Full Terms & Conditions of access and use can be found athttp://www.tandfonline.com/action/journalInformation?journalCode=rjsp20

Download by: [UNIVERSITAT DE BARCELONA] Date: 17 March 2016, At: 11:25

Journal of Sports Sciences

ISSN: 0264-0414 (Print) 1466-447X (Online) Journal homepage: http://www.tandfonline.com/loi/rjsp20

Quantifying the offensive sequences that result ingoals in elite futsal matches

Hugo Sarmento, Paul Bradley, M. Teresa Anguera, Tiago Polido, Rui Resende& Jorge Campaniço

To cite this article: Hugo Sarmento, Paul Bradley, M. Teresa Anguera, Tiago Polido,Rui Resende & Jorge Campaniço (2016) Quantifying the offensive sequences thatresult in goals in elite futsal matches, Journal of Sports Sciences, 34:7, 621-629, DOI:10.1080/02640414.2015.1066024

To link to this article: http://dx.doi.org/10.1080/02640414.2015.1066024

Published online: 17 Jul 2015.

Submit your article to this journal

Article views: 200

View related articles

View Crossmark data

Quantifying the offensive sequences that result in goals in elite futsal matchesHugo Sarmento1,2, Paul Bradley3, M. Teresa Anguera4, Tiago Polido5, Rui Resende6 and Jorge Campaniço5

1Centre for the Study of Education, Technologies and Health, CI&DETS, School of Education, Polytechnic of Viseu, Viseu, Portugal; 2Research Centerin Sports Sciences, Health Sciences and Human Development, CIDESD, University Institute of Maia, ISMAI, Maia, Portugal; 3Carnegie School ofSport, Leeds Metropolitan University, Leeds, UK; 4Institute for Research on the Brain, Cognition and Behaviour (IR3C), Faculty of Psychology,University of Barcelona, Barcelona, Spain; 5Department of Sport Sciences, University of Trás-os-Montes e Alto Douro, Vila Real, Portugal; 6ResearchCenter in Sports Sciences, Health Sciences and Human Development, CIDESD, University Institute of Maia, ISMAI, Maia, Portugal

ABSTRACTThe aim of this study was to quantify the type of offensive sequences that result in goals in elite futsal.Thirty competitive games in the Spanish Primera Division de Sala were analysed using computerisednotation analysis for patterns of play that resulted in goals. More goals were scored in positional attack(42%) and from set pieces (27%) compared to other activities. The number of defence to offense“transitions” (n = 45) and the start of offensive plays due to the rules of the game (n = 45) were themost common type of sequences that resulted in goals compared to other patterns of play. The centraloffensive zonal areas were the most common for shots on goal, with 73% of all goals scored from theseareas of the pitch compared to defensive and wide zones. The foot was the main part of the bodyinvolved in scoring (n = 114). T-pattern analysis of offensive sequences revealed regular patterns of play,which are common in goal scoring opportunities in futsal and are typical movement patterns in thissport. The data demonstrate common offensive sequences and movement patterns related to goals inelite futsal and this could provide important information for the development of physical and technicaltraining drills that replicate important game situations.

ARTICLE HISTORYAccepted 22 June 2015

KEYWORDSIndoor soccer; matchanalysis; patterns of play

1. Introduction

Futsal is the indoor 5-a-side equivalent of soccer and wasdeveloped in the 1930s (Barbero-Alvarez, Soto, Barbero-Alvarez, & Granda-Vera, 2008). More recently, FédérationInternationale de Football Association (FIFA) and Union ofEuropean Football Associations (UEFA) have standardisedthe rules of the game, thus allowing both domestic andinternational competitions to be popularised with an expo-nential increase in the number of participants involved in thegame.

Despite its increasing popularity in the last 10 years, thereis scant scientific evidence on the sport compared to otherfootballing codes (Duarte, Batalha, Folgado, & Sampaio, 2009).Some researchers have examined the physiological demands(Dittrich, Silva, Castagna, Lucas, & Guglielmo, 2011; Oliveira,Leicht, Bishop, Barbero-Alvarez, & Nakamura, 2013; Schultz deArruda, De Freitas, De Moura, Aoki, & Moreira, 2013) and thephysical characteristics of elite and sub-elite futsal players(Castagna, D’Ottavio, Granda Vera, & Barbero Alvarez, 2009;Dogramaci, Watsford, & Murphy, 2011). Although less workhas been carried out on the match activities of futsal players,particularly the quantification of fundamental movement pat-terns of players during offensive sequences of the play(Jovanovic, Sporis, & Milanovic, 2011; Lapresa, Álvarez, Arana,Garzón, & Caballero, 2013; Travassos, Araújo, Vilar, & McGarry,2011).

