Relationships between performance test and match-related physical performance parameters

A study in professional soccer players across three seasons
  • Stefan Altmann
  • Maximilian Kuberczyk
  • Steffen Ringhof
  • Rainer Neumann
  • Alexander Woll
Main Article
  • 203 Downloads

Abstract

Background

The purpose of this study was to analyze the relationship between performance test parameters and match-related physical performance in professional soccer players.

Methods

To determine physical capacity, 28 male soccer players underwent several performance tests at the start of the seasons 2013/2014, 2014/2015, and 2015/2016. The following parameters were assessed: maximum running velocity (vmax), fixed (v4mmol/l) and individual anaerobic threshold (vIAS) during an incremental treadmill test; 30-m sprint time in a linear sprint test (LS30m); in a repeated-sprint test, the 30-m sprint time (RST30m) and performance decrement (RSTdecr); and countermovement jump height (CMJ). Match physical performance was quantified during the first ten official matches of each season using a computerized, camera-based tracking system. The following measures of match physical performance were considered: top running speed (TS), mean running speed (vØ), total distance covered (TD), number of sprints (SP), number of high-intensity running (HIR), and aerial duels won (AD+). Pearson correlation coefficients were used for statistical analysis.

Results

Moderate to very large correlations were found between the majority of performance test parameters and match performance variables, with a variability of correlations across the three seasons. Large relationships across all three seasons were only observed between vmax and TD, vmax and vØ, LS30m and TS as well as RST30m and TS.

Conclusion

This study demonstrates the relationship between several performance test parameters and match-related physical performance in professional soccer players, thereby supporting the test parameters’ criterion validity. vmax, LS30m, and RST30m seem to be the most consistent parameters.

Keywords

Exercise test Football Athletic performance Validity Computer assisted image processing 

Zusammenhänge zwischen leistungsdiagnostischen Parametern und physischer Spielleistung

Eine Studie bei Profifußballern über drei Spielzeiten

Zusammenfassung

Hintergrund

Ziel der vorliegenden Studie war es, den Zusammenhang zwischen Leistungstestparametern und der spielbezogenen körperlichen Leistung bei Berufsfußballern zu analysieren.

Methoden

Zur Bestimmung der körperlichen Leistungsfähigkeit absolvierten 28 männliche Fußballspieler verschiedene Leistungstests zu Beginn der Spielzeiten 2013/2014, 2014/2015 und 2015/2016. Folgende Parameter wurden ermittelt: maximale Laufgeschwindigkeit (vmax), fixe (v4mmol/l) und individuelle anaerobe Schwelle (vIAS) in einem Laufbandstufentest; 30-m-Sprint-Zeit in einem linearen Sprinttest (LS30m); 30-m-Sprint-Zeit (RST30m) und Leistungsrückgang (RSTdecr) in einem Sprintwiederholungstest; Sprunghöhe in einem Countermovement-Jump (CMJ). Die spielbezogene physische Spielleistung wurde während der ersten zehn offiziellen Spiele jeder Spielzeit mithilfe eines computergestützten, kamerabasierten Tracking-Systems quantifiziert. Die folgenden Maße der spielbezogenen physischen Spielleistung wurden berücksichtigt: Spitzenlaufgeschwindigkeit (TS); mittlere Laufgeschwindigkeit (vØ); zurückgelegte Gesamtstrecke (TD); Anzahl der Sprints (SP); Anzahl der hochintensiven Läufe (HIR); gewonnene Luftzweikämpfe (AD+). Für die statistische Analyse wurden Pearson-Korrelationskoeffizienten herangezogen.

Ergebnisse

Moderate bis sehr starke Korrelationen wurden zwischen der Mehrzahl der leistungsdiagnostischen Parameter und der physischen Spielleistung gefunden, wobei eine Variabilität der Korrelationen über die drei Spielzeiten hinweg erkennbar war. Starke Zusammenhänge über alle drei Spielzeiten zeigten sich nur zwischen vmax und TD, vmax und vØ, LS30m und TS sowie RST30m und TS.

Schlussfolgerung

Die vorliegende Studie zeigt den Zusammenhang zwischen verschiedenen leistungsdiagnostischen Parametern und der physischen Spielleistung bei Profifußballern. Damit stützt sie die Kriteriumsvalidität der leistungsdiagnostischen Parameter. vmax, LS30m und RST30m scheinen die beständigsten Parameter zu sein.

