High metabolic load distance in professional soccer according to competitive level and playing positions

Introduction

Match physical demands monitoring process aims to determine soccer players’ demands during their match practice to replicate them during training sessions and, consequently, optimize their preparation (Castillo et al., 2019; de Dios-Álvarez et al., 2021; Rojas-Valverde et al., 2019). In this vein, the emergence of player tracking technologies has facilitated access to knowledge related to the external demands of professional soccer matches (Cummins et al., 2013). Specifically, the Global Position System (GPS) is considered a valid instrument to record external load demands in elite soccer during training and matches (Mallo et al., 2015). However, different professional leagues use different tracking systems, and not all of them have been validated against a gold-standard instrument (Castellano, Alvarez-Pastor & Bradley, 2014). In addition, these devices have presented problems related to the signal reception and can be uncomfortable for soccer players (Hennessy & Jeffreys, 2018). An interesting alternative to solve GPS-derived problems is the video camera tracking system, which has shown an almost perfect agreement with GPS (Pons et al., 2019). This system is composed of eight super cameras placed strategically throughout each stadium and allows calculating the positions for each player on several axes (Pons et al., 2019). Prior studies conducted in soccer have focused on reporting distances covered using fixed absolute velocity-based thresholds (e.g., sprinting: distance covered above 24.0 km·h⁻¹; Bush et al., 2015; Malone et al., 2015). However, the intermittent nature of soccer match-play implies that analyses based only on velocity thresholds may be incomplete, and it is necessary to consider accelerations/decelerations and the energy cost of changing velocities (Young et al., 2019).

Metabolic power is a theoretical model to estimate the energy cost of acceleration and deceleration in team sports, including soccer (Hoppe et al., 2017; Osgnach et al., 2010). It proposes that accelerated running on level ground is energetically equivalent to running uphill at a constant speed (de Hoyo et al., 2018; di Prampero et al., 2005). In this regard, very high correlations between aerobic fitness variables and metabolic power estimations of high-power distance during professional matches have been found (Manzi, Impellizzeri & Castagna, 2014). However, metabolic power involves all actions, without discriminating them according to their intensity, although the relationship between high-intensity actions and soccer success has been proven (Faude, Koch & Meyer, 2012; Yang et al., 2018). In this context, a novel variable called high metabolic load distance (HMLD) is considered (Tierney et al., 2016). This variable refers to the distance covered with a power consumption above 25.5 W·kg⁻¹ (Tierney et al., 2016). Specifically, this value corresponds to running at a constant velocity of 5.5 m·s⁻¹ or 19.8 km·h⁻¹ on grass. Accelerations or decelerations (e.g., accelerating from 2 to 4 m·s⁻² for 1 s) also are included in HMLD. This variable provides global information about the players’ total high-intensity activities, as it not only includes high velocity running but also accelerations and decelerations, justifying its analysis in soccer (Martín-García et al., 2018).

Recent studies have been conducted focusing on external demands in professional soccer players including HMLD variables. Specifically, Smpokos, Mourikis & Linardakis (2018) analyzed the seasonal changes of the external demands in professional Greek soccer players. These authors reported a maximum value of 1,880 m of HMLD in the most demanding phase of the season (January). On the other hand, Tierney et al. (2016) compared the external demands in professional soccer players according to their playing formation. These authors reported maximum HMLD values of 2,025.0 ± 304 m when 3-5-2 playing formation was used, presenting significantly lower values when other formations were selected (e.g., 1,568.0 ± 257 m with 4-4-2 playing formation). In addition, these authors also reported that forwards achieved the highest HMLD values (2,476 ± 1,339 m) compared to their counterparts. Martín-García et al. (2018) analyzed the most demanding passages of play in soccer matches of professional players and observed values of MHLD per min of 26 ± 6.1 m in 10-min passages. Additionally, they found that wide midfields presented the greatest HMLD/min values. Despite these studies, specific knowledge about HMLD variables is scarce, as the aforementioned studies are relatively old, so the demands of HMLD could be different because the current demands, mainly related to high-intensity actions, have increased in recent years (Pons et al., 2021). Furthermore, there are no studies that combine the analysis of playing positions and levels.