A recent systematic review of the literature pertaining tomatch analysis in the footballing codes identified thatprevious work has focused mainly on the description of phy-siological aspects of match-play (Sarmento et al., 2014). Acurrent challenge in this area of research involves creatingsuitable movement sequences that can clearly identify andcategorise individuals and behaviours over time. A valid pre-dictor of sporting behaviour involves examining tactical stra-tegies of individuals within a team, with the aim to identifycommon movement patterns during particular sequences(James, 2012).

Futsal is a game with a random intermittent nature, wherebycritical elements of the game are sometimes determined bychance. Thus, the training process should aim to develop regularindividual/team behaviour sequences. Attempting to predictfuture performance on the basis of previous performances is achallenging task yet important for match analysts (James, 2012).Typically, the basis for any prediction model is that performanceis repeatable, to some degree. In other words, events that havepreviously occurredwill occur again in some predictablemanner.This type of prediction is based on the principle that any perfor-mance is a consequence of factors like prior learning, inherentskills and situational variables. Although this is a challenge giventhat movement patterns and technical performance indicatorsvary substantially from game to game in most footballing codes(Van Winckel et al., 2014).

CONTACT Hugo Miguel Borges Sarmento hg.sarmento@gmail.com Centre for the Study of Education, Technologies and Health, CI&DETS, School ofEducation – Polytechnic of Viseu, Viseu, Portugal.

JOURNAL OF SPORTS SCIENCES, 2016VOL. 34, NO. 7, 621–629http://dx.doi.org/10.1080/02640414.2015.1066024

© 2015 Taylor & Francis

Dow

nloa

ded

by [

UN

IVE

RSI

TA

T D

E B

AR

CE

LO

NA

] at

11:

25 1

7 M

arch

201

6

Researchers in this area should adopt a multifactorialapproach when employing statistical analyses, thus improvingthe ability to find associations between variables and theeffect of different interactions (Sarmento et al., 2014). Onepromising area of research that has been recently employedat investigating common movement patterns in team sports isT-pattern analysis (Camerino, Chaverri, Anguera, & Jonsson,2012; Lapresa, Álvarez, et al., 2013; Lapresa, Anguera,Alsasua, Arana, & Garzón, 2013; Sarmento et al., 2014). Afterstudying the existing methods and software and running intotheir limitations regarding the analysis of naturally occurringbehaviour as complex real-time processes, Magnusson (1978)set out to develop new structural concepts and tools and inparticular for the discovery of hidden patterns of behaviour.

The detection algorithm used in the software Theme allowsthe detection of repeated temporal and sequential structures inreal-time behaviour records that cannot be fully detectedthrough unaided observation or with the help of statisticalmethods (Borrie, Jonsson, & Magnusson, 2002). A temporalpattern is essentially a combination of events, which occur, inthe same order with temporal distances between each other,which remain relatively invariant in relation to the null hypoth-esis that each component is independent and is distributedrandomly in time (for review, see Casarrubea et al. (2014)). Asstated by Magnusson (2000, p. 94), “that is, if A is an earlier andB a later component of the same recurring temporal patternthen after an occurrence of A at t, there is an interval [t+d1, t+d2] (d2≥d1≥d0) that tends to contain at least one occurrenceof B more often than would be expected by chance”.

This software has been extensively used in several areas ofknowledge both in animal (Feenders & Bateson, 2012; Nicol,Segonds-Pichon, & Magnusson, 2015) and human studies(Sandman, Kemp, Mabini, Pincus, & Magnusson, 2012;Woods, Yefimova, & Brecht, 2014).

Although no research exists that has examined T-patternanalysis of elite futsal match play during sequences that resultin goals. Taking into account that goal scoring is the ultimateobjective measure of offensive effectiveness in futsal matchplay, the aim of the present study was to quantify the type ofoffensive sequences that result in goals in elite futsal.

2. Methods

2.1. Players

Data were collected from a single Spanish team in the PrimeraDivisión de Sala and consisted of 30 games and 17 individualplayers. The team qualified to the championship playoffs.Original data files included 126 goals scored. Ethical approvalwas granted from the appropriate institutional ethics committee.

