From a physiological point of view, the player's high-intensity effort could be determined, among others, through the oxygen uptake, blood lactate concentration, or heart rate (HR) values. The HR is usually employed as a valid indicator of the physiological load that professional players bear during training and competition (2,3,9,14) because, compared with other indications of exercise intensity, the HR is easy to monitor, is relatively cheap, and can be used in most situations, although it can also be influenced by a number of factors such as dehydration, hyperthermia, and mental stress (1,3).
Nevertheless, generally, the HR behavior is a result of the type of movement, speed displacement, and also the player's game situation. It has been suggested that the maximum elevation of HR values appeared after the player was in direct contact with the opponent in situations such as a 1 × 1 or a tackle (11) and that there are significant differences between the mean HRs during friendly matches, modified games, tactical training, and technical training activities (10). It has also been shown as to how the HR behavior, and other physical and physiological load indicators, is in different small-sided games that vary in space, duration, number of players, or the possibilities of possession of the ball (8,18–20,22,23). The HR values for determining the physiological load of soccer players have usually been expressed as a percentage function of the maximum HR (HRmax) or reserve HR (HRres).
On the other hand, the analysis of maximum intensity efforts is the one that possibly describes best the most important actions in soccer, as they are related to the most critical phases of the match. Although the profiles of activity in soccer are noncyclical, given that discrete actions rarely follow a particular pattern, high-intensity activities differ very little across games (21), despite the short time spent in high-intensity efforts by soccer players (3). With regard to the HR, an effort about 90–95% of HRmax or HRres can be considered a high-intensity effort. However, it could be also considered that a high-intensity effort has been performed when a marked increase in the HR values is experienced in a very short time (i.e., when the HR increases from 145 to 160 b·min−1 in only 5 seconds).
This situation, which we have labeled as marked increase in heart rate (MIHR) has been identified and occurs several times during the competition for professional players, although its magnitude seems to depend on the tactical position of the player in the field (13). Despite its potential utility, a few studies have been found in the scientific literature in which MIHR values have been used to determine the relationship between high-intensity efforts and game situations (12). Partly, this is because the HR monitors do not have functions to measure the MIHR easily.
The purpose of this article is to detect, describe, and compare the occurrence of both MIHR and a more traditional measurement of high-intensity effort: the percentage of maximum HRres (MHRR) to assess the relationship between both measurements and the game situations. Our hypothesis is that MIHR and MHRR (both considered high-intensity physiological efforts) are triggered by different mechanisms, that is, different game situations. Testing for this hypothesis will help in the understanding of the relationship between the physiological response and situational keys of the game, thus helping to develop more appropriate training strategies.
Experimental Approach to the Problem
To detect both types of high-intensity efforts (MIHR and MHRR) and determine the game situations in which they both take place, a research design was developed using a combination of selective and observational methodologies.
Selective Methodology. Unit of Analysis: Heart Rate Values
The unit of analysis considered here is the game situation in which the soccer player is involved when his HR values exhibit 1 of 2 possible high-intensity situations: (a) HR values are close to HRres for that player (MHRR): HR ≥95% of HRres for, at least, 5 seconds. (b) HR values show a marked increase (MIHR): HR rising over HRres by ≥10%, for a maximum time of 15 seconds and HR rising over HRres by ≥5%, for a maximum time of 5 seconds.
The use of HRres as an individualized parameter of physiological effort seems to be one of the most suited measurements, because it implies taking into account the base HR values to calculate the selected HR values for competition. This allows for a better differentiation between players when estimating high-intensity efforts than using HRmax values only.
Observational Methodology. Unit of Observation: The Game Situation
To determine the game situation profile, systematized observational methodology has been employed after the guidelines provided by Jonsson et al. (15) and Castellano et al. (4). The present research uses a diachronic, idiographic, and multidimensional design, with limited follow-up, both intersessions and across sessions. Seven criteria, each of them referring to a different aspect of the game, were taken simultaneously into account to comprehensively describe the game situation. The observation instrument was created using the Match Vision Studio Premium (5) from a combination of field formats and category systems, all of them developed and optimized from simple “ad hoc” categories, which can be easily recognized within the game situation. The criteria and derived categories are summarized in Table 1. For each criterion, the HR condition had to be assigned to 1 of several mutually exclusive categories. Thus, each HR condition was given 7 binary scores, 1 in each condition. These scores described (a) if the ball was in play, (b) the location of the player with respect to the game center, (c) the role of the player, (d) the presence or absence of opposition, (e) the location of the player, (f) the time, and (g) the result.