Schlüsselwörter

Belastungstest Fußball Sportliche Leistung Validität Computergestützte Bildverarbeitung 

Introduction

The multidimensional nature of soccer requires its players to complete different movement patterns, such as jogging, sprinting, turning, and jumping. These patterns are performed repeatedly throughout the game, thereby demanding the players’ full physical capacities (e.g., endurance, speed, and power) (Mohr, Krustrup, & Bangsbo, 2003). To assess and monitor individual changes in these capacities, several performance tests have been developed (Chamari et al., 2004; Stølen, Chamari, Castagna, & Wisløff, 2005; Svensson & Drust, 2005). Over the past years, performance testing has become an increasingly fundamental part of professional soccer (Turner et al., 2011). Concomitantly, the use of technology in soccer to monitor, forecast, and adjust the cardiovascular and neuromuscular stress imposed by training or matches is increasing in relevance (Bradley et al., 2009; Halson, 2014). Therefore, physical performance parameters, such as running distances at different intensities and frequency of accelerations, are assessed during training and matches (Bradley et al., 2009; Halson, 2014; Trewin, Meylan, Varley, & Cronin, 2017).

Many widely-used physical performance tests focus on endurance, speed, and power (Faude, Schlumberg, Fritsche, Treff, & Meyer, 2010; Stølen et al., 2005; Turner et al., 2011). Their relevance predominantly arises from trying to mimic demands during matches. In this context, speed tests investigate the ability to perform (repeated) sprints (Turner et al., 2011). (Incremental) endurance tests examine the ability to cover distances at various intensities and have been shown to represent the demands on different energy systems of matches (Turner et al., 2011). In addition, power tests may be expected to reflect jumps or rapid turns (Stølen et al., 2005).

The parameters assessed are dependent on the specific performance test conducted. During speed tests, times to complete shorter (e.g., 5 or 10 m) and longer (e.g., 30 or 40 m) distances while sprinting are usually measured. In addition, the ability to maintain high speeds over the course of a series of repeated sprints is another point of interest. Results can be reported as mean times over the sprint series or performance decrements (e.g., best vs. worst sprints) (Dawson, 2012; Faude et al., 2010; Turner et al., 2011). In terms of aerobic endurance tests, the maximum velocity reached is frequently reported. When possible, the addition of further parameters gives a more comprehensive view on the athletes’ aerobic endurance capacity. These parameters include the anaerobic threshold, using lactate analysis, or maximal oxygen uptake and running economy, using gas analysis (Faude et al., 2010; Hoppe et al., 2013; Turner et al., 2011). Finally, power is commonly examined via vertical jump tests, with jump height being a standard parameter assessed (Faude et al., 2010; Turner et al., 2011).

The criterion validity of physical performance tests describes their ability to make inferences about performance in competitive conditions, which for soccer is match-related physical performance (Currell & Jeukendrup, 2008). Based on the frequent use and the acceptance of many performance tests, such as endurance, speed, and power tests, one may assume that their criterion validity has already been investigated. As mentioned above, for a long time, the relevance of the tests only followed their logical validity, as the test requirements reflect the demands during matches (Rampinini et al., 2007a).

As a consequence, an increasing amount of research has been published in recent years, focusing on verifying possible associations between field or laboratory-based tests with match performance (Aquino, Palucci Vieira, Paula Oliveira, Cruz Goncalves, & Pereira Santiago, 2017; Buchheit, Mendez-Villanueva, Simpson, & Bourdon, 2010; Castagna, Impellizzeri, Cecchini, Rampinini, & Alvarez, 2009; Castagna, Manzi, Impellizzeri, Weston, & Barbero Alvarez, 2010; Mendez-Villanueva & Buchheit, 2011; Pekas, Trajković, & Krističević, 2016; Rampinini et al., 2007a; Rebelo, Brito, Seabra, Oliveira, & Krustrup, 2014). In professional youth soccer players, small to very large correlations were found between distances covered at several running intensities during matches and different aerobic and anaerobic endurance-based field tests (Aquino et al., 2017; Buchheit et al., 2010; Castagna et al., 2009; Castagna et al., 2010; Mendez-Villanueva & Buchheit, 2011; Pekas et al., 2016; Rebelo et al., 2014). Large relationships between high-intensity running during matches and both aerobic and anaerobic endurance capacity have also been observed in top-level soccer players (Bradley et al., 2011b; Krustrup et al., 2003; Rampinini et al., 2007a). However, no study has investigated if these associations vary over more than one season. In fact, this issue is especially important since match-related performance is highly variable because of several influencing factors (Trewin et al., 2017). Therefore, relationships between physical capacities and match-related performance are complex and might possibly vary over a longer time period. Furthermore, the studies mentioned above have mainly focused on aerobic and anaerobic endurance, while associations of sprint and power abilities with match performance have scarcely been investigated. This becomes especially evident when considering professional soccer players. In this context, Rampinini et al. (2007a) reported large relationships between the mean time of a repeated-sprint test and sprinting distance, but not for the best trial of the same test and top speed during games, as well as power (assessed by a vertical jump test) and any match variables.

