Reliability of physical, physiological and tactical measures in small-sided soccer Games with numerical equality and numerical superiority

Confiabilidade das medidas de demanda física, fisiológica e tática em pequenos jogos com superioridade e igualdade numérica no futebol

Abstract

The aim of this study was to investigate the physical, physiological (HR) and tactical demands of small-sided soccer games (SSG) with numerical equality (3x3) and numerical superiority (4x3). Eighteen male players of a soccer team played against each other both types of SSG, both played as two 4-minute bouts interspersed with 4-minute rest. Physical (distances, velocities and accelerations) and HR demands were recorded by a GPS system (15hz) and a heart rate monitor during SSG. The individual tactical behaviors proposed by the System of Tactical Assessment in Soccer were registered by an observer who analyzed the SSG videos. Reliability was verified calculating the intraclass correlation coefficient (ICC) and the absolute and percentage standard error of measurement (SEM and %SEM, respectively), both considering athletes’ individual scores in each SSG bout. As a result, higher number of variables related to physical and physiological demands presented significant ICC values, being classified as moderate to excellent in relation to variables related to tactical demands. Considering these variables that presented significant ICC values in both SSG, the SEM% values found for physical and physiological variables ranged from 3.0 to 42.0%, while those found for the tactical variables ranged from 31.6 to 75.0%. It could be concluded that several variables related to physical and physiological demands present moderate to excellent reliability. Differently, the majority of variables related to tactical demands were not reliable in both types of SSG.

Key words
Reliability; Soccer; Task performance analysis

Resumo

O objetivo do presente estudo foi investigar a confiabilidade das demandas física, fisiológica (FC) e tática em configurações de pequenos jogos (PJ) em igualdade (3x3) e superioridade numérica (4x3) no futebol. Dezoito jogadores do sexo masculino de uma equipe de futebol jogaram entre si os dois tipos de PJ, com duas séries de 4 minutos e 4 minutos de pausa entre elas. Durante os jogos, as variáveis relacionadas à demanda física (distâncias, velocidades, acelerações) e fisiológica foram registradas utilizando-se um equipamento de GPS (15hz) e um cardiofrequencímetro. Os comportamentos táticos individuais propostos pelo Sistema de Avaliação Tática no Futebol foram registrados por um avaliador partir da análise da filmagem dos jogos. A confiabilidade das medidas foi verificada considerando os valores individuais de cada atleta em cada série de PJ, por meio do coeficiente de correlação intraclasse (CCI) e do erro padrão de medida absoluto e percentual (EPM e EPM%, respectivamente). Como resultado, um maior número de variáveis relacionadas às demandas física e fisiológica apresentaram valores de CCI significativos com classificação moderada a excelente em relação às variáveis relacionadas à demanda tática. Considerando as variáveis com valores de CCI significativos em ambas as configurações de jogo, o EPM% variou entre 3,0 e 42,0% para as variáveis físicas e para a FC, e de 31,6 a 75,0% para as variáveis relativas à demanda tática. Conclui-se que diversas variáveis relacionadas às demandas física e fisiológica apresentaram confiabilidade moderada à excelente. Diferentemente, a maioria das medidas relacionadas à demanda tática não foi confiável em ambos PJ.

Palavras-chave
Análise de desempenho e tarefas; Confiabilidade; Futebol

INTRODUCTION

Small-sided games (SSG), as a means of training in soccer, enables technical, physical and physiological demands to occur associated to tactical demands similar to those found in the formal game¹. Researching the reliability of the different variables that represent these demands provides data for understanding the variability of responses when athletes are exposed to the same training condition. Reliability refers to the consistency or reproducibility of a measure when the individual performs a repeated test condition². Thus, knowledge on the variability of responses in SSG supports the prescription of this type of training.

Previous studies have investigated the reliability of variables related to physiological and physical demands of soccer SSG^3-5. In these, heart rate (HR) showed high reliability, while blood lactate concentration and rate of perceived exertion showed low to moderate reliability^3-6. In terms of physical demand, the distance traveled at low speeds shows high reliability^5,6. However, actions of acceleration and displacements at high speeds have low reliability in soccer SSG^4,5, possibly due to the low sampling frequency of GPS receivers^7,8, the natural variation of these responses between exercise bouts and training sessions⁹ and differences among SSG configurations¹⁰.