2.2. Data coding system

The present study used an observational instrument that has acombination of a field format and a system of categories ascriteria (Anguera, 2003; Anguera, Magnusson, & Jonsson,2007); from a validated list of the most important futsal activ-ities. This observational instrument enabled all notationalevents to be coded effectively. The finalised list of notational

activities was selected based on an extensive review of futsalliterature and by consulting a panel of experts in the area(coaches and researchers in the game). Based on the sugges-tions from the panel, the following observational instrumentwas developed (Table I). This instrument included the follow-ing categories that were used to code notational events: (1)start of the offensive process; (2) phases of the offensiveprocess; (3) development of the offensive process; (4) spatialcharacterisation of the field (Figure 1); (5) body part thatperformed the shot on goal; (6) number of the player. Thevalidated coding instrument mentioned above was used tocode all futsal activities in each game and data were subse-quently placed in Microsoft Excel. The reliability of the datacollected was subjected to intra- and inter-observeragreement analysis using Cohen’s kappa (Cohen, 1960).Values >0.90 were achieved for all criteria.

2.3. Data analysis

Specialised software (THÈME V. 5.0) was used for the detection oftemporal patterns of play during futsal matches. This algorithm isbased on the assumption that the flow of complex humanbehaviour (e.g. sports performance) is based on the sequentialstructure as a function of time, and has a discrete nature that isnot fully detectable without the use of standardised statisticaland behavioural methods (Borrie et al., 2002).

The following criteria were used in order to detect theT-patterns during games: (1) a minimum frequency of threeoccurrences was set in each match dataset; (2) the level ofsignificance was set at P < 0.05 and (3) the T-patterns detectedwere only accepted if the software detects them among all theadditional randomly generated relationships. To reduceagainst this artefact, the software used well described algo-rithms to detect temporal patterns of importance (Magnusson,1996, 2000, 2005, 2006).

Taking into account the variability of actions that character-ise indoor soccer (like in football), the investigators in thisresearch area (e.g. Camerino et al., 2012; Lapresa, Álvarez,et al., 2013; Lapresa, Anguera, et al. 2013; Sarmento et al.,2014) have considered that an action that is repeated threetimes, established a significance level <0.05, is relevant forbeing consider as a regular behaviour. Additionally, if theresearchers failed to opt for the above criteria it could haveincreased the degree of difficulty in identifying completedT-patterns (i.e. behaviours that include the beginning, develop-ment and end of the offensive sequences).

All actions in games were registered in a systematicmanner taking into consideration the breakdown of succes-sive actions. Despite this limitation, it was still possible forthis software to quantify repetitive T-patterns of behaviour(Anguera, 2004, 2005). The most valuable contribution ofT-patterns arises from the possibility of detecting particulartypes of temporal structures (Borrie et al., 2002). Given thatpatterns facilitate the detection of hidden structures, theyare of significant importance in the analysis of a futsalgame. This type of analysis allows the representation of aspecific movement pattern which corresponds to theactions that occur in that specific order. Statistical Package

622 H. SARMENTO ET AL.

Dow

nloa

ded

by [

UN

IVE

RSI

TA

T D

E B

AR

CE

LO

NA

] at

11:

25 1

7 M

arch

201

6

for the Social Sciences (SPSS) 20.0 was used to calculatedescriptive statistics.

3. Results

3.1. Temporal and zonal analysis

The greatest number of goals was scored in attacking positions(42%), from set pieces (27%) and fast counterattacks (27%)

compared to other activities such as offensive (4%) and defensive(2%) play in a “4 vs 5” situation. The central offensive zonal areaswere the most common for shots on goal (n = 92) compared toother areas, with 73% of all goals scored from these areas of thepitch compared to defensive and wide zones. When breakingdown the goals into zonal areas it seems that only 2 goals werescored from the defensive zone, 14 goals from the midfielderzone, 47 from the offensive zone and 52 goals from the ultra-offensive zone. It is noteworthy that 44 goals were scored insidethe penalty area with the offensive and ultra-offensive zonesaccounted for a total of 37 and 44 goals, respectively. From thepoints 1 (Z1) and 2 (Z2) were performed 11 goals. Finally, 18goals were scored from the side corridors (left and right) of theoffensive and ultra-offensive zones (Figure 2).

The number of defence to offense “transitions (IPOt)”(n = 45), the ball recovery possession by interception (n = 23)and the start of offensive plays due to the rules of the game(n = 45) were the most common types of sequences thatresulted in goals compared to other patterns of play. The footwas the main part of the body involved in scoring (n = 114)compared to other body parts, with the breakdown of goalsscored with the foot involving the inside edge (n = 68), instep(n = 34), toe (n = 8) and other parts of the foot (n = 4).

3.2. T-pattern analysis

The data analysis revealed the existence of 345 differentT-patterns, ranging from 1 to 5 levels and including 2–8

Table I. Summarised description of the variable criteria within the observa-tional toll.

