TACTICS OF ATTACK OF FOOTBALL TEAMS IN THE CHAMPIONS LEAGUE KNOCKOUT PHASE IN SEASONS OF 2015/2016 AND 2016/2017

The aim of this study is to find out characteristics of tactics of attack of teams which play in the finals of the Champions Leagule by analyzing competition activities and also to use comparative analysis for the same competition in two competition seasons and point to the potential tendencies in tactical preparation of teams for the very end of the competition. The analysis involved quarter-finals, semi-finals and finals in seasons 2015/2016 and 2016/2017, i.e. 13 matches in each season. The tactical action was analyzed by monitoring variables that describe the ways of beginning, realization and completion of successful attacks, but also the overall characteristics of play when a team was in ball possession (scope, success and speed of passing the ball). Results of this study showed that that there are no significant differences in tactical demonstration of teams that play the final round of the competition in two competition seasons. The application of continuous attacks is dominant, beginning by cutting a pass and winning “the second ball” on the opponent’s half, while the final pass is in most cases the centre and back lateral passing. However, significand difference was discovered only in efficiency during the passing game, which was significantly greater in the 2016/2017 season (p=0.013) and partially in the longer ball possession also in the 2016/2017 season. The kind of analysis in the field of foodball tactics is valuable because it provides a possibility to define the criteria for achieving results on the highest level, but to predict potential tendencies in developing offensive foodball tactics.


INTRODUCTION
Moments of the game that more or less influence the final result can be identified by analyzing competition activities in football on the individual or the team level (Castelano et al., 2012; Liu & Gomez., 2014; . Studies that deal with this kind of analyses in a longer period of time aim at discovering these technical and tactical parameters of the game correlated with success, and, on the other hand, serve to envisage new trends, and above all, football game tactics . Results of these studies and their explanation function to define theoretical procedures in evolutive development of football, but they also have a significant role in technological process and preparation for certain football competitions.
A high level of technical, tactical, physical and psychological preparation is a necessary prerequi-site for competition on the highest level. The training process for each of the types of preparation, through evolution of the football game has been changing and adapting according to demands and characteristics of the same game. Evolution of the foodball game is moving towards the intensified activities of players with and without the ball that directly influence the group and team actions in certain standard situations of the game . Significantly higher scope and intensity of movement of players during the game, and on the other hand, reduction of time of the active play are indicators of changes in demands of players necessary for competition on the elite level. Besides, significantly greater closeness of players on the area but also the speed of movement of the ball on the pitch point to changes in organization of the game on the group and the team level .
One of the most systematic overviews of methods of football game analysis was presented by Sarmento et al. (2014), who divided the research activity in football into descriptive, comparative and predictive analysis. The space of description is related to the description of technical, tactical and physical performances of a football player; the comparative analysis mostly take into consideration the level/type of the competition and position of a player in a team, while the prediction of actions based on the already seen is a space in which standard protocols are defined. For all types of analyses, towards determining characteristics of offensive tactics, the structure of attack actions that ended with the shot on the opponent's goal was observed ( Based on the current directions in the football game analysis, a need arose to continuously monitor tactical manifestation at the highest quality competitions of modern football in order to accurately define directions of the game development from the tactical prospective, but also to identify those technical and tactical elements of the game that can be characterized as the coach's surprise for a certain match.
Therefore, the object of this research was the highest quality club competition in football, the Champions League, in two succesive seasons of competition, that is, the tactics of attack of teams which played in the finals. The specificity is the fact that the FC Real Madrid was the winner of the competition in both seasons, so it was interesting to see whether and to what extent Real and other teams changed their tactical plan of the game.
Thus, the aim of this paper was to determine characteristics, as well as to define modern tendencies in relation to the tactics of the attack of teams that compete on the highest level, in this case in the Champions League. Based on the results of studies completed so far, it was expected that certain characteristics of the offensive tactics specific for the competition itself would be sorted out, but also that parameters of the game related to individual preparation, the idea of the coach, specificity of teams that have qualified for the finals of the competition etc. would be discovered.

Research sample
The sample of matches that were analyzed for needs of this study included the final matches of the Champions League (CL) in seasons 2015/16 and 2016/17. Matches played in quarter-finals, semi-finals and finals were observed, i.e. 13 matches of this competition for each season, therefore the total of 26 matches were analyzed.