The reliability of physical and physiological demands has been observed only in configurations with numerical equality (1x1 to 6x6)^3,4,6,10. However, considering that the formal game presents situations with numerical unbalance among players¹¹ (e.g. offensive numerical superiority) and that such situations may have different demands from those with numerical equality^12,13, researching SSG reliability with numerical superiority can provide subsides for coaches regarding training prescription. In addition, the reliability of tactical demand in SSG has not been investigated, limiting the understanding of the variability of these responses during SSG.

Therefore, the aim of this study was to investigate the reliability of physical, physiological and tactical demands in soccer SSG with numerical equality (3x3) and numerical superiority (4x3).

METHODOLOGICAL PROCEDURES

Sample

Eighteen U-17 male players were selected from a professional soccer club (age: 16.4 ± 0.4 years, weight: 68.4 ± 8.0 kg). All participated in competitions at national/international level and performed, on average, seven weekly training sessions. Players and their legal guardians signed the informed consent form with information about the research. The study was approved by the Ethics Research Committee of the Federal University of Minas Gerais (CAAE: 29215814.8.0000.5149).

Procedures

Each team was composed of a defender, a midfielder and a forward with similar levels of procedural tactical knowledge, as shown in Figure 1.

Figure 1
Procedures for the composition of teams G1 and G2 = Athletes with higher and lower performance in PTKT, respectively; D = Defender; M = midfielder; F = Forward. Superscript number indicates the classification in the PTKT performance ranking in each positional category from 1st to 6th place.

To assess the procedural tactical knowledge of players in each position, the Procedural Tactical Knowledge Test (PTKT)¹⁴ was applied to each group of six athletes from the same position (e.g., six defenders). In each of these groups, two teams of three athletes (randomly allocated) played a game that was filmed for later analysis of the procedural tactical knowledge. Based on the PTKT results, a performance ranking was established within each positional status (1^st to 6^th place). Then, the sample was divided into Group 1 (G1) and Group 2 (G2) composed of three players each positional status (total of 9 athletes each) with greater and lesser performance on the PTKT, respectively. Each group was divided into three teams (A, B and C and D, E and F) of three athletes, one defender, one midfielder and one forward. In order to make balanced teams, they could not have two athletes with the same classification in the PTKT performance ranking.

After the composition of teams, players performed 3x3 and 4x3 SSG in two familiarization sessions. The collection occurred every other day, always at the same time and after standardized preparatory activity. Teams from the same group randomly performed each type of SSG in all possible games (i.e. AXB, AXC, BXC, DXE, DXF and EXF). No games were played between G1 and G2 teams. For each game, two SSG (bouts) of each type were performed, totaling 12 games in each group (total of 24 games). Each SSG had total duration of four minutes with a four-minute rest between them¹⁰. To verify reliability, two SSG of the same type were carried out between the same teams (e.g., two 3x3 SSG between teams A and B).

Instruments

• Small-sided games

In both SSG, pitch size was 36x27 meters, with goals of 5x2 meters¹⁵. All the rules of formal soccer game were used, including offside. Auxiliary balls were at hand for a quick ball replacement. Two coaches provided external encouragement³ to teams. In the 4x3 configuration, an additional floater player was included (midfielder of same tactical knowledge level of the team that was not playing - e.g., midfielder of team C in the AXB game in G1). After the loss of the ball by a team, the player was transferred to the other team and conducted the offensive actions for the new team. This player could perform all actions of the offensive process, except free kicks and ball replacement on sides and corners.

• Physical demand

GPS units brand GPSports Systems model SPIProX2 (15Hz) with coupled triaxial accelerometer (100Hz) were attached between athletes’ shoulder blades. The validity and reliability of this equipment have been reported in previous studies⁷. Variables related to physical demand were total distance traveled, percentage of the total distance traveled in speed ranges from 0 to 7.2 km/h (% D0-7.2), 7.3 to 14.3 km/h (%D7.3-14.3), and 14.4 to 21.5 km/h (%D14.4-21.5)¹⁶, peak speed, peak acceleration, number of accelerations greater than 2.0 m/s² and distance traveled in accelerations above 2.0 m/s.

• Physiological demand

HR was recorded using Polar^®, FS1 transmitters (Finland), compatible with the GPS interface. Data relating to rest between bouts were excluded.

Mean heart rate (HR_MEAN) expresses the mean values recorded by the equipment during the four minutes of SSG. Peak heart rate (HR_PEAK) is the highest value recorded in the four minutes.