Category Code Sub-category Code

Start of the offensiveprocess

IPO Interception IPOiDisarm IPOdGoalkeeper action IPOgrDue the rules of the game IPOijBy transition IPOt

Phases of the offensiveprocess

FJ Positional attack FJapSet-pieces FJbpFast attack FJarCounterattack “1 vs 0” FJca10Counterattack “1 vs 1” FJca11Counterattack “2 vs 0” FJca20Counterattack “2 vs 1” FJca21Counterattack “2 vs 2” FJca22Counterattack “3 vs 0” FJca30Counterattack “3 vs 1” FJca31Counterattack “3 vs 2” FJca32Counterattack “3 vs 3” FJca33Counterattack “4 vs 3” FJca43

Development of theoffensive process

DPO Pass back DPOptPass forward DPOpfConduction of the ball DPOcbReception/control DPOrcDribble “1 vs 1” DPOdrDuel DPOduLong pass performed bygoalkeeper

DPOgrlo

Short pass performed bygoalkeeper

DPOgrcu

Pass performed by thegoalkeeper with the hand

DPOmao

Pass to the second goalpost DPO2pshot DPOrShot with goal scored DPOgolself goal DPOagol

Spatial characterisation ofthe field

Z Right defensive zone ZDDLeft defensive zone ZDERight medium zone ZMDLeft medium zone ZMERight offensive central zone ZOCDLeft offensive central zone ZOCERight offensive zones ZOCDLeft offensive zones ZOERight ultra-offensive zones ZUODLeft ultra-offensive zones ZUOERight ultra-offensive centralzones

ZUOCD

Left ultra-offensive central zones ZUOCEPoint 1 Z1Point 2 Z2Point 3 Z3Point 4 left side Z4EPoint 4 right side Z4D

Body part that performedthe shot on goal

Sf Head SFcInside edge of the foot SFpiOther part of the foot SFpeToe SFpoInstep SFppOther body parst SFo

Number of player J From player with number 1 toplayer with number 15

J1 to J15

Figure 1. Spatial characteristics and pitch dimensions of a typical futsal pitch.

Figure 2. The analysis of the various pitch zones that were scored during futsalmatches.

JOURNAL OF SPORTS SCIENCES 623

Dow

nloa

ded

by [

UN

IVE

RSI

TA

T D

E B

AR

CE

LO

NA

] at

11:

25 1

7 M

arch

201

6

events. Data were selected that included complete T-patterns(n = 5), which comprised of the offensive sequence from startto its finish. T-patterns that were associated with goalsscored on at least three separate occasions were analysedin greater detail in the software and subsequentlyrepresented in Figures 3–7.

Figure 3 represents the first complete T-pattern which wasrepeated five times during sequences that ended in a goal.This figure presents a graphical representation of the offensivesequence that results in a goal, and the Theme output of thedetected pattern, composed of three boxes: (1) the box in thetop left corner of the screen highlights the hierarchical con-struction of the pattern using a dendrogram. The right-handedge details the technical action taking place within the pat-tern. The left-hand side details the structure of the pattern and

how a combination of simple patterns combines to form amore complex pattern; (2) the upper right hand box containsinformation about the frequency and real-time distribution ofevent types in the pattern. The dots represent event occur-rences and the zig-zag lines connecting the dots representpattern occurrences. Thus, the number of lines illustrates howoften the pattern occurs; (3) the bottom box contains informa-tion about the real-time structure of the offensive pattern. Itshows lines displaying the connections between events, thetimes at which they take place, and how much time passesbetween event occurrences and pattern occurrences.

This comprised of a positional attack pattern and includedsix events: (1) the pattern was initiated by a player transition-ing in the left midfield zone (ZME); (2) followed by a throughpass (DPOp) from the left midfield zone (ZME), to another

Figure 3. Graphical representation of T-pattern 1.

624 H. SARMENTO ET AL.

Dow

nloa

ded

by [

UN

IVE

RSI

TA

T D

E B

AR

CE

LO

NA

] at

11:

25 1

7 M

arch

201

6

player; (3) that player then controls the ball (DPOrc) in theright midfield zone (ZMD); (4) this player keeps moving for-ward into a more offensive position and then makes a throughpass (DPOp) from the right midfield zone (ZMD) in the direc-tion of another player; (5) this player follows the same direc-tion of the ball (DPOcb) from the same zone (ZMD) and (6) thissequence is finished with a shot from the right central zone(ZOCD) that results in a goal (DPOgol).

