Predictive factors of the attack efficacy: the case of the Brazilian women’s Volleyball League champion team

– The aim was to identify the predictive factors of the attack efficacy of the Brazilian women’s League Champion team according to the level of performance of the opposing team. The sample was composed of the observation of 1137 attacks in 21 games of the 2015-2016 Brazilian women’s Volleyball League champion team. The results were significant to identify the predictive factors of the attack efficacy against high ( x 2 =110.562; p <0.00001), intermediate ( x 2 =64.134; p <0.00001) and low ( x 2 =62.137; p <0.00001) performance opponents. Thus, it is concluded that when playing against high-performance teams, powerful attack for position 6 and second attack tempo increase the chances of attack point; powerful attacks for position 1 and 5 reduce the chances of game continuation; and powerful attack for position 6 increases the chances of blocking. In addition, in games against intermediate-performance teams, re-ception that allows organized attack without all attack options reduces the chances of attack point and game continuation, while the reception effect did not seem to be a predictive factor of the attack efficacy in games against high-and low-performance opponents and, finally, in games against low-performance teams, powerful attack for position 1 reduces the chances of game continuation and attack blocking.


INTRODUCTION
Notational analysis in sports is not new 1,2 and allows understanding which aspects of the game influence sports performance 3,4 .In the case of volleyball, it is observed that, in males, attack efficacy is predicted by excellent receptions that allow organized attack with all attack options 5 .In addition, 1 st and 2 nd attack tempo 6,7 and powerful parallel and diagonal attacks 8,9 predict attack efficacy in both sexes.However, analysis regarding sex shows that there are differences in the type of game played, derived from the power used in attacks, as well as the speed of the female game 10,11 , that is, in women's volleyball, attack has lower power and the game is slower compared to men's volleyball.
In analyzing the attack efficacy in complex 1 [12][13][14] , that is, in the game procedures regarding reception, setting and attack, it is verified that the result of the reception action in men's volleyball 6,13 , as well as the attack tempo 13,15 and attack type 10,13,16 in both sexes, influence the attack efficacy and that the type of game practiced changes according to the level of performance of the opposing team 12 .
Thus, the quality of the opposing team enables managing risks in decision-making 17 , and the quality of the game results in changes in the type of game practiced 12 , suggesting that situational constraints must be taken into account when analyzing volleyball game 18,19 .In this context, there is a gap in literature, since most studies had been devoted to the comprehension of the game in a generalized manner 12,13,20 , but the present research aimed to identify the predictive factors of the attack efficacy of the Brazilian women's League champion team according to the level of performance of the opposing team, indicated by the final classification of teams.

Sample
The convenience sample consisted of the observation of 1137 attacks in 21 games of the 2015-2016 Brazilian women's Volleyball League champion team, of the total of 27 games played in that season.This sample is relevant since the Brazilian women's volleyball is internationally ranked among the top four teams and most of its players participate in the Brazilian Volleyball League.

Variables
Level of performance: for the analysis of this variable, teams were divided according to classification at the end of the championship.Thus, teams that finished competition between the 1 st and 4 th positions were considered high performance, teams that finished competition between the 5 th and 8 th positions were considered intermediate performance and those that finished competition between the 9 th and the 12 th positions were considered low performance.
Reception Efficacy: reception efficacy was analyzed using the instrument proposed by Maia and Mesquita 20 : • Poor reception (C): reception that did not allow organized attack.
• Moderate Reception (B): Reception that allowed organized attack, although not all attackers were available for the attack.• Excellent reception (A): reception that allowed organized attack with all attackers available for the attack.
Attack tempo: for this analysis, the instrument proposed by Afonso et al. 15 was used: • 1 st attack tempo: the attacker jumped during or immediately after setting, in which one step can occur after setting; • 2 nd attack tempo: the attacker made two or three steps after setting; • 3 rd attack tempo: the attacker waited for the ball to reach the peak of the upward trajectory and only then began the attack step.
Attack type: attack type was analyzed according to instrument proposed by Costa et al. 21.Attacks exploiting block, when carried out with power, were grouped into powerful attacks on the parallel or diagonal, while Off-speed Attack that exploited block were grouped into the Offspeed Attack category: Number of blockers: Corresponds to the number of blockers that opposes the opposing attack.Thus, the following categories were obtained: • No block: corresponds to the absence of blockers due to the excellence of setting.