• Tactical demand

Games were analyzed using the Soccer Analyser® software. Two observers assessed the frequency of occurrence of individual tactical behavior of players using the System of Tactical Assessment in Soccer (FUT-SAT)¹⁵, considering the location of actions on the playing field.

Intra- and inter-observer concordances were considered “perfect”¹⁷, with Kappa coefficient values above 0.9 (standard error <0.006).

Statistical analysis

Reliability was calculated through the intraclass correlation coefficient 3,1 (ICC_3,1) and standard error of measurement (SEM) considering the individual values obtained in each bout of each type of SSG.

ICC_3,1 values⁹ were accompanied by their respective 95% confidence intervals and classified as weak (<0.4), moderate (0.4 to 0.59), good (0.6 to 0.74) and excellent (0.75-1)¹⁸. SEM⁹ was relativized by the average values of variables, resulting in percentage SEM (%SEM)⁹. All analyses were performed using SPSS 20.0 software (Chicago, IL). The statistical significance level was 5%.

RESULTS

ICC, SEM, %SEM values and ICC classifications are shown in Tables 1 and 2. The number of accelerations and the distance traveled in accelerations above 2.0 m/s showed significant reliability values only in 3x3 SSG. For some variables, the absolute ICC values were higher in 4x3 SSG compared to 3x3 SSG.

Table 1
ICC, SEM, %SEM values and ICC classification for variables related to physical and physiological demands in 3x3 and 4x3 SSG.

	Small-sided games
		3x3			4x3
	ICC (95% CI)	ICC classification	SEM (%)	ICC (95% CI)	ICC classification	SEM (%)
HRmean	0.72* (0.46 - 0.86)	Good	8.9 (5.4)	0.87* (0.74 - 0.93)	Excellent	6.7 (4.2)
HRpeak	0.51* (0.23 - 0.71)	Moderate	5.6 (3.0)	0.61* (0.32 - 0.79)	Good	6.0 (3.3)
Total distance	0.68* (0.21 - 0.86)	Good	22.6 (5.2)	0.71* (0.50 - 0.84)	Good	29.5 (7.2)
%D0-7.2	0.38* (0.07 - 0.63)	Weak	11.2 (6.6)	0.42* (0.12 - 0.65)	Moderate	13.2 (7.9)
%D7.3-14.3	0.56* (0.09 - 0.79)	Moderate	22.1 (11.2)	0.74* (0.55 - 0.86)	Good	24.0 (15.1)
%D14.4-21.5	0.54* (0.25 - 0.74)	Moderate	17.3 (30.0)	0.28* (-0.06 - 0.56)	Weak	21.3 (42.0)
Peak speed	0.08 (-0.24 - 0.39)	Fraco	2.3 (10.6)	0.09 (-0.26 - 0.40)	Weak	2.4 (10.8)
N. Accel. > 2.0 m/s2	0.66* (0.42 - 0.81)	Good	1.4 (17.3)	0.24 (-0.09 - 0.53)	Weak	2.1 (30.5)
% Dist. Accel. > 2.0 m/s2	0.51* (0.23 - 0.72)	Moderate	21.3 (27.1)	0.27 (-0.07 - 0.55)	Weak	22.5 (34.4)
Peak acceleration	-0.29 (-0.58 - 0.05)	Weak	0.4 (12.5)	0.24 (-0.09 - 0.53)	Weak	0.4 (12.0)

• ^*
  = statistical significance for the ICC values. HR_MEAN and HR_PEAK = mean heart rate and peak heart rate; %0-7.2, %7.3-14.3, %14.4-21.5 = percentage of the total distance in speed ranges from 0 to 7.2 km/h, 7.3 to 14.3 km/h and 14.4 to 21.5 km/h, respectively; N. Accel.> 2.0 m/s² and % Dist. Accel.> 2.0 m/s² = number of accelerations above 2.0 m/s² and percentage of the total distance traveled in accelerations above 2.0 m/s², respectively.

Table 2
ICC, SEM, %SEM values and ICC classification for variables related to tactical demand in 3x3 and 4x3 SSG.