Figure 4 represents the second complete T-pattern whichwas repeated five times during sequences that ended in a goal.This comprised of a second type of positional attack patternand included four events: (1) players start by transitioning (IPOt)in the right midfield zone (ZMD); (2) a pass is produced (DPOp)from the right midfield zone (ZMD) to a second player; (3)followed by a through pass (DPOcb) from the left midfieldzone (ZME) in the direction of the opponent goal and (4) thissequence is finished with a shot from the left central ultra-offensive zone (ZUOCE), which results in a goal (DPOgol).

Figure 5 represents the third complete T-pattern, whichwas repeated five times during sequences that ended in agoal. This comprised of a third type of positional attack pat-tern that included four events: (1) players start by transitioning(IPOt) in the right midfield zone (ZMD); (2) a pass is produced(DPOp) from the right midfield zone (ZMD) to a second player;(3) ball is moved (DPOcb) in the direction towards the

opponent goal from the right midfield zone (ZMD) and (4)this sequence is finished with a shot from the right centraloffensive zone (ZOCD), which results in a goal (DPOgol).

Figure 6 represents the fourth complete T-pattern, whichwas repeated five times during sequences that ended in agoal. This comprised of a fourth type of positional attackpattern that included three events: (1) players start by transi-tioning (IPOt) into the left midfield zone (ZME); (2) a pass isproduced (DPOp) from the midfield zone (ZME) and (3) thissequence is finished with a shot from the ultra-offensive cen-tral zone (ZUOCD), which results in a goal (DPOgol).

Figure 7 represents the final complete T-pattern, which wasrepeated five times during sequences that ended in a goal.This comprised of an offensive sequence from a set-piece thatincluded two events: (1) an infringement has occurred and theteam in possession have a free kick 10 m from goal (IPOj) and(2) a shot from the free kick (Z2) results in a goal (DPOgol).

4. Discussion

The aim of this study was to quantify the type of offensivesequences that result in goals in elite futsal.

The T-patterns that were detected in this study and subse-quently analysed represent 21% of all goals scored. These

Figure 4. Graphical representation of T-pattern 2.

JOURNAL OF SPORTS SCIENCES 625

Dow

nloa

ded

by [

UN

IVE

RSI

TA

T D

E B

AR

CE

LO

NA

] at

11:

25 1

7 M

arch

201

6

included basic movement patterns in the game such as pivotcombinations, parallels and diagonals along with set-pieces.These data demonstrate common offensive sequences andmovement patterns related to goals in elite futsal and thiscould provide important information for the development ofphysical and technical training drills that replicate importantgame situations for all levels and ages.

The first four patterns start in the midfield offensive zone(right and left), through transitions that are developed bypositional attacks. In the first pattern, the offensive sequenceis developed through a pass to the pivot player (in futsal thisfield position is characterised by the most offensive centralzone). Normally, the pivots are characterised by their capacityto receive and control the ball and by their decision-makingcapabilities (Braz, 2006). In this situation, the pivot receivedthe ball in the ZME and performed a pass to the ZMD, whereanother player finished and scored. In the second pattern, thepivot player receives a pass but in this case, through theconduction of the ball, this player moves into a zone nearthe opponent goal where they performed a shot that resulted

in a goal. The third pattern was characterised by an offensivesequence developed through a basic and specific movementof futsal, the parallel. This is characterised by a parallel to thelateral line penetration movement without the ball in thedirection of the opposition goal, by a player from the defen-sive sector of the team (Voser, 2001). They then receive andmove the ball into the ZOCD where a shot occurs that resultsin a goal. Another basic movement pattern within futsal is thediagonal. In this case, a player from the defensive sector of theteam, moves into the offensive half of the pitch through adiagonal run (Voser, 2001) until the ZUOCD is reached and ashot occurs.

The final pattern found displayed very different character-istics to the other patterns since all seven goals were scoredfrom a free kick 10 m away from the opposition goal. This typeof set piece usually occurs due to an accumulation of fouls bythe opposition team. For each foul committed by the oppo-nent, a 10 m free kick is awarded to any outfield player(Lozano, 1995). Thus this should be considered a fundamentalstrategic situation that is only found in futsal and not in other

Figure 5. Graphical representation of T-pattern 3.

626 H. SARMENTO ET AL.

Dow

nloa

ded

by [

UN

IVE

RSI

TA

T D

E B

AR

CE

LO

NA

] at

11:

25 1

7 M

arch

201

6

sports. Professional coaches advocate the use of systematic setpieces training as part of the modern game and our datatherefore support that assertion (Lozano, 1995; Sarmentoet al., 2014; Voser, 2001).