Data collection procedure
All games were recorded from the top perspective, i.e., about 7-9 meters behind the bottom line of the court and the camera was positioned approximately three meters above ground level for better viewing of video scenes.A Sony camera with 1080p HD resolution and 60Hz frequency was used for data collection.Observers were physical education professionals with at least 5-year experience in the function of observational analyzers.For the reliability calculation, 20% of actions were re-analyzed, exceeding the reference value of 10% 23 .Cohen's Kappa values for inter-and intra-observer reliability, respectively, were: reception efficacy = 0.98 and 0.98; attack tempo = 0.96 and 0.94; attack type = 1.00 and 1.00; attack efficacy = 1.00 and 1.00.In this sense, the reliability values are above the reference value of 0.75 24 .

Statistical Procedures
Descriptive analysis and multinominal logistic analysis were performed, observing the relationship of independent variables with the dependent variable one by one.In this context, attack efficacy was considered as dependent variable and attack efficacy, attack tempo and attack type were independent variables, since the number of blockers was not associated with attack efficacy.In order to avoid inconsistencies in the prediction model due to the low occurrence, APP2 and APP4 attacks were excluded.In addition, the analysis of the predictive attack factors was stratified according to the level of performance of opponents, obtaining an analysis for each of them.In data treatment, 5% significance level (p≤0.05) was adopted and SPSS software version 20.0 was used.

RESULTS
The descriptive analysis is presented in table 1 and showed that the occurrences of game procedures varied, although the distribution tendency modified little according to the level of performance of the opposing team.The analysis of predictive factors of the attack efficacy in confrontations against high-performance opponents was statistically significant (x 2 = 110.562;p <0.00001), as shown in table 2.
The analysis of predictive factors of the attack efficacy in confrontations against intermediate-performance opponents was statistically significant (x 2 = 64,134; p <0.00001), as shown in table 3.
The analysis of predictive factors of the attack efficacy in confrontations against low-performance opponents was statistically significant (x 2 = 62.137; p <0.00001), as shown in table 4.

