best from the rest. J Quant Anal Sports 2010; 6 pp. cessful and unsuccessful teams in the 1998 World Cup.
1–17. Insight (FA Coaches) 1998; 2: 21–24.
52. Kuhn T. Changes in professional soccer: a qualitative 59. Castellano J, Masach J and Zubillaga A. Cuantificación
and quantitative study. In: Reilly T, Cabri J and Araújo del esfuerzo fı́sico del jugador de fútbol en competición.
D (eds) Science and football V. London: E & FN Spon, Training Fútbol: Revista técnica professional, 1996, vol. 7,
2005, pp.179–193. pp.27–42 September 1996.
53. Seaton M and Campos J. Distribution competence of a 60. Castellano J and Zubillaga A. Análisis de los goles Mun-
football clubs goalkeepers. Int J Perform Anal Sport dial USA’94 (2ª parte). El Entrenador Español 1995; 65:
2011; 11: 314–324. 46–58.
54. Bode G. World Cup 2010: giving away the goals. Success
Soccer 2011; 14: 6–10.

View publication stats