To address the gap identified in the literature, we analyzed the external demands, in terms of HMLD, of Spanish male professional soccer players during two competitive seasons (from 2018/19 to 2019/20). Specifically, this study aimed to examine the Spanish professional soccer players’ HMLD profile, comparing competitive level and playing positions. Based on previous studies (Martín-García et al., 2018; Tierney et al., 2016), we hypothesized that forwards and wide midfields would achieve the highest values of HMLD, whereas higher HMLD values would be observed during First Division matches.

Materials and Methods

Results

Table 1 presents the information about the HMLD variables and the comparison of competitive levels (i.e., First division vs Second division). No significant differences between competitive levels were found for any variable except for HMLD_min and HMLD_min2. Conditional r² values ranged from 0.54 to 0.64 in all models showing a moderate effect of competitive level on match physical demands.

Differences according to playing positions are shown in Tables 2, 3 and Fig. 1. CB presented the lowest values in all variables compared to their counterparts in both levels (p < 0.001). Significant differences were also observed when First Division CB were compared with Second Division CB in the HMLD, HMLD1, and HMLD1_min variables (p < 0.001). Regarding FB, significant differences with CM were observed in all variables (p < 0.05) except for the HMLD2 in the Second Division. Differences between FB and WM were also observed in all variables (p < 0.05) except for HMLD and HMLD2 in the First Division. Differences between FB and FW (p < 0.05) were found in HMLD, HMLD1 (only in the First Division), HMLD2, HMLD1_min, and HMLD2_min (only in the Second Division). Moreover, significant differences were observed when FW in the First Division were compared with the FW in the Second Division in HMLD and HMLD1 variables (p < 0.05). CM presented significant differences (p < 0.05) with WM in all variables except for HMLD_min and HMLD2_min, whereas differences in HMLD1_min were observed only in the First Division. Significant differences between CM and FW were observed in all variables and levels (p < 0.05). In all variables, CM in the First Division presented higher values compared to CM in the Second Division (p < 0.001). Attending to WM, significant differences with FW were found in all variables and levels (p < 0.01). Significant differences between WM in both levels were also found in HMLD and HMLD2 (p < 0.05). Significant differences between FW in both levels were only observed in HMLD1_min (p < 0.05). Finally, conditional r² values ranged from 0.54 to 0.64 in all models showing a moderate effect of playing position on match physical demands.

Discussion

The main aim of this study was to examine Spanish professional soccer players’ HMLD profile, comparing competitive level and playing positions. This is the first research that compares the match physical demands of competitive level and playing positions attending to this variable, which reflects the high-intensity profile of soccer players. No differences between competitive levels were found in any variable when all players were analyzed conjointly except for HMLD_min and HMLD_min2. However, when playing positions were considered, differences between competitive levels were observed in all positions, mainly in the HMLD and HMLD1 variables. Several differences between playing positions were also observed, with CB presenting the lowest values in all variables compared to their counterparts in both competitive levels, whereas CM in the First Division and WM in the Second Division showed the highest values in the HMLD variables.

This novel variable has gained relevance in recent years but the research literature has revealed few studies with professional soccer players. In our study, mean values of 2,347 ± 10.9 m were found when both competitive levels were considered conjointly. These values are higher than those reported by Tierney et al. (2016) and Smpokos, Mourikis & Linardakis (2018), who analyzed England sub-23 league and professional Greek soccer players in official matches and presented HMLD values of 2,025.0 ± 304 m and 1,880 m, respectively. These differences could be because the aforementioned studies are older than ours. Also, several authors have recently concluded that the distance covered at high intensity is increasing in elite soccer (Pons et al., 2021). Additionally, non-significant differences were observed when the first half was compared to the second half, similar to the report of Casamichana et al. (2019), which is justified by the fact that lower distances at high-intensity are covered by soccer players during second halves in official matches (Pons et al., 2021; Rivilla-García et al., 2019). Finally, regarding the comparison of match physical demands between professional soccer leagues, no significant differences were found when the First and Second Division were compared attending to HMLD variables, except for HMLD_min and HMLD_min2. On the contrary, previous research have showed differences in match physical demands between soccer leagues. For example, the Premier League soccer players covered less total distance and high intensity running distances than the lower leagues players (Bradley et al., 2013; Di Salvo et al., 2013). In addition, recent studies have analyzed the Spanish professional soccer leagues with the same aim, and they concluded that there are significant differences, finding that the top-tiered leagues were more physically demanding than the lower (Castellano et al., 2015; Gomez-Piqueras et al., 2019; Pons et al., 2021). However, our results could be influenced by considering all the players conjointly, without differentiating playing positions.