DISCUSSION
The aim of the present study was to identify the predictive factors of the attack efficacy of the Brazilian women's League champion team according to the level of performance of the opposing team.Descriptive analysis showed that, regardless of level performance of the opposing team, there was higher occurrence of A reception, second attack tempo, powerful attack for position 1 and double compact block.These results corroborate literature, which shows greater occurrence of receptions that allow organized attack 8,12,21 , second attack tempo 8,9,13,25 , powerful attack 13 and double block 26 .In this context, it was observed that the offensive game pattern of the Brazilian League champion team did not change according to the level of performance of the opponent and therefore, to a certain extent, do not corroborate results that indicate that the quality of the opposing team and the quality of confrontation enable managing risks in decision-making 17 , as well as changes in the type of game practiced 12 .Thus, it is suggested that the complex I, composed of reception, setting and attack, suffered little influence from the opposing team when considering the game procedures analyzed.However, these results do not allow inferring about the offensive strategy used, a factor that can contribute to similarities observed in the descriptive analysis and that, according to Ramos et al. 19 , changes according to the opponent and moment of the game.Analysis of predictive factors of the attack efficacy on games against high-performance teams showed that powerful attack for position 6 and second attack tempo increased the chances of attack point; powerful attacks for positions 1 and 5 reduced the chances of game continuation; and powerful attack for position 6 increased the chances of blocking.These results partially corroborate literature, since 1 st and 2 nd attacks tempo are predictive factors of the attack point, as well as the powerful attacks 8,11,27 .Regarding block, there is partial agreement with the study by Conti et al. 9 , who demonstrated increased chances of blocking after powerful attacks … continue performed with parallel and diagonal.In this way, it was observed that to set faster, specifically in second attack tempo and powerful attack are essential in high-performance women's volleyball.However, the fact that the powerful attack for position 6 has increased the chances of block to occur suggests greater concern with tactical procedures to protect this type of attack by the attacking team.Analysis of predictive factors of the attack efficacy on games against intermediate-performance teams showed that reception B reduced the chances of attack point and game continuation.The results corroborate the study by Conti et al. 9 , who demonstrated reduction of the chances of point after receptions that do not allow attack with all attack options.In addition, the study by Costa et al. 10 demonstrated that defeat on the set is predicted by non-scoring attacks and poor quality receptions.Thus, it is possible to suggest that receptions that do not allow attack organization with all attack options limit distribution and, consequently, restrict the action of the attacker, making difficult obtaining the point.
Analysis of predictive factors of the attack efficacy on games against low-performance teams showed that powerful attack for position 1 reduced the chances of game continuation and attack block.These results partially corroborate the study by Costa et al. 8 , who demonstrated reduction of game continuation after powerful attacks performed for position 1.On the other hand, the results disagree with those obtained by Conti et al. 9 , who observed an increase in the chances of game continuation occurring after powerful attack for position 1.Although there are divergences in the predictive factors pointed out in the present research, it was observed that this fact can be related to the specific analysis of the aforementioned studies, since they analyzed attacks made from positions 4, 2 and 1 of the men's national league.However, since literature points out that the game played in women's volleyball is supportive, that is, it consists of a greater amount of defense than the game played in men's volleyball 6,12,28 , it can be suggested that attacks made from position 1 aim to limit the offensive construction, considering that the setter occupies such a position in the defensive system when in the defense zone.
The present study did not show reception efficacy as a predictive factor of the attack efficacy in games against high-and low-performance opponents, and did not show the 1 st attack tempo as a predictive factor of the attack efficacy, regardless of opponent's level of performance.These results contradict the available literature in both sexes 9,14,26,29 , which demonstrated the influence of reception efficacy on the attack efficacy, as well as the 1 st attack tempo on point achievement.According to the Brazilian Volleyball Confederation (CBV) 30 , the efficacy of attackers of the champion team was higher than the average of the competition and 3 attackers of the champion team were ranked among the 10 most effective attackers of the competition, which suggests that when playing against high-performance opponents, high-quality receptions are a prerequisite and teams should have specific offensive tactical strategies based on situational constraints.On the other hand, in confrontations with low-performance teams, reception efficacy may not be predictive due to discrepancies in the tactical-technical performance, especially in attack, a game procedure that is shown to be a determining factor for the point achievement and winning the set 5,14 .With regard to attack tempo, it is possible to infer that playing in 1 st attack tempo is an essential factor and that the level of performance does not change this game procedure.Thus, it was observed that high-quality receptions and fast game are fundamental for point achievement and are not differentiating factors of sport performance.
Finally, it is hoped that the information contained in the results obtained allows adaptations of the tactical concepts used by coaches in structuring specific training to improve both attack performance through training of technical and tactical actions that potentiate the attack efficacy and in a defense situation, consolidating technical-tactical actions capable of neutralizing the opponent's actions.However, the study presents as limitations the analysis of the attack efficacy as a whole, without distinguishing distinct periods of the game (e.g., differences between sets or punctuation intervals within sets), and did not consider counterattacks.

CONCLUSION
Based on the aims, applied methods and results found, the present study showed how the performance of the opposing team influences the offensive construction of the attack, that is, in complex 1.In this way, confrontations with low-and intermediate-performance teams were influenced by the reception efficacy and attack type, respectively.On the other hand, confrontations with high-performance teams were influenced by powerful attacks for position 1, suggesting the need to limit the opponent's offensive construction.

Table 1 .
Descriptive analysis of game procedures in relation to the performance of the opposing team

Table 2 .
Predictive factors of the attack efficacy against high-performance opponents aThe reference category for the dependent variable is the attack error; b The reference category for the independent variable; * difference for p <0.05

Table 3 .
Predictive factors of the attack efficacy against intermediate-performance opponents Note. a The reference category for the dependent variable is the attack error; b The reference category for the independent variable; * difference for p <0.05

Table 4 .
Predictive factors of attack efficacy against low-performance opponents The reference category for the dependent variable is the attack error; b The reference category for the independent variable; * difference for p <0.05