	Small-sided games
	3x3						4x3
		ICC (95% CI)		ICC classification		SEM (%)		ICC (95% CI)		ICC classification		SEM (%)
Penetration		0.07 (-0.27 - 0.39)		Weak		1.9 (40.6)		0.14 (-0.20 - 0.45)		Weak		1.9 (50.1)
Offensive coverage		0.40* (0.08 -0.64)		Moderate		2.0 (55.1)		0.16 (-0.18 - 0.47)		Weak		2.8 (72.1)
Width and length (without ball)		0.06 (-0.28- 0.39)		Weak		3.3 (45.9)		0.43* (0.14 - 0.66)		Moderate		2.9 (37.6)
Width and length (with ball)		0.17 (-0.17-0.47)		Weak		1.1 (82.0)		0.21 (-0.14 - 0.50)		Weak		1.0 (87.0)
Depth Mobility		0.35* (0.02- 0.61)		Weak		1.4 (71.5)		0.40* (0.08 - 0.65)		Moderate		2.1 (75.0)
Offensive unit		0.40* (0.08 - 0.64)		Moderate		3.5 (38.0)		0.23 (-0.11 - 0.52)		Weak		5.1 (43.7)
Delay		0.01 (-0.33 - 0.33)		Weak		2.1 (41.1)		-0.13 (-0.45 - 0.21)		Weak		2.6 (53.2)
Defensive coverage		0.36* (0.04 - 0.61)		Weak		2.2 (47.8)		0.19 (-0.15 - 0.49)		Weak		3.1 (53.8)
Defensive balance		0.02 (-0.32 - 0.34)		Weak		3.3 (60.4)		0.01 (-0.32 - 0.34)		Weak		3.0 (55.6)
Recovery balance		0.35* (0.02 - 0.61)		Weak		1.4 (70.9)		0.17 (-0.13 - 0.46)		Weak		1.9 (80.7)
Concentration		0.078 (-0.25 - 0.39)		Weak		1.8 (76.6)		0.07 (-0.27 - 0.39)		Weak		2.1 (78.0)
Defensive unit		0.22 (-0.12 - 0.51)		Weak		3.7 (30.9)		0.20 (-0.09 - 0.47)		Weak		4.0 (30.0)
(A) MC. OF.		0.38* (0.05 - 0.63)		Weak		4.4 (41.4)		0.69* (0.47 - 0.83)		Moderate		3.7 (31.6)
(A) MC. DEF.		0.36* (0.04 - 0.61)		Weak		6.4 (31.3)		0.23 (-0.11 - 0.51)		Weak		7.7 (35.5)
(D) MC. OF.		0.16 (-0.18 - 0.46)		Weak		6.5 (49.5)		0.34* (0.01 - 0.60)		Weak		5.2 (36.6)
(D) MC. DEF.		0.07 (-0.27 - 0.39)		Weak		5.7 (31.1)		0.27 (-0.05 - 0.55)		Weak		5.1 (26.5)

• ^*
  = statistical significance for ICC values; (A) MC. OF. = Tactical attack actions carried out in the offensive midfield; (A) MC. DEF. = Tactical attack actions carried out in the defensive midfield; (D) MC. OF. = Tactical defense actions carried out in the offensive midfield; (D) MC. DEF. = Tactical defense actions carried out in the defensive midfield.

Table 2 indicates that 3x3 and 4x3 SSG show two and three variables, respectively, with significant ICC values which were classified as moderate. These variables are different in 3x3 and 4x3 SSG.

It is observed in Tables 1 and 2 that among variables with significant ICC in both SSG, %SEM values of variables related to physical and physiological demands ranged from 3.0 to 42.0% while those for variables related to tactical demand ranged from 31.6 to 75.0%.

DISCUSSION

The ICC value reflects the magnitude of the variability between subjects and the consistency of this variability in various measures of the same test⁹. Low ICC values indicate that the differences between subjects are not maintained between repeated conditions of a test. The SEM value reflects the degree of fluctuation of the scores of an individual in a test/condition, indicating the expected natural variability (random error) for the response of a given variable⁹. Higher SEM values indicate higher ranges of expected scores for a certain variable, which makes it difficult to perceive significant changes in the scores of an individual following a systematic intervention (e.g., training).

Considering both SSG in the present study, several variables related to physical and physiological demands showed moderate or good ICC values and one variable showed excellent ICC, suggesting that individuals who had higher scores in the first bout continued to have higher scores in the second bout. In relation to HR, other studies with SSG corroborate the good consistency of this variable^3,4, and Hill-Haas et al.⁵ showed a percentage typical error from 2 to 4% for HR in 4x4 SSG with athletes aged 17 years. In this study, %SEM (equivalent to percentage typical error) showed similar values from 4.2 to 5.4 and 3.0 to 3.3 for mean HR and peak HR, respectively. In addition, Ngo et al.¹⁰ presented ICC of 0.95 and 0.96 for mean HR in 3x3 SSG with soccer players of the same age.