Therefore, the aim of this study was to analyze the relationship between performance tests, including aerobic endurance, sprint, and power-related assessments, and selected match-related physical performance parameters in professional German soccer players over three consecutive seasons. The findings of this study could help practitioners to decide which tests and specific parameters to choose during physical performance testing and, further, to design appropriate physical training programs.

Methods

Subjects

A total of 28 professional male soccer outfield players (age: 24 ± 2 years, height: 179 ± 7 cm, body mass: 74.6 ± 8.7 kg) were included in the study. All players were members of the same elite soccer club playing in the second German division.

Procedures

At the beginning of the competitive seasons 2013/14, 2014/15, and 2015/16, all players underwent a performance testing battery. The reliability (expressed as a coefficient of variation) of the tests included has been reported in literature as follows: 0.9% for an incremental treadmill test, 0.9% for a linear sprint test, 0.7–11.0% for a repeated-sprint test (depending on the parameter analyzed), and 2.0% for a vertical jump test (Hopkins, Schabort, & Hawley, 2001; Oliver, 2009). The tests were performed in two separate sessions over two consecutive days. The incremental treadmill test and vertical jump test were completed on the first day, while the linear sprint test and repeated-sprint test were performed on the second day. All players were familiar with the tests used in this study and had performed them previously. The first ten official matches of each season were used for the analysis of match performance parameters.

Performance tests

Incremental treadmill test.

All athletes completed an incremental treadmill test until exhaustion on a Woodway treadmill (Woodway GmbH, Weil am Rhein, Germany). The test started with a running speed of 6 km/h which was increased by 2 km/h every 3 min. Between each stage, capillary blood was collected from the earlobe within 30 s. In addition, heart rate was measured with a Polar system (Polar Electro Oy, Kempele, Finland) throughout the whole test. Subjects were encouraged to complete as many stages as possible, whereat the test was finished at the point of volitional exhaustion. Trials in which players were not able to perform the test until maximum exhaustion (e.g., because of illness or injury) were excluded from analysis. Blood lactate concentration was analyzed with Biosen C-Line Sport (EKF-diagnostic GmbH, Barleben, Germany). Besides the maximal velocity reached (vmax), velocities at the fixed and individual anaerobic thresholds (v4mmol/l and vIAS, respectively) were automatically determined by Ergonizer Software (K. Roecker, Freiburg, Germany). While v4mmol/l was set at a blood lactate concentration of 4 mmol/l for each athlete, vIAS was computed by adding a lactate concentration of 1.5 mmol/l to the lactate threshold for each athlete individually according to Wasserman, Hansen, Sue, Casaburi, and Whipp (1999) and Roecker, Striegel, and Dickhuth (2003).

Linear sprint test.

All players completed three valid sprint trials over 30 m with a minimum of 3 min recovery between. The starting position was a split start with the starting distance to the first timing light being 30 cm (Altmann et al., 2015). A start signal was not given. For a trial to be valid, players had to adopt the correct starting position and accelerate without prior rocking movements. Furthermore, a trial was considered invalid when players decelerated before reaching the finish timing light. Prior to the test, all athletes performed a standardized warm-up. Single-beam timing lights (TAG Heuer, La-Chaux-de-Fonds, Switzerland) were mounted at a height of 64 cm (Altmann et al., 2017) and were placed at the start and finish line in order to collect sprint times. The tests took place indoors on a polyvinyl-chloride-running surface. The results of all three attempts were averaged and used for the evaluation (LS30m) (Al Haddad, Simpson, & Buchheit, 2015).

