Effects of Match Location, Quality of Opposition and Match Outcome on Match Running Performance in a Portuguese Professional Football Team

his study was to analyze the effects of match location, quality of opposition and match outcome on match running performance according to playing position in a Portuguese professional football team. Twenty-three male professional football players were monitored from eighteen Portuguese Football League matches during the 2019–2020 season. Global positioning system technology (GPS) was used to collect time-motion data. The match running performance was obtained from ve playing positions: central defenders (CD), fullbacks (FB), central mid elders (CM), wide mid elders (WM) and forwards (FW). Match running performance was analyzed within speci c position and contextual

were monitored from eighteen Portuguese Football League matches during the 2019–2020 season. Global positioning system technology (GPS) was used to collect time-motion data. The match running performance was obtained from ve playing positions: central defenders (CD), fullbacks (FB), central mid elders (CM), wide mid elders (WM) and forwards (FW). Match running performance was analyzed within speci c position and contextual factors using one-way analysis of variance (ANOVA) for repeated measures, standardized (Cohen) differences and smallest worthwhile change. CM and WM players covered signi cantly greater total distance (F = 15.45,p= 0.000, 2= 0.334) and average speed (F = 12.79,p< 0.001, 2= 0.294). WM and FB players covered higher distances at high-speed running (F = 16.93,p= 0.000, 2= 0.355) and sprinting (F = 13.49;p< 0.001, 2= 0.305). WM players covered the highest number of accelerations (F = 4.69,p< 0.001, 2= 0.132) and decelerations (F = 12.21, p< 0.001, 2= 0.284). The match running performance was in uenced by match location (d =0.06–2.04; CI: 0.42–2.31; SWC = 0.01–1.10), quality of opposition (d =0.13–2.14; CI: –0.02–2.60; SWC = 0.01–1.55) and match outcome (d =0.01–2.49; CI: 0.01–2.31; SWC = 0.01–0.35). Contextual factors in uenced the match running performance with differential effects between playing positions. This study provides the rst report about the contextual in uence on match running performance in a Portuguese professional football team. Future research should also integrate tactical and technical key indicators when analyzing the match-related contextual in uence on match running performance. Keywords:physical performance; activity pro le; time-motion; match analysis; team sports 1. Introduction Football is an intermittent team sport characterized by high physiological demands [1]. Elite players were found to cover 9–14 km in total during an of cial football match [2,3]. The high-intensity activity (>19.8 km h –1 ) represents 8–10% of the total distance completed, since most movement activities are carried out in low-intensity zones [4,5]. The distances covered at high intensities are a valid indicator to evaluate physical performance in profes- sional football given its relationship with the training process [6,7]. High-speed running, Entropy2021,23, 973.

) represents 8–10% of the total distance completed, since most movement activities are carried out in low-intensity zones [4,5]. The distances covered at high intensities are a valid indicator to evaluate physical performance in profes- sional football given its relationship with the training process [6,7]. High-speed running, Entropy2021,23, 973.