To facilitate the comprehension and reproducibility of this study, we shall define in a more detailed fashion the different categories and criteria depicted in Table 1:
Ball in play or ball out of play: According to rule IX of soccer, the ball is in play all throughout the match duration, except when it passes a bounding line, when an offence occurs or when play is stopped by the referee.
Game Center: (a) Player inside the Game Center. A player is inside the Game Center when he is interacting with the ball, or ready to interact with it, either to assist or to oppose. (b) Player going to the Game Center/Game Center approaching the player. A player approaches the Game Center when he is moving toward the player interacting with the ball. The Game Center approaches a player when the trajectory of the ball reaches the area occupied by that player. (c) Player outside the Game Center. A player is outside the Game Center when he cannot participate in the Game Center (the area where the player interacts with the ball).
Role: (a) The player close to the ball. The player is close to the ball when he is interacting with it, when he is trying to impede the progression of the player who is in possession of the ball, or when no one is interacting with the ball, but he is in such a position that he can interact with the ball before any other player. (b) Attacking player without ball. An attacking player without the ball belongs to the team in possession of the ball but he is not playing it. (c) Opponent of the player without the ball. This is a member of the team without ball, who is accompanying the attacking player without the ball and intercepting the reception of the ball by the attacking player without the ball.
Opposition: (a) Direct opposition. When the opposition action takes place in the immediate interaction area of the player, against one or more opponents, with the aim of counteracting the opponent's action. (b) Indirect opposition. Same as the previous case but the opposition action takes places outside the immediate interaction area of the player. Indirect opposition also occurs when opposition takes place in the immediate interaction area of the players, but they are far from the ball and from the Game Center. (c) No opposition. This happens when none of the above is observed.
Area: (a) Defensive Area. The whole width of the field from the penalty box to the ending bound line. (b) Middefensive Area. The whole width of the field from the limit of the penalty box to the line tangential to the Center Circle. (c) Central Area. The whole width of the field comprised between the 2 lines tangential to the Center Circle. (d) Midattack Area. Same as Mi-defensive Area for the opposite side of the field. (e) Attack Area. Same as Defensive Area for the opposite side of the field.
The present research involved a total of 441 games situations (241 MIHRs and 200 MHRRs), taken from HR records and individual filming of 12 healthy elite soccer players from 2 Spanish Professional League Teams (4 defenders, 4 midfields, and 4 forwards). All the players had an experience of >100 matches played in First Division. All the players, together with technical staff and supervisors, signed an informed consent form beforehand. The research protocol followed the guidelines provided by the Helsinki Statement about Biomedical Research for Humans (18th Medical Assembly, 1964; revised 1983 in Italy and 1989 in Hong Kong), at the international level, and the Arrangement for Preservation of Human Rights and Dignity with Respect to Biological and Medical Applications (IR 1999; B.O.E. 251, 1999), at the national level. The study was approved by the local ethics committee.
The HRmax values, and basal HR values, were gathered for all the players beforehand, from the Course-Navette test (17) performed for the season by the team technicians 3 weeks before the data gathering. The HR values for all the players were determined from these, as percentages with respect to HRmax and HRres, as defined by Karvonen and Vuorimaa (16). A brief description of the sample of elite soccer players participating in the study is as follows: age 26.88 ± 3.41 years; HRmax 193.7 ± 3.74; base HR 42.6 ± 1.23; HRmax in competition 85.29 ± 3.18%; and HRres in competition 81.3 ± 4.01%.
The players were simultaneously filmed and HR monitored along 7 preseason games (between the end of August and the beginning of September), under very similar field and weather conditions (dry field and an average temperature of 26.16 ± 1.89° C) and against other professional teams. The players did not present symptoms of dehydration o hyperthermia. For the recording of the HR condition, an HRM Polar RS400 was attached to each player when in competition and registered R-R intervals. Subsequently, HR data were dumped in the Polar Precision Performance software, and from this, they were transferred to Microsoft Excel, where were detected manually, among all values included, the 3 behaviors of high-intensity HR described in units of analysis, detecting along the 7 games a total of 441 events of high-intensity efforts (241 MIHRs, and 200 MHRRs). The HR responses varied widely among players (100–192 b·min−1; 52.08–98.49% of HRmax; 38.26–98.1% of HRres).