Physical demands showed moderate to good ICC values. For the total distance traveled, %SEM values between 5 and 7% corroborate the results of Ade et al.⁴, who showed coefficients of variation from 6 to 8% for this variable in 1x1 and 2x2 SSG. For the distance traveled between 14.4 and 21.5 km/h, higher %SEM values were found (30.0 to 42.0%), but also close to those found by Ade et al.⁴ (40-60%) in a similar speed range (14.5-19.8 km/h). In fact, studies have shown a decrease in reliability for higher speeds in SSG using 1Hz, 5Hz, and 10Hz GPS^7,8,19. This may be related to the low sampling frequency of GPS devices, which limits the capture of short movements at high speeds²⁰.

Comparing SSG, only 3x3 configuration showed significant reliability for the number of accelerations and the distance traveled in accelerations above 2.0 m/s². Considering that these variables represent the most intense actions during SSG, greater reproducibility of these actions might be related to the increased need for intense confrontations among players in game with numerical equality. The other variables showed equal or greater reliability in the 4x3 configuration, suggesting that numerical superiority may show greater reproducibility in less intense demands, which are facilitated by the presence of the additional player. However, to prove the existence of statistical differences between absolute ICC values of the different variables in each SSG, an inferential analysis would be necessary and, consequently, a different experimental design.

For tactical variables with significant and moderate ICC values, %SEM valuesranged from 31.6 to 75.0%, considering both SSG. These data indicate a high fluctuation of measures and low consistency of differences between subjects in bouts 1 and 2 of each SSG. These results may be related to inherent characteristics of the game, such as unpredictability and complexity, requiring players to adapt their actions to tactical situational demands²¹.

Moreover, it is possible that the frequency of occurrence of certain tactical behavior does not reflect the actual response of each tactical behavior of U-17 category athletes in the investigated conditions. For example, one of the athletes from team C had 9 penetration actions in the first SSG and only 2 in the second. Another athlete had seven defensive coverage actions in the first SSG and 17 in the second. Such data impact both the consistency of tactical behaviors as well as its variability and may have been influenced by the game duration²². In this study, the duration of 4 minutes was determined considering previous data suggesting that 4 minutes are sufficient for the manifestation of different tactical principles¹⁵. However, if four minutes are sufficient for such events to occur with a frequency capable of representing the expected variability for each principle remains an issue to be investigated.

Another important result indicates that variables with significant ICC in 3x3 SSG were not the same that presented significant ICC in 4x3 SSG. Therefore, the reproducibility of the tactical demand seems to be influenced by SSG configuration. For example, the offensive unit may have had greater reproducibility in 3x3 SSG due to greater participation of players behind the line of the ball in attack due to the difficulty to progress with the ball in game with numerical equality. The presence of the additional player in 4x3 SSG may have allowed the three athletes of the team to be free to move between the ball line and the line of the last defender, providing moderate reliability in width and length (without the ball) principle. In 3x3 SSG, only two athletes would be able to perform this movement. These examples allow us considering that the ICC values of variables related to tactical demands may be associated to the frequency of a certain behavior, allowing it to present a “representative” variability for the investigated conditions.

The number of participants in this study may be a limitation to the generalization of results. In addition, considering that the level of tactical knowledge can be different between athletes of different ages²³, different patterns of tactical behavior in different categories are expected, thus limiting the conclusions of this study to the U-17 category.

CONCLUSION

The results indicate moderate to excellent reliability for several variables related to physical and physiological demands in soccer SSG, while most variables related to tactical demand have low reliability. These results suggest that coaches should be aware that 3x3 and 4x3 SSG as a means of training in soccer generate physical and physiological demands with variability around 3-42% and variability of up to 75% for tactical responses of athletes.