Entropy2021,23, 973 2 of 16 sprints, tackles, impact accelerations and decelerations occur intermittently in a match-play, which require greater physiological and neuromuscular demands [8]. Researchers' and practitioners' interest in the physical performance has been growing over the last four decades at the professional football level [9,10]. Monitoring players' work-rate pro les during training and competition has become more practicable with computer-aided time-motion approaches [5–7]. Additionally, using tracking systems to monitor match demands has become a hot topic of research, referring to work rate, activity pro le or match running performance [5,9,11,12]. Several studies quanti ed the match running performance across national professional leagues, such as the English [13–21], Italian [3,22,23], Spanish [19,24–26], French [20,27,28], German [29–32], Brazilian [33,34], Norwegian [35,36], Danish [37] and Australian leagues [38,39]. The literature also focused on the European Champions League [40–44], UEFA Cup/Europe League [41,44] and the World Cup [45–47]. Current research has also demonstrated an in uence of position on the players' match demands [15,19,25,26,48], and further, the football game's evolution has demonstrated a position-speci c physical increase over time [11,49,50]. Generally, central mid elders covered more distance, and wide mid elders covered more distance at high-intensity zones [13,51]. The central defenders and wide defenders covered more distance at low-intensity zones [51]. Forwards sprint signi cantly less frequently than central defenders [21]. Central defenders performed signi cantly fewer explosive and leading sprints [13]. Accelerations contributed to 7–10% of the player workload for all playing positions during a match-play, while decelerations represented 5–7% [52]. Nevertheless, interpreting match running performance should consider the in uence of contextual, environmental or situational factors [24,53–55]. Studies have pointed to a strong in uence of contextual factors on the match running performances from top football national leagues and continental competitions [24,30,56–64]. Hence, independent and interactive potential effects have been reported for match running performance according to match location, quality of opposition and match status in professional football [59,65]. Contextual factors have a potential in uence on the relationship between match running and the overall performance dimension [55]. Thus, match running performance shall be adjusted according to the intended contextual factors

[24,30,56–64]. Hence, independent and interactive potential effects have been reported for match running performance according to match location, quality of opposition and match status in professional football [59,65]. Contextual factors have a potential in uence on the relationship between match running and the overall performance dimension [55]. Thus, match running performance shall be adjusted according to the intended contextual factors underlying the match-play [24,53,55]. Indeed, elite players normally cover less high-intensity distances when winning [66]. Total distance covered by players was found to be higher when playing at home and against high-ranked teams [24,59]. Linking players' behaviors and match outcomes in speci c contexts has been identi ed as a crucial insight to develop speci c game strategies or training designs [11]. To the best of our knowledge, no studies have analyzed the in uence of contextual factors on match running performance in a professional Portuguese football competition. Therefore, the aim of this study was to analyze the effects of match location, quality of opposition and match outcome on match running performance according to playing position in a Portuguese professional football team. It was hypothesized that the contextual factors and speci c playing positions in uence the match running performance. 2. Materials and Methods 2.1. Participants and Match Sample Twenty-three male professional football players (age: 32.02 1.19 years; height: 1.82 0.01 m; weight: 74.74 0.53 kg) participated in eighteen Portuguese Second League (Leadman LigaPro ® ,Lisbon, Portugal) matches (8 home and 10 away) during the 2019–2020 season. The sampled players were characterized to one of ve playing positions (goalkeeper was excluded): central defenders (CD), fullbacks (FB), central mid elders (CM), wide mid elders (WM) and forwards (FW). The numbers of subjects in the different subgroups were: CD (n= 6), FB (n= 4), CM (n= 5), WM (n= 5) and FW (n= 3). The playing positions were organized into ten dyads: CD vs. FB, CD vs. WM, CD vs. CM, CD vs. FW, FB vs. WM, FB vs. CM, FB vs. FW, CM vs. WM, CM vs. FW and WM vs. FW. The match data correspond to the observations of the

(n= 4), CM (n= 5), WM (n= 5) and FW (n= 3). The playing positions were organized into ten dyads: CD vs. FB, CD vs. WM, CD vs. CM, CD vs. FW, FB vs. WM, FB vs. CM, FB vs. FW, CM vs. WM, CM vs. FW and WM vs. FW. The match data correspond to the observations of the seven out eld players for each match in the