Additionally to the T-pattern analysis, the frequency ana-lysis of the data also suggest that the teams frequently startoffensive sequences that result in a goal, through the transi-tional play and set plays. Therefore, the set pieces, posi-tional attack and the capacity to maintain ball possessionare strengths that characterise the style of play of this team.The central offensive zonal areas were the most common forshots on goal compared to other areas, with 64% of allgoals scored from these ZUOCE, ZUOCD, ZOCE and ZOCDzones of the pitch compared to defensive and wide zones.This finding is similar to the previous work by Álvarez,Manero, Manonelles, and Puente (2004), which concludesthat 90% of the 1771 analysed goals in the Spanish FutsalLeague were scored from a similar area. Similarly, Martín

(2009), Alves (2010) and Lapresa, Álvarez, et al. (2013) con-firmed this zonal trend in futsal, while Yiannakos andArmatas (2006) found a similar tendency in football. Thesedata, in addition to the detected regular structures of beha-viour (T-patterns), are even more important given that theanalysed team can be characterised as a successful team asit has qualified to the championship playoffs, thus thesetendencies of the game can help the coach and teamachieve success.

The present findings concerning the area of the body usedto score goals are in line with previous research (Álvarez et al.,2004). Players typically scored more goals with the inside edgeof the foot (54%), followed by the instep (27%). However,Lapresa, Álvarez, et al. (2013) demonstrated that the SpanishNational team in the 2010 UEFA Futsal Championship scoredmore goals with the instep (55%), and the inside edge of thefoot (22%). These results are consistent with the current litera-ture (Castelo, 2009) that state that using the inside edge of the

Figure 6. Graphical representation of T-pattern 4.

JOURNAL OF SPORTS SCIENCES 627

Dow

nloa

ded

by [

UN

IVE

RSI

TA

T D

E B

AR

CE

LO

NA

] at

11:

25 1

7 M

arch

201

6

foot increases ball precision and placement, with the instep ofthe foot allowing players to take shots from any area of the fieldwhile influencing the ball speed (Lapresa, Álvarez, et al., 2013).

Determining which style of play is the most effective has longbeen disputed in football performance (Hughes & Franks, 2005;Tenga, Holme, Ronglan, & Bahr, 2010a, 2010b). Although thepresent study suggests that positional attack was the mostefficient. We emphasise that this situation is the result of thecomplex interaction of different factors such as the philosophy ofthe playing style, the tradition, identity and history of the club,the quality of the players, as well as the specific environment thatcharacterises the game (e.g. the quality of opposition, matchstatus) (Sarmento, Barbosa, Campaniço, Anguera, & Leitão, 2011).

5. Conclusions

Futsal tactical analysis should not only include the action ofthe players but also the sequences of game play resultingfrom these actions, particular the patterns that result ingoals. Despite the importance attributed to tactical–technicalfactors to the success of futsal teams, this element of thegame is not well understood. The complexity of the T-patterns

observed in this study highlight the intricate nature of futsalperformance and a greater understanding of this area isneeded to optimise the performance in the game. Our datademonstrate that the greatest number of goals were scored inattacking positions and from set pieces and are scored fromcentral offensive zonal areas by using the inside edge andinstep of the foot. T-pattern analysis of offensive sequencesrevealed regular patterns of play that are common in goalscoring opportunities in futsal and are typical movement pat-terns in the sport. This information could help develop physi-cal and technical training drills that replicate important gamesituations.

Disclosure statement

No potential conflict of interest was reported by the authors.

References

Álvarez, J., Manero, J., Manonelles, P., & Puente, J. (2004). Analysis of theoffensive actions resulting in goal of professional league of Spanishfutsal. Revista de entrenamiento deportivo, 18(4), 27–32.

Figure 7. Graphical representation of T-pattern 5.

628 H. SARMENTO ET AL.

Dow

nloa

ded

by [

UN

IVE

RSI

TA

T D

E B

AR

CE

LO

NA

] at

11:

25 1

7 M

arch

201

6

Alves, L. (2010). Descriptive study of the level of technical-tactical goal-keeper futsal world cup in 2008 (Master’s thesis). Federal University ofMinas Gerais: School of Physical Education, Physiotherapy andOccupational Therapy, Brazil.

Anguera, M. T. (2003). Observational methods (General). In R. Fernández-Ballesteros (Ed.), Encyclopedia of psychological assessment (Vol. 2, pp.632–637). London: Sage.