A simultaneous and synchronized filming of the players during the game was also performed to be able to relate the game situation in the film with HR conditions. Digital video cameras (Sony DCR-PC 250E, miniDV format, with a sample frequency of 60 Hz) were placed at a central and elevated position, and the filming was performed using observers who were trained to be always focused on the player being monitored, together with the ball, and the widest possible portion of the game field.
In a second phase, each sequence of the game, belonging to the 441 high-intensity efforts detected, was digitized and incorporated into the Match Vision Studio Premium (5), and each of them was assigned, by the observers, to one of the existing system of categories for the game situation (Table 1), taking into account the game situation immediately precedent to the observed HR condition. This was done because there is a time gap between the beginning of a high-intensity effort, and the HR condition, which reflects such effort. In this sense, each player seems to show a characteristic pattern for expressing his effort through the HR. This pattern was outlined along several measurements of the time lapse during different high effort training tasks, 3 repetitions, with complete rest, sprint 20 m with ball and without ball, which allowed for effectively adjusting the players' HR behavior to the different game situations.
Observation and coding were performed by a team of 4 observers, who were experts in soccer. With respect to interrater reliability, a high agreement among observers was found in all criteria, with values for Cohen's kappa (7), always >0.985.
At the bivariate level, tests of independence (chi-square statistic) and strength of association were performed to determine the relationship between HR measurements (MIHR and MHRR) and each of the different game criteria (Ball in Play, Game Center, Role, Opposition, Area, Time, and Result). At the multivariate level, those criteria that had a significant bivariate association with HR behavior (p < 0.01) were included as predictors in a stepwise logistic regression analysis, using as inclusion criteria Wald's statistic (p < 0.01).
At the descriptive level, Table 2 shows the percentage of game situations in which the players reach HR values close to the maximum (MHRR) or where marked increases in the HR values have been reached (MIHR). Given that each HR condition has to be assigned to only 1 category on each criterion, all percentages add up to 100% for each category and HR condition.
At this first level of analysis, several interesting differences between both HR conditions can be seen. As has already been noted, for the Ball in play criteria, 28% of situations in which the players reach HR values close to reserve maximum (MHRR), the ball is out of play; for MIHR, the corresponding percentage is 0%. For the Game Center criteria, the player is outside the Game Center in 61.1% of MHRR cases; for MIHR, the corresponding percentage is only 0.9%. On the other side, only in 22.2% of MHRR situations the player is close to the ball (i.e., he has the ball, or gets the ball, or is opposing the player who has the ball); however, for both criteria (Role and Opposition), the percentage is almost 70% for MIHR. No differences between MHRR and MIHR are apparent for the 3 remaining criteria (Area, Time, and Result).
In the second phase, the significance of the association between HR condition and the different game situations was statistically tested. The results are shown in Table 3.
As can be seen in Table 3, and as descriptive analysis already suggested, 4 out of 7 criteria of game situation had a significant association with HR condition (MHRR vs. MIHR): Ball in play, Game Center, Role, and Opposition. These results allow for a characterization of the situational profile for both HR conditions. The MHRR situation is associated with a game situation in which the player is mainly outside the Game Center (61.1%), is not close to the ball (77.8%), and is neither getting nor performing direct opposition (77.8%). Conversely, when the player physiologically shows an MIHR condition in his HR values, it will be associated with a game situation where he is in the Game Center (81.7%), close to the ball (69.6%), and where he gets or performs direct opposition (68.7%).
Among the association measures also represented in Table 3, the uncertainty coefficient expresses in which proportion the uncertainty in assigning 1 player to 1 of the HR conditions is reduced by knowing the corresponding game situation he is involved in. The results showed that the best criteria in reducing the prediction error for the HR situation is Game Center (39%), followed by Opposition (23%), Role (17%), and Ball in play (16%).
These results strongly suggest that it is possible to predict which high-intensity physiological effort the player is bearing according to the game situation he is involved in. Thus, a logistic regression model was tested, using the HR condition as the dependent measure for high-intensity physiological effort, and 3 categories of game situation as predictors. The saturated model can be expressed as follows:
where (a) α high-intensity physiological effort is a dichotomous variable (MHRR vs. MIHR); (b) βGC is the polytomous criteria Game Center; (c) βRo is the polytomous criteria Role; and (d) βOp is the polytomous criteria Opposition. The Ball in play criteria was not included as a predictor in the logistic regression model given that it is a necessary condition for MIHR (100% of MIHR cases appear when the ball is in play; Table 2).