Entropy2021,23, 973 3 of 16 same team (n= 128). The analysis has only considered the players who were part of the starting line-up and performed the entire match duration. The substituted players and non-starting players were not analyzed. The number of observations per position role was: CD (n= 36), FB (n= 31), CM (n= 33), WM (n= 19) and FW (n= 9). The match data showed 3 wins, 9 draws and 5 loses, with a total of 13 goals scored and 15 goals conceded by the sampled team. The matches (2 45 0 ) were performed in of cial stadiums (FIFA standard, natural grass, ~100 70 m), between 10:00 AM and 08:00 PM, and the mean environment temperature was 14.9 5.3 C. All participants were informed about the aim and risks in the investigation. The study includes only the players that have signed the informed consent, and was conducted according the ethical standards of the Declaration of Helsinki. The experimental approach was approved and followed by the Technical and Scienti c Board of the Douro Higher Institute of Educational Sciences. 2.2. Data Collection and Procedures The seven main players were monitored in each match using a portable GPS through- out the whole match duration (STATSports Apex ® , Newry, Northern Ireland). The GPS device provides raw position velocity and distance at 10 Hz sampling frequencies, includ- ing accelerometer (100 Hz), magnetometer (10 Hz) and gyroscope (100 Hz). Each player wore the micro-technology inside a mini-pock of a custom-made vest supplied by the manufacturer, which was placed on the upper back between both scapulae. All devices were activated 30 min before the match data collection to allow an acceptable clear recep- tion of the satellite signal. Respecting the optimal signal to the measurement of human movement, the match data considered eight available satellite signals as the minimum for the observations [67]. The validity and reliability of the global navigation satellite systems (GNSS), such as the GPS tracking, have been well-established in the literature [67–69]. The current variables and thresholds have a small error of around 1–2% reported for the 10 Hz

the measurement of human movement, the match data considered eight available satellite signals as the minimum for the observations [67]. The validity and reliability of the global navigation satellite systems (GNSS), such as the GPS tracking, have been well-established in the literature [67–69]. The current variables and thresholds have a small error of around 1–2% reported for the 10 Hz STATSports Apex ® devices [68]. 2.3. Contextual Factors Contextual factors were codi ed based on three independent variables: match loca- tion, quality of opposition and match outcome. These contextual dimensions have been extensively documented in the literature [54,65]. Match location was split into “home” and “away”, based on when the team under analysis was playing at home or away. Quality of opposition was classi ed from “high-ranking” (i.e., from 1st to 5th position in the league ranking), “medium-ranking” (i.e., from 6th to 12th position in the league ranking) and “low-ranking” (i.e., from 13th to 18th position in the league ranking). Quality of opposition was classi ed according to the nal standing of the 2019–2020 season. Match outcome was analyzed according to “lose”, “draw” or “win” at the end of the match-play. 2.4. Physical Load Measures The match running performances were obtained with the following time-motion data using physical load measures: total distance (TD) covered (m), average speed (AvS) expressed in distance covered per minute (m min 1 ), high-speed running (HSR) distance (m), number of sprints (SPR), number of accelerations (ACC) and number of decelerations (DEC). The GPS software only provided information about the locomotor categories above 19.8 km h 1 : HSR (19.8–25.1 km h 1 ) and SPR (>25.1 km h 1 ). Both acceleration variables (ACC and DEC) considered the movements made in the maximum intensity zone (3 m s 2 ): ACC (>3 m s 2 ) and ACC (<3 m s 2 ). The high-intensity activity thresholds were adapted from previous studies [6,7]. 2.5. Statistical Analysis For descriptive statistics, the Kolmogorov–Smirnov and Levene's tests were used to test the normality and homogeneity, where a normal distribution was observed. Differences

(3 m s 2 ): ACC (>3 m s 2 ) and ACC (<3 m s 2 ). The high-intensity activity thresholds were adapted from previous studies [6,7]. 2.5. Statistical Analysis For descriptive statistics, the Kolmogorov–Smirnov and Levene's tests were used to test the normality and homogeneity, where a normal distribution was observed. Differences