REFERENCES

• ¹
  Davids K, Araújo D, Correia V, Vilar L. How small-sided and conditioned games enhance acquisition of movement and decision-making skills. Exerc Sport Sci Rev 2013; 41(3):154–61.
• ²
  Hopkins WG. Measures of reliability in sports medicine and science. Sport Med 2000; 30(1):1–15.
• ³
  Rampinini E, Impellizzeri FM, Castagna C, Abt G, Chamari K, Sassi A, et al. Factors influencing physiological responses to small-sided soccer games. J Sports Sci 2007; 25(6):659–66.
• ⁴
  Ade JD, Harley JA, Bradley PS. Physiological response, time-motion characteristics, and reproducibility of various speed-endurance drills in elite youth soccer players: Small-sided games versus generic running. Int J Sports Physiol Perform 2014; 9(3):471–9.
• ⁵
  Hill-Haas S, Coutts A, Rowsell G, Dawson B. Variability of acute physiological responses and performance profiles of youth soccer players in small-sided games. J Sci Med Sport 2008; 11(5):487–90.
• ⁶
  Hill-Haas S, Rowsell G, Coutts A, Dawson B. The Reproducibility of Physiological Responses and Performance Profiles of Youth Soccer Players in Small-Sided Games. Int J Sports Physiol Perform 2008; 3(3):393–6.
• ⁷
  Köklü Y, Arslan Y, Duffield R. Accuracy and reliability of SPI ProX global positioning system devices for measuring movement demands of team sports. J Sports Med Phys Fitness 2015; 55(2):1-2.
• ⁸
  Varley MC, Fairweather IH, Aughey RJ. Validity and reliability of GPS for measuring instantaneous velocity during acceleration, deceleration, and constant motion. J Sports Sci 2012; 30(2):121–7.
• ⁹
  Weir JP. Quantifying test-retest reliability using the Intraclass Correlation Coefficient and the SEM. J Strength Cond Res 2005; 19(2):231–40.
• ¹⁰
  Ngo JK, Tsui MC, Smith AW, Carling C, Chan GS, Wong DP. The effects of man-marking on work intensity in small-sided soccer games. J Sport Sci Med 2012; 11(1):109–14.
• ¹¹
  Travassos B, Vilar L, Araújo D, Mcgarry T. Tactical performance changes with equal vs unequal numbers of players in small-sided football games. Int J Perform Anal Sport 2014; 14(2):594–605.
• ¹²
  Sampaio JE, Lago C, Gonçalves B, Maçãs VM, Leite N. Effects of pacing, status and unbalance in time motion variables, heart rate and tactical behaviour when playing 5-a-side football small-sided games. J Sci Med Sport 2014; 17(2):229–33.
• ¹³
  Bekris E, Mylonis E, Sarakinos A, Gissis I, Anagnostakos A, Kombodieta N. Supernumerary in small sided games 3Vs3 & 4Vs4. J Phys Educ Sport 2012; 12(3):398–406.
• ¹⁴
  Greco PJ, Aburachid LMC, Silva SR, Morales JCP. Content validation of tactical-technical actions of the Tactical Procedure Knowledge Test - Sporting Orientation. Motrici 2014; 10(1):38–48.
• ¹⁵
  Teoldo I, Garganta J, Greco PJ, Mesquita I, Maia J. System of tactical assessment in Soccer (FUT-SAT): Development and preliminary validation. Motri 2011; 7(1):69–83.
• ¹⁶
  Owen AL, Wong DP, Paul D, Dellal A. Physical and technical comparisons between various-sided games within professional soccer. Int J Sports Med 2014; 35(4):286–92.
• ¹⁷
  Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977; 33(1):159–174.
• ¹⁸
  Cicchetti DV. Guidelines, criteria, and rules of thumb for evaluating normed and standardized assessment instruments in psychology. Psychol Assess 1994; 6(4):284–90.
• ¹⁹
  Gray AJ, Jenkins D, Andrews MH, Taaffe DR, Glover ML. Validity and reliability of GPS for measuring distance travelled in field-based team sports. J Sports Sci 2010; 28(12):1319–25.
• ²⁰
  Jennings D, Cormack S, Coutts A, Boyd L, Aughey R. The Validity and Reliability of GPS Units for Measuring Distance in Team Sport Specific Running Patterns. Int J Sports Physiol Perform 2010; 5(10):328–41.
• ²¹
  Garganta JM. Trends of tactical performance analysis in team sports: bridging the gap between research, training and competition. Rev Port Ciênc Desporto 2009; 9(1):81–9.
• ²²
  Mateus N, Gonçalves G, Abade E, Liu H, Torres-Ronda L, Leite N, et al. Game-to-game variability of technical and physical performance in NBA players. Int J Perform Anal Sport 2015; 15(3):764–76.
• ²³
  Giacomini DS, Silva EG, Greco PJ. Comparação do conhecimento tático declarativo de jogadores de futebol de diferentes categorias e posições. Rev Bras Ciênc Esporte 2011; 33(2):445–63.

Publication Dates

• Publication in this collection
  Sep-Oct 2016

History

• Received
  05 July 2016
• Accepted
  15 Sept 2016