Repeated-sprint test.

Following the linear sprint test, participants rested for 5 min before they completed a repeated-sprint test (RST). Using the same measurement set-up as in the linear sprint test, the RST consisted of five repeated maximal 30-m sprints, starting every 20 s. After each 30-m sprint, athletes had to jog back to the starting line in time. The last 10 s prior to the next sprint were counted down to provide the players with feedback regarding the beginning of the next sprint. Before the start of each sprint, participants remained in the ready position and waited for the start signal. The subjects were required to repeat each trial of the RST with maximum effort. The average time of the fourth and fifth sprint (RST30m) as well as the absolute performance decrement (RSTdecr) were used for analysis. The average of the fourth and fifth sprint rather than all five sprints was used in order to capture the sprints where fatigue is expected to be at its highest level. RSTdecr was calculated by subtracting LS30m from RST30m (Dawson, 2012).

Vertical jump test.

For the vertical jump test, all participants performed a countermovement jump (CMJ). The players were instructed to place their feet hip-width apart and to jump for maximal vertical height with their hands fixed at the hips. The athletes performed three valid maximum jumps with 1 min rest between the attempts. CMJ evaluation was determined using a force plate (AMTI, Watertown, MA, USA) and commercially available software (Templo, Contemplas GmbH, München, Germany), which estimated jump height by the vertical impulse. The best value was used for the final statistical analysis (Al Haddad et al., 2015).

Match performance parameters

Match-related physical performance data were collected for each match within a 10-week period following the performance tests using a multicamera, computerized tracking system (TRACAB, Chyron Hego, Melville, NY, USA). The system consists of two multicamera units, mounted side-by-side 10 m apart along the pitch long-side. Each unit contained three remote-controlled cameras sampling at 25 Hz. TRACAB uses image processing to track players (Chyron Hego; 2018; document not publicly accessible). The raw data obtained were further processed by OPTA Sports (London, UK), which is the official match data provider of the German Bundesliga. The reliability of the capture process and data accuracy for some of the parameters (e.g., aerial duels won) has been proven by Liu, Hopkins, Gómez, and Molinuevo (2013). For the present study, top running speed (maximal speed obtained during the whole match; TS), total distance covered (TD), as well as mean running speed (total distance divided by the number of minutes played; vØ) were determined. In addition, number of high-intensity runs (≥4 m/s for at least 2 s while reaching 5.0 m/s; HIR), number of sprints (≥4 m/s for at least 2 s while reaching 6.3 m/s; SP), and percentage of aerial duels won (AD+) were calculated and used for further statistical analyses. All definitions are based on the catalogue of the German soccer league (Deutsche Fußball Liga – DFL, 2014; document not publicly accessible).

To be included in analysis, players had to complete at least two games within the relevant 10-week period, where upon all games completed were pooled together for each player. Only games where athletes played at least 85 min were used for analysis.

Statistical analysis

Descriptive data are presented as mean values and standard deviations (SD). Statistical tests were performed using SPSS statistical software version 23 (SPSS Inc., Chicago, IL, USA). Assumptions of normality were verified using the Kolmogorov–Smirnov test. T‑tests were run to determine if differences existed between the three seasons in terms of performance test parameters and match-related physical performance variables. Pearson’s product–moment correlations were used to examine the relationships between performance test parameters and match-related physical performance variables. Based on previous studies (e.g., Bradley et al., 2011b; Castagna et al., 2010; Mendez-Villanueva, Buchheit, Simpson, Peltola, & Bourdon, 2011; Rampinini et al., 2007a) and theoretical considerations, the following parameters were correlated:
  • Aerobic endurance ability: vmax, v4mmol/l, and vIAS were correlated with TD, vØ, HIR, and SP,

  • Sprint ability: LS30m was correlated with TS, HIR, and SP,

  • Repeated-sprint ability: RST30m and RSTdecr were correlated with TS, TD, vØ, HIR, and SP,

  • Jump ability: CMJ was correlated with TS and AD+.