Sample of variables and ways of data collection
Since the result represents the basic measure of success in football, which is the consequence of successfully completed attack actions of a team, to describe the tactics of attack, variables that point to the structure of successful attacks were observed (all offensive actions were finished by the shoot on the opponent's goal): − Number of shots on the goal (successful attacks); − Accuracy of successful attacks -inacurrate, accurate and efficient; − Time interval of realization: 0 -15 minutes, 15 -30 min., 30 -45 min., 45 -60 min., 60 -75 min., 75 -90 min. and 90 -120 min.; − Zone of initiation of successful attacks -attacks started from the defence, middle or attack zone; − The way the successful attacks started -"cutting the ball", winning the tackle, throw-in (the attack that started by throwing the ball in), free kick play (the attack started after the free kick, and the ball was not directed to the opponent's penalty area), "the second ball" (coming into possession after "winning" the ball that had never been in control by either team), offensive stoppage and attacks started by the goalkeeper's pass; − Types of successful attacks -continuous, counter-attacks and set pieces attacks; − Structure of successful attacks -number of participant players, number of passes, duration of an attack; − The way of coming into position for the shot on the goal (assistance) -individual attack, "the second ball", centre, back pass, long pass; − Zones from which shots on the goal were directed -0 -5 m, 5 -11 m, 11 -16 m and >16 m; − The segment with which the shot was madehead, left leg, right leg.
Together with successful attacks, towards defining the attacking tactics, variables that point to the orientation of the game in situations when the team was in ball possession were monitored: − Total duration of the ball possession during the match, number of ball possession situations, duration of ball possession; − Number of successful passes, percentage of successful passes, the speed of passes (number of precise passes in the unit of time by the achieved possession of the ball). The analysis of previously recorded matches started with creation of the observing protocol (Carling et al., 2005). That protocol was used for each match and each team separately. While observing matches, each successful attactk and action that ended with the shot on the goal was recorded in the observation sheet. Video materials were taken from the TV channels Premium sport -HD, Bein sports -HD, Sky sports -HD, and Premium Calco -HD.

Data processing
Each variable was processed by standard descriptive statistics (total and mean value, as well as standard deviation). Differences in distribution of certain tactical indicators within one group between com-petition seasons 2015/2016 and 2016/2017 were assessed by nonparametric chi-square test, while for examining differences between two independent competitions the Mann-Whitney U test was used in certan variables. The level of statistical significance was p<0.05. All statistical tests were processed using SPSS 17.0 program (SPSS INC Chicago, IL).

RESULTS
Results of this study have shown that teams that appeared in the finals of the competition in the 2016/2017 season were slightly more offensive, which was confirmed by the data according to which the average number of successful attacks was 11.5 compared to the previous season -9.5. The ratio of inacurrate, accurate and efficient shoots towards the goal was 48% -41% -11% in the 2015/2016 season, and 49% -38% -13% in the 2016/2017 season.
Observing successful attacks realized in certain time intervals, it can be noted that the number of such attacks in seasons 2015/16 and 2016/17 was equal, however, in the 2016/2017 season more attacks were realized in the introductory parts of the first and the second halftime ( Figure 1). When the type of the attack is in question, the analysis showed that the application of continuous attacks, counter attacks and successful actions after the break of the game is proportionally represented in both seasons (Table 1). Somewhat smaller percentage of shots to the goal after the break in the season 2016/17 was manifested through the increase in the number of successful attacks from the game (continuous attacks and counter attacks). The largest number of shots on the goal occurred after organization of continuous attacks, and the fact in connection with this data is the average duration of successful attacks that was 13.9 seconds in the 2015/16 season, and 13.57 seconds in the second competition season ( Table 2).
The teams most often started successful attacks in both competition seasons from the attacking third of the pitch, then from the middle zone, and the least number of successful actions started from the defence zone (Table 1). Besides, the analysis shows that in the second season (2016/2017) the number of attacks started from the attacking third is slightly larger compared to the first season (46% -49%) while on the other hand fewer attacks started from the middle zone (36% -32%). Table 1. Zones from which successful actions were initiated, types of realized successful attacks, zones of direction and segments used to shoot on the goal in the Champions League (seasons 2015/ 16 and 2016/17) Results show that shots on goal in both seasons most rarely came from zones 0 -5m and 11 -16m, as well as that players in the observed matches most often shot from the 5 -11m zone and the area >16m (Table 1). However, in the 2015/16 season, the percentage of realized shots from the 5 -11m zone was smaller than from the area >16m, while in the season of 2016/17 the number of shots from the 5 -11m zone was larger compared to shots outside the 16-meter area. Shots performed with a leg were dominant, in most cases the shot was realized using right leg (46% -48%), then left leg (34% -33%) and finaly using head (20% -19%) ( Table 1).
Winning "the second ball" and cutting the ball are the most often used forms to start a successful attack, and the significant number of actions that ended with a shot towards the goal started after organizing an offensive break ( Figure 2).
The analysis of the attack structure, observed according to the number of players who participated and the number of passes during a successful ac-tion showed that in both seasons teams performed successful attacks with 3 -5 players, at which most often there were 4 passes before a shot on the goal ( Table 2).
The total duration of possession of the ball in one game was slightly longer in the 2016/17 season and the difference is 45 seconds per a match, besides, the teams were in ball possession slightly longer in the second season, in average 16.8 seconds ( Table 2).
The dynamics of the play in both seasons was equal, which was confirmed by the results of the speed of passes, but the teams in the second season were more successful in passing the ball from one player to another ( Table 2).
Results further show that the end of successful attacks was mostly different after assistences of centre shots, then after individual actions, long passes, back passes and finally after "the second ball" won. Together with the listed, a significant increase in the case of assistences with centre shots and back passes in the second season is also interested (Figure 3).