Summary and goodness-of-fit statistics for the logistic regression analysis are shown in Table 4. The amount of variance accounted for by the model ranges from 41 to 56%, if we take pseudo R-squares. The percentage of correctly classified cases is 84.5%, but it is almost perfect if we only take into account MIHR cases (99.1%).
With respect to the model itself, only 1 of the predictors, the Game Center criteria, was retained in the final equation (Table 5), which means that this criterion represents the game situation most relevant to explain the HR condition that a soccer player is facing during competition. The Odds ratios show that there is an extremely high association between MIHR and closeness to the Game Center, so that the chances that a player is experiencing MIHR in the HR values are 172.33 times higher when he is in the Game Center than when he is outside, and 220 times higher when he is approaching the Game Center, or the Game Center is approaching him, than when he is outside.
This study has assessed the type of game situations in which 2 different types of high-intensity efforts (MIHR and MHRR) take place. Our results showed that both types of efforts arise in different game situations. The MIHR is highly associated with situations in which the ball is in play, the player is inside the game center, the player is close to the ball, and performing direct opposition. In summary, the game situation related to the MIHR turns out to be very similar to those described by Yamanaka et al. (24) and Cazorla and et Farhi (6) when they mention high-intensity efforts taking place in crucial parts of the game, directly related to the reception of the ball, shooting, 1 × 1 with and without the ball, and fighting for the ball, and it is also in keeping with the results of Ferret et al. (11), who determined that the maximum elevation of the HR values appeared after the player was in direct contact with the opponent, in situations such as a 1 × 1 or a tackle.
On the other side, the MHRR is highly associated with situations in which the player is outside the game center, far from the ball, and performing indirect or no opposition. A possible interpretation for this fact could be that the player reaches these high HR values as a “physiological debt” because of the accumulation of previous efforts.
Additionally, logistic regression analysis has shown that Game Center is the most important criteria when predicting the occurrences of MIHR vs. MHRR, with 84.5% of correctly classified HR behaviors. These results are consistent with those found by García García et al. (12), where Game Center was the best predictor of the game situation where MIHR values appear; in fact, as we have seen in the odds ratio, the chances of having MIHR values is about 172 times higher when the player is inside the Game Center than when he is outside of it. A possible explanation for this finding is that entering and keeping oneself inside the Game Center requires a great deal of extra effort within a very short time.
These results support our hypothesis that these HR behaviors, both considered high-intensity physiological efforts, are triggered by different mechanisms, that is, game situations. These results highlight that the intensity of the physiological load the player is bearing during competition could be advantageously analyzed using MIHR as one HR indicator, together with the traditional percentages for HRmax or HRres. If we only consider as high-intensity effort those generating an HR response of 90–95% of HRmax or HRres, we are overlooking those efforts situated below that intensity but leading to an MIHR. Furthermore, this would also lead one to underestimate the physiological load the player is actually bearing in competition as a response to relevant game situations.
In summary, our research supports the advantages of using MIHR values as an indicator of the presence of physiological load for professional soccer players in competition. With respect to the game situations associated with MIHR, Game Center is the most important predictor of the different HR values (MHRR vs. MIHR). Players' Role and Opposition are also relevant, although secondary, criteria.
From an applied point of view, these results suggest the design of training exercises in which marked increases in the HR values occur, to get the player familiar with the demands of competition. In this sense, small-sided games seem to be the most useful option. This has been stressed by Rampinini et al. (20), when considering that the intensity of soccer training tasks in small settings can be managed using different types of exercise, different sizes for the game field, and even different incentives are offered by the trainer, encouraging the creation of tasks where the player is alternately inside and outside of the game center, and when inside, the game situation demands that the player is close to the ball and performing direct opposition.
On the other side, the quantification of MIHR provides unique information about the intensity of the competition or training session. The more MIHR a player experiences, the higher the intensity of that competition or session. Nevertheless, the quantification of MIHR can be time consuming for the technical staff, because, currently, there are no HR monitors that can execute this function, and we require a manual search of the behaviors described of the HR. A possible solution would be the inclusion of a specific function to quantify MIHR for monitoring devices.
1. Achten J, Jeukendrup AE. Heart rate monitoring: Applications and limitations. Sports Med 33: 517–538, 2003.
2. Bangsbo J. The physiology of soccer—With special reference to intense intermittent exercise. Acta Physiol Scand 151(Suppl. 619): 1–155, 1994.