Anguera, M. T. (2004). Hacia la búsqueda de estructuras regulares en laobsservación del fútbol: Detección de patrones temporales. Cultura,Ciencia y Deporte, 1(1), 15–20.

Anguera, M. T. (2005). Microanalysis of T-patterns. Analysis of symmetry/asymmetry in social interaction. In L. Anolli, S. Duncan, M. Magnusson,& G. Riva (Eds.), The hidden structure of social interaction. From genomicsto culture patterns (pp. 51–70). Amsterdam: IOS Press.

Anguera, M. T., Magnusson, M. S., & Jonsson, G. K. (2007). Instrumentos noestándar. Avances en Medición, 5(1), 63–82.

Barbero-Alvarez, J., Soto, V., Barbero-Alvarez, V., & Granda-Vera, J. (2008).Match analysis and heart rate of futsal players during competition.Journal of Sport Sciences, 26, 63–73.

Borrie, A., Jonsson, G., & Magnusson, M. (2002). Temporal pattern analysisand its applicability in sport: An explanation and exemplar data. Journalof Sports Sciences, 20(10), 845–852.

Braz, J. (2006). Organização do jogo e do treino em Futsal: Estudo com-parativo acerca das concepções de treinadores de equipas de rendi-mento superior de Portugal, Espanha e Brasil (Master´s thesis).Faculdade de Desporto. Universidade do Porto, Porto.

Camerino, O., Chaverri, J., Anguera, M. T., & Jonsson, G. (2012). Dynamicsof the game in soccer: Detection of T-patterns. European Journal ofSport Science, 12(3), 216–224.

Casarrubea, M., Jonsson, G., Faulisi, F., Di Giovanni, G., Benigno, A.,Crescimano, G., & Magnusson, M. (2014). T-pattern analysis for thestudy of temporal structure of animal and human behavior: A compre-hensive review. Journal of Neuroscience Methods, 239, 34–46.

Castagna, C., D’Ottavio, S., Granda Vera, J., & Barbero Alvarez, J. C. (2009).Match demands of professional futsal: A case study. Journal of Scienceand Medicine in Sport, 12(4), 490–494.

Castelo, J. (2009). Futebol – Organização dinâmica do jogo. Lisboa: Centrode Estudos de Futebol da Universidade Lusófona de Humanidades eTecnologia.

Cohen, J. (1960). A coefficient of agreement for nominal scales.Educational and Psychological Measurement, 20, 37–46.

Dittrich, N., Silva, J., Castagna, C., Lucas, R., & Guglielmo, L. (2011). Validityof Carminattiʼs test to determine physiological indices of aerobic powerand capacity in soccer and futsal players. Journal of Strength andConditioning Research, 25, 3099–3106.

Dogramaci, S. N., Watsford, M. L., & Murphy, A. J. (2011). Time-motionanalysis of international and national level futsal. Journal of Strengthand Conditioning Research, 25(3), 646–651.

Duarte, R., Batalha, N., Folgado, H., & Sampaio, J. (2009). Effects of exerciseduration and number of players in heart rate responses and technicalskills during futsal small-sided games. The Open Sports Sciences Journal,2, 37–41.

Feenders, G., & Bateson, M. (2012). The development of stereotypic beha-vior in caged europeanstarlings, sturnus vulgaris. DevelopmentalPsychobiology, 54, 773–784.

Hughes, M., & Franks, I. (2005). Analysis of passing sequences, shots andgoals in soccer. Journal of Sports Sciences, 23(5), 509–514.

James, N. (2012). Predicting performance over time using a case study inreal tennis. Journal of Human Sport and Exercise, 7(2), 421–433.

Jovanovic, M., Sporis, G., & Milanovic, Z. (2011). Differences in situationaland morphological parameters between male soccer and futsal – Acomparative study. International Journal of Performance Analysis inSport, 11(2), 227–238.

Lapresa, D., Álvarez, L., Arana, J., Garzón, B., & Caballero, V. (2013).Observational analysis of the offensive sequences that ended in ashot by the winning team of the 2010 UEFA Futsal Championship.Journal of Sports Sciences, 31(15), 1731–1739.

Lapresa, D., Anguera, M. T., Alsasua, R., Arana, J., & Garzón, B. (2013).Comparative analysis of T-patterns using real time data and simulateddata by assignment of conventional durations: The construction of

efficacy in children’s basketball. International Journal of PerformanceAnalysis in Sport, 13(2), 321–339.

Lozano, J. (1995). Futbol sala, experiencias tácticas. Madrid: Real FederaciónEspañola de Futbol.