Previous studies have determined that professional soccer players’ external demands during match-play vary according to their specific playing position (Martín-García et al., 2018). Therefore, it is necessary to analyze HMLD considering this differentiation for a deeper knowledge of the match physical demands of the competition. This would allow optimizing the decision-making process for the prescription and manipulation of the training load of professional soccer players. In this study, differences in HMLD were observed according to playing positions, with the CB players presenting the lowest HMLD values in all the variables, in both the First and Second Division. These results support those found by Martín-García et al. (2018), who observed that CB players had the lowest HMLD values (i.e., m·m⁻¹) in all the passages analyzed (i.e., 1, 3, 5 and 10 min) compared to their counterparts. Tierney et al. (2016) also obtained similar results in CB players. This consensus seems to be due to these players’ predominantly defensive roles on the field (e.g., situations near to the goal without longer distances covered at high intensity), which favors their HMLD values being significantly lower than those of the rest of the playing positions. On the other hand, the highest HMLD distances are covered by different playing positions as a function of competitive level, with CM players obtaining the highest values in the First Division, whereas the highest values in the Second Division were observed in WM players. These differences between competitive levels could be related to the differences in the playing styles followed in each division (Pons et al., 2021). However, our results disagree with those obtained in the study conducted by Tierney et al. (2016), in which FW presented the highest HMLD values. When the most demanding passages of match-play are analyzed (Martín-García et al., 2018), WM presented the highest values. These players could be involved in the most determinant actions in soccer, such as goals, which are characterized by the presence of sprints (Faude, Koch & Meyer, 2012; Yang et al., 2018). Finally, when playing positions and competitive levels interacted, differences between competitive levels were observed in all positions, mainly in HMLD and HMLD1 variables. These findings suggest that it is necessary to develop an individualized profile attending to the interaction of positions and competitive levels and thus achieve individualized training tasks to meet the demands of each player.

This study is not without limitations. Firstly, this investigation was conducted only in the top two Spanish professional soccer leagues, so the findings obtained cannot be extrapolated to different leagues, as external demands vary according to the country. Secondly, the teams were not differentiated according to different contextual variables, such as the match location, the quality of the opponent, or the playing style. In addition, would be interesting considered the match status (Andrzejewski et al., 2016) and the evolution of score-line (Redwood-Brown et al., 2012), since research have showed their influence on position-specific match physical demands, showing that central defenders and fullbacks covered shorter distances at high intensity in won matches than in lost matches. Finally, it would also be appropriate to differentiate between teams of high and low competitive density because considering them conjointly could influence the results.

Conclusions

In conclusion, this study has confirmed the lack of HMLD differences between competitive levels when all the players were considered conjointly, except for HMLD_min and HMLD_min2. Differences between competitive levels were obtained in all positions when analyzed isolatedly, mainly in the HMLD and HMLD1 variables. Several differences between playing positions were also observed, with CB presenting the lowest values in all variables compared to their counterparts in both levels, whereas CM in the First Division and WM in the Second Division obtained the highest values in the HMLD variables. In practical terms, this study could serve as a reference for the analysis of HMLD in professional soccer players, improving the adaptation and individualization of the training process in this profile of team sport athletes according to the competitive level and specific playing position of each player.

Supplemental Information