DISCUSSION
Results of this study showed that the finals of the Champions League demand a certain level of technical and tactical manifestation, taking into account the fact that there are no significan differences in the structure of successful attacks of teams that ranked into the quarter-finals of this competition in two competition seasons. However, certain technical and tactical elements of play sorted out which are different to a certain extent and in that way they point to certain innovations in preparations of teams for the finals of the competition.
Realization of successful attacks in relation to time intervals of duration showed that there are no differences in the dynamics of the play between two competition seasons, however, it is noted that in the first 15  Application of continuous attack, according to results of this study, represents a dominant form of offensive action organization. If we take into account the data according to which is directy linked with the positive result (Leontijević et al., 2015), it can be said that for ranking in the finals of the competition a tactical commitment in the play according to which passing game and ball possession resulted in the shot on the opponent's goal. For such a tactical plan it is necessary to have players who are firstly technically skilled for such way of playing a game. Thus, teams that base their offensive actions exclusively on a direct play (counter attack) or rely on a break cannot achieve good results continuously. However, such a statement can be further examined in some future studies if we take into account the fact that these matches are eliminatory, and, at the same time, their importance is significantly greater, the play itself is more cautious, teams tend to risk less, primarily trying to score a goal, they lose ball possession less frequently in the phase of the initiation of the attack, etc.
The quality in the game by applying the continous attack, among other things, can be analysed by efficiency in the ball possession manifested through the success in passing the ball. The 2016/2017 competition season was more efficient in relation to the previous one (a larger number of goals), probably contributed by the data according to which the teams were in ball possession for 45 seconds longer during the match. Besides, the second season is characterized by significantly greater success in passing game (p=0.013), while the speed of the ball during passing was equal. Thus, the conclusion is imposed that it is necessary to technically prepare a player for a game as for the successful passing the ball and for playing under pressure in order to achieve results in football at the highest level.

CONCLUSION
The analysis of the finals of the Champions League in seasons 2015/2016 and 2016/2017 showed that there is no clearly defined structure of attacking actions that end with a shot on goal for this level of competition. The application of continuous attacks, a high percentage of success in the passing game with the distribution of the ball, especially in the attacking third of the pitch represent dominanat forms of realization of successful attacks. Also, it can be concluded, based on some differences, that there is an increasing tendency in the offensive tactics to play through the wing positions, to employ the players of impeccable technical skill which increases accuracy in ball passing, and affects efficiency of the final pass and the shot on goal. Besides, there is an increase in number of attacks inititated in the attacking third of the pitch, that is, the application of high-pressing at the beginning of the opponent's attack which has become more frequent choice in the defensive orgranization.
To provide a more detailed analysis and more precise discovering of the game mechanisms, it is necessary to examine the structure of continuous attacks, counter attacks and set pieces in the future studies. Also, using comparative analysis, elimination matches and the group phase of the competition, it is necessary to examine whether and to which extent the importance of the game affects the tactical plan of the game. Of course, the analysis of each subsequent competition season will only determine certain standards in the offensive tactics of football on the highest level, but also point to the actual tendencies of development and improvement.