Magnusson, M. (1978). The human ethological probabilistic structuralmultivariateapproach to the study of children’s nonverbal communica-tion (Thesis). University of Copenhagen’s Silver Medal, Copenhagen.

Magnusson, M. (1996). Hidden real-time patterns in intra- and inter-indi-vidual behavior: Description and detection. European Journal ofPsychological Assessment, 12, 112–123.

Magnusson, M. (2000). Discovering hidden time patterns in behavior:T-patterns and their detection. Behavior Research Methods,Instruments, & Computers, 32, 93–110.

Magnusson, M. S. (2005). Understanding social interaction: Discoveringhidden structure with model and algorithms. In L. Anolli, S. Duncan,M. Magnusson, & G. Riva (Eds.), The hidden structure of social interac-tion. From genomics to culture patterns (pp. 4–24). Amsterdam: IOSPress.

Magnusson, M. S. (2006). Structure and communication in interaction. In G.Riva, M. T. Anguera, B. K. Wiederhold, & F. Mantovani (Eds.), From com-munication to presence: Cognition, emotions and culture towards the ulti-mate communicative experience (pp. 127–146). Amsterdam: IOS Press.

Martín, J. (2009). Analysis of goalkeeper fails in indoor football. RevistaInternacional de deportes colectivos, 2, 36–57.

Nicol, A., Segonds-Pichon, A., & Magnusson, M. (2015). Complex spikepatterns in olfactorybulb neuronal networks. Journal of NeuroscienceMethods, 239, 11–17.

Oliveira, R. S., Leicht, A. S., Bishop, D., Barbero-Alvarez, J. C., & Nakamura, F.Y. (2013). Seasonal changes in physical performance and heart ratevariability in high level futsal players. International Journal of SportsMedicine, 34(5), 424–430.

Sandman, C., Kemp, A., Mabini, C., Pincus, D., & Magnusson, M. (2012). Therole ofself-injury in the organisation of behaviour. Journal of IntellectualDisability Research, 56, 516–526.

Sarmento, H., Anguera, M. T., Pereira, A., Marques, A., Campaniço, J., &Leitão, J. (2014). Patterns of play in the counterattack of elite footballteams – A mixed method approach. International Journal ofPerformance Analysis in Sport, 14(2), 411–427.

Sarmento, H., Barbosa, A., Campaniço, J., Anguera, M. T., & Leitão, J.(2011). T-patterns detection in the counter-attack of the F. C.Barcelona. Scientific Report Series Physical Education and Sport, 15(1),12–16.

Sarmento, H., Marcelino, R., Anguera, M. T., Campaniço, J., Matos, N., &Leitão, J. (2014). Match analysis in football: A systematic review. Journalof Sports Sciences, 32(20), 1831–1843.

Schultz de Arruda, A., De Freitas, C., De Moura, N., Aoki, M., & Moreira, A.(2013). Immune-endocrine responses to a futsal match. Motriz-RevistaDe Educação Física, 19(2), 460–466.

Tenga, A., Holme, I., Ronglan, L., & Bahr, R. (2010a). Effect of playing tacticson achieving score-box possessions in a random series of team posses-sions from Norwegian professional soccer matches. Journal of SportsSciences, 28(3), 245–255.

Tenga, A., Holme, I., Ronglan, L., & Bahr, R. (2010b). Effect of playing tacticson goal scoring in Norwegian professional soccer. Journal of SportsSciences, 28(3), 237–244.

Travassos, B., Araújo, D., Vilar, L., & McGarry, T. (2011). Interpersonalcoordination and ball dynamics in futsal (indoor football). HumanMovement Science, 30(6), 1245–1259.

Van Winckel, J., Tenney, D., Helsen, W., McMillan, K., Meert, J., & Bradley, P.(2014). Fitness in soccer – The science and practical aplicattion. Levene:Moveo Ergo Sum.

Voser, R. (2001). Futsal: Princípios técnicos e tácticos. Rio de Janeiro:Editorial Sprint.

Woods, L., Yefimova, M., & Brecht, M. (2014). A method for measuringperson-centeredinterventions: Detecting and characterizing complexbehavioral symptoms ofpersons with dementia. Clinical Gerontologist,37, 139–150.

Yiannakos, A., & Armatas, V. (2006). Evaluation of the goal scoring patternsin European championship in Portugal 2004. International Journal ofPerformance Analysis in Sport, 6(1), 178–188.

JOURNAL OF SPORTS SCIENCES 629

Dow

nloa

ded

by [

UN

IVE

RSI

TA

T D

E B

AR

CE

LO

NA

] at

11:

25 1

7 M

arch

201

6