Entropy2021,23, 973 4 of 16 between playing positions, contextual factors and match running performance were tested with one-way analysis of variance (ANOVA) for repeated measures. When a signi cant difference occurred, Bonferroni post-hoc tests were used to identify localized effects. Dun- nett's T3 post-hoc tests were applied if variances were not homogeneous. Bonferroni post hoc was performed to evaluate TD, rHSR, SPR and AvS. The Dunnett's T3 post-hoc was executed for ACC and DEC. Standardized effect sizes (ES) were calculated by Cohen's d, and the thresholds were classi ed as: 0.2, trivial; 0.6, small; 1.2, large; >2.0, very large [70,71]. Smallest worthwhile change (SWC) was calculated as 0.2 multiplied by standard deviation (SD). Additionally, trivial area was calculated from the SWC determined as 0.2 times the between-playing positions [72]. Statistical signi cance was set atp< 0.05. Data are presented as the mean SD. Mean differences (D) are presented in absolute values. All statistical analyses were conducted using IBM SPSS Statistics for Windows (Version 27.0., IBM Corp, Armonk, NY, USA). ES calculations were performed with G*Power (Version 3.1.5.1 Institut für Experimentelle Psychologie, Düsseldorf, Germany). Data visualization was produced using GraphPad Prism (GraphPad Software, Inc., San Diego, CA, USA). 3. Results 3.1. Effects of Contextual Factors on Match Running Performance The descriptive statistics of match running performance according to competitive stage, match location quality of opposition and match outcome are presented in Table. Table 1.Mean match running performance according to contextual factors. Match Location (n= 128) Quality of Opposition (n= 128) Match Outcome ( n= 128) Measures Away (n= 60) Home (n= 68) Low-Rank (n= 36) Medium- Rank (n= 41) High-Rank (n= 51) Lose (n= 61) Draw (n= 36) Win (n= 31) TD (km) 10.91 0.83 10.95 0.81 10.90 0.79 10.86 0.73 10.99 0.91 10.89 0.84 10.92 0.78 11.00 0.85 AvS (m min 1 ) 0.63 0.23 0.66 0.25 0.59 0.24 0.62 0.24 0.69 0.24 0.064 0.23 0.66 0.28 0.64 0.24 rHSR (m) 68.62 15.23 64.17 20.41 69.94 15.73 61.00 19.95 68.57 16.91 67.90 15.41 61.56 21.27 69.61 17.64 SPR (n) 88.74 23.48 81.32 23.60 85.53 20.39 84.41 26.61 85.75 23.93 87.29

10.99 0.91 10.89 0.84 10.92 0.78 11.00 0.85 AvS (m min 1 ) 0.63 0.23 0.66 0.25 0.59 0.24 0.62 0.24 0.69 0.24 0.064 0.23 0.66 0.28 0.64 0.24 rHSR (m) 68.62 15.23 64.17 20.41 69.94 15.73 61.00 19.95 68.57 16.91 67.90 15.41 61.56 21.27 69.61 17.64 SPR (n) 88.74 23.48 81.32 23.60 85.53 20.39 84.41 26.61 85.75 23.93 87.29 20.41 78.08 23.89 89.58 28.21 ACC (n) 40.32 13.48 42.03 15.33 39.28 14.15 39.95 14.09 43.37 14.66 40.48 13.44 42.28 16.19 41.07 14.19 DEC (n) 0.09 0.01 0.09 0.01 0.09 0.01 0.09 0.01 0.09 0.01 0.09 0.01 0.09 0.01 0.09 0.018 ACC—accelerations; ALL—overall independent position group; AvS—average speed; CD—central defenders; CM—central mid eld- ers; DEC—decelerations; FB—fullbacks; FW—forwards; rHSR—relative high-speed running; SPR—sprints; TD—total distance; WM— wide mid elders. Tables according to playing positions. Standardized (Cohen) differences, 95% CI and SWC for each contextual factor are presented in Figure trivial to very large effects by match location (d =0.06–2.04; CI: 0.42–2.31; SWC = 0.01–1.10), quality of opposition (d =0.13–2.14; CI: 0.02–2.60; SWC = 0.01–1.55) and match outcome (d =0.01–2.49; CI: 0.01–2.31; SWC = 0.01–0.35). Quality of opposition's influence had a very large effect on TD for WM vs. FW (d= 2.14, CI: 1.88–2.40; SWC = 0.30). Match outcome had a very large effect on rHSR for CD vs. FB (d= 2.12, CI: 1.97–2.27; SWC = 0.17) and CD vs. WM (d= 2.49, CI: 2.38–2.60; SWC = 0.13). CD vs. WM also showed a very large result of the quality of the opposition's influence for DEC (d= 2.14, CI: 1.97–2.31; SWC = 0.19).