3. Bangsbo J, Mohr M, Krustrup P. Physical and metabolic demands of training and match-play in the elite football player. J Sports Sci 24: 665–674, 2006.
4. Castellano J, Hernández Mendo A, Morales V, Anguera MT. Optimizing a probabilistic model of the development of play in soccer. Quality & Quantity. Int J Meth 41: 93–104, 2007.
5. Castellano J, Perea A, Alday L, Hernández Mendo A. Measuring and observation tool in sports. Behav Res Meth 40: 898–905, 2008.
6. Cazorla G, et Farhi A. Exigences physiques et physiologiques actuelles. EPS 273: 60–66, 1998.
7. Cohen J. Weighted kappa: Nominal scale agreement with provision for scaled disagreement of partial credit. Psychol Bull 70: 213–220, 1968.
8. Coutts AJ, Rampini E, Marcora SM, Castagna C, Impellizzeri FM. Heart rate and blood lactate correlates of perceived exertion during small-sided soccer games. J Sci Med Sport 10: 79–84, 2009.
9. Ekblom B. Applied physiology of soccer. Sports Med 3: 50–60, 1986.
10. Eniseler N. Heart rate and blood lactate concentrations as predictors of physiological load
on elite soccer players during various soccer training
activities. J Strength Cond Res 19: 799–804, 2005.
11. Ferret JM, Eclache J, Quard S, Jacquet A, Viret A. Approche énergétique de l´entrâinement d´uneéquipe professionnelle de football. Cinésiologie 77: 15–25, 1980.
12. García García O, Ardá Suárez A, Rial Boubeta A, Domínguez lago E. The behaviour of the heart rate of the professional football player in competition. Is possible to explain it from the context of the game situation
? Motricidad. Eur J Hum Mov 19: 37–68, 2007.
13. García García O, Ardá Suárez A, Rial Boubeta A, Domínguez lago E. Is different the behaviour of the professional football players heart rate in competition according to the tactical position of the player in the field? APUNTS. Phys Educ Sports 90: 42–50, 2007.
14. Hoff J, Wisloff U, Engen L, Kemi O, Helgerund J. Soccer specific aerobic endurance training. Br J Sports Med 33: 218–221, 2002.
15. Jonsson GK, Anguera MT, Blanco-Villaseñor A, Losada JL, Hernández-Mendo A, Ardá T, Camerino O, Castellano J. Hidden patterns of play interaction in soccer using SOF-CODER. Behav Res Meth 38: 372–381, 2006.
16. Karvonen J, Vuorimaa T. Heart rate and exercise intensity during sports activities. Practical application. Sports Med 5: 303–311, 1988.
17. Leger LA, Mercier D, Gadoury C, Lambert J. The multistage 20 meter shuttle run test for aerobic fitness. J Sports Sci 6: 93–101, 1988.
18. Little T, Williams AG. Measures of exercise intensity during soccer training
drills with professional soccer players. J Strength Cond Res 21: 367–371, 2007.
19. Mallo J, Navarro E. Physical load imposed on soccer players during small-sided training games. J Sports Med Phys Fitness 48: 166–171, 2008.
20. Rampinini E, Impellizzeri FM, Castagna C, Abt G, Chamari K, Sassi A, Marcora SM. Factors influencing physiological responses to small-sided soccer games. J Sports Sci 25: 659–666, 2007.
21. Reilly T. The physiological demands of soccer. In: Soccer & Science. In An Interdisciplinary Perspective. Bangsbo J. ed. Copenhagen, Denmark: Munksgaard Institute of Exercise and Sport Sciences. University of Copenhagen, 2000. pp. 91–105.
22. Sampaio J, García G, Macas V, Ibañez S, Abrantes C, Caixinha P. Heart rate and perceptual responses to 2x2 and 3x3 small-sided youth soccer games. J Sports Sci Med 6: 121–122, 2007.
23. Tessitore A, Meeusen R, Piacentini MF, deMarie S, Capranica L. Physiological and technical aspects of “6-a-side” soccer drills. J Sports Med Phys Fitness 46: 36–43, 2006.
24. Yamanaka K, Haga S, Shindo M, Narita J, Koseki S, Matsuura Y, Eda M. Time and motion analysis in top class soccer games. In: Science and Football. Reilly T., Lees A., Davids K., Murphy W.J., eds. London, United Kingdom: E & FN. SPON, 1988. pp. 334–340.