Entropy2021,23, 973 5 of 16 Table 2.Cohen'sd, 95% con dence intervals and smallest worthwhile changes for the in uence of match location on match running performance according to playing positions. Variables Playing Positions Measures Inference CD vs. FB CD vs. WM CD vs. CM CD vs. FW FB vs. WM FB vs. CM FB vs. FW CM vs. WM CM vs. FW WM vs. FW TD (km) d 0.66 0.81 1.57 0.21 0.73 0.97 0.54 0.33 1.64 1.58 95% CI 0.55–0.77 0.57–1.05 1.32–1.82 0.17–0.25 0.60–0.86 0.80–1.14 0.38–0.70 0.27–0.39 1.41–1.87 1.29–1.87 SWC 0.13 0.28 0.29 0.04 0.15 0.20 0.19 0.07 0.27 0.34 AvS (m min 1 ) d 0.55 0.58 0.92 0.06 0.49 0.98 1.17 0.52 1.56 1.26 95% CI 0.37–0.73 0.52–0.64 0.87–0.97 0.03–0.09 0.44–0.54 0.92–1.04 1.14–1.20 0.51–0.53 1.55–1.57 1.23–1.29 SWC 0.21 0.06 0.06 0.03 0.06 0.06 0.04 0.01 0.01 0.03 rHSR (m) d 0.66 0.81 1.57 0.21 0.73 0.97 0.54 0.33 1.64 1.58 95% CI 0.50–0.82 0.78–0.84 1.52–1.62 0.12–0.30 0.70–0.76 0.96–0.99 0.42–0.57 0.19–0.47 1.62–1.66 1.55–1.61 SWC 0.18 0.03 0.06 0.10 0.04 0.01 1.10 0.16 0.02 0.03 SPR (n) d 1.38 1.29 0.51 0.43 0.79 0.59 0.79 1.21 0.15 1.39 95% CI 1.33–1.43 1.27–1.31 0.49–0.53 0.38–0.48 0.78–0.80 0.58–0.59 0.23–1.35 1.14–1.28 0.07–0.23 1.00–1.78 SWC 0.06 0.03 0.03 0.05 0.02 0.01 0.64 0.08 0.09 0.45 ACC (n) d 0.38 0.33 0.20 0.59 0.87 0.30 0.52 0.92 0.59 0.67 95% CI 0.51–0.61 1.27–1.37 0.56–0.70 0.48–0.49 1.92–2.16 0.55–0.78 0.67–0.85 0.97–1.01 1.60–1.71 1.04–1.08 SWC 0.06 0.06 0.08 0.01 0.14 0.13 0.10 0.02 0.07 0.03 DEC (n) d 0.56 0.56 0.63 0.49 2.04 0.66 0.76 0.99 1.65 1.06 95% CI 0.01–1.14 1.98–2.31 1.14–1.27 0.55–0.62 1.26–1.62 0.53–0.63 0.51–0.61 1.25–1.34 1.89–1.99 0.67–1.40 SWC 0.08 0.19 0.08 0.04 0.21 0.06 0.05 0.05 0.06 0.42 Abbreviations: ACC—accelerations; AvS—average speed; CD—central defenders; CI—con dence intervals; CM—central mid elders; d—Cohen differences; DEC—decelerations; FB—fullbacks; FW—forwards; rHSR—relative high speed running; SPR—sprints; SWC—smallest worthwhile changes; TD—total distance; WM—wide mid elders.

This study analyzes how match context affects running performance in professional football.