According to Hopkins (2002), the magnitude of the correlation coefficient was considered as small (0.1 ≤ r < 0.3), moderate (0.3 ≤ r < 0.5), large (0.5 ≤ r < 0.7), very large (0.7 ≤ r < 0.9), and nearly perfect (r ≥ 0.9).

The level of statistical significance for all tests was set to p < 0.05.

Results

Descriptive statistics (mean ± SD) of the performance test and match-related physical performance parameters for each season are reported in Table 1. The results showed no differences in the respective parameters between the three seasons (0.07 < p < 0.99). Therefore, comparisons between the three seasons seem justified. The average number of matches per player included in analysis was 7.9 ± 2.2 in season 2013/14, 6.4 ± 2.4 in season 2014/15, and 7.5 ± 1.9 in season 2015/16.
Table 1

Mean values (±SD) of the performance test and match-related physical performance parameters in each season

Parameter

2013/2014

2014/2015

2015/2016

Incremental treadmill test

Vmax [km/h]

17.6 ± 1.0

17.7 ± 1.0

17.7 ± 0.8

v4mmol/l [km/h]

15.0 ± 1.0

15.1 ± 0.9

15.2 ± 1.1

vIAS [km/h]

13.5 ± 0.9

13.6 ± 0.7

13.8 ± 1.1

Linear sprint test

LS30m [s]

4.12 ± 0.11

4.11 ± 0.13

4.15 ± 0.11

Repeated-sprint test

RST30m [s]

4.58 ± 0.16

4.52 ± 0.14

4.54 ± 0.10

RSTdecr [s]

0.48 ± 0.09

0.42 ± 0.09

0.39 ± 0.07

Vertical jump test

CMJ [cm]

40 ± 4

38 ± 4

36 ± 4

Match-related parameters

TS [km/h]

30.6 ± 1.0

30.0 ± 1.4

30.6 ± 1.3

TD [km]

10.9 ± 1.0

10.8 ± 0.8

10.9 ± 0.9

vØ [km/h]

7.1 ± 0.5

6.9 ± 0.5

7.1 ± 0.6

HIR

51.2 ± 15.6

55.1 ± 17.3

54.6 ± 17.5

SP

15.6 ± 7.0

16.9 ± 8.6

16.7 ± 7.0

AD+ [%]

60.7 ± 11.4

53.0 ± 29.5

49.9 ± 20.1

Performance test parameters: v max  maximal velocity reached in incremental treadmill test, v 4mmol/l  fixed anaerobic threshold in incremental treadmill test, v IAS  individual anaerobic threshold in incremental treadmill test, LS 30m  30-m time in linear sprint test; RST 30m  30-m time in repeated-sprint test, RST decr  absolute performance decrement in repeated-sprint test, CMJ countermovement jump height

Match-related physical performance parameters: TS top running speed, TD total distance covered, v Ø  mean running speed, HIR number of high-intensity running, SP number of sprints, AD+ aerial duels won

The range of players included in the correlation analysis was 8 to 10 in season 2013/14, 9 to 11 in season 2014/15, and 9 to 10 in season 2015/16, depending on the specific parameters investigated.

Correlations between performance test parameters and physical match performance variables are presented in Tables 234 and 5 separately for each season. Most of the investigated parameters showed a large correlation (r ≥ 0.5) for at least one of the three evaluated seasons. However, large to very large relationships (r ≥ 0.5 and r ≥ 0.7, respectively) across all three seasons were only observed between the following: vmax and TD; vmax and vØ; LS30m and TS as well as RST30m and TS.
Table 2

Pearson correlations (r) between incremental treadmill test and match-related physical performance parameters for each season

Parameters

2013/2014

2014/2015

2015/2016

Vmax and TD

r = 0.51

r = 0.74*

r = 0.83*

p = 0.20

p = 0.01

p < 0.01

n = 8

n = 11

n = 10

Vmax and vØ

r = 0.73*

r = 0.69*

r = 0.85*

p = 0.04

p = 0.02

p < 0.01

n = 8

n = 11

n = 10

Vmax and HIR

r = 0.26

r = 0.54

r = 0.59

p = 0.54

p = 0.09

p = 0.07

n = 8

n = 11

n = 10

Vmax and SP

r = 0.05

r = 0.47

r = 0.24

p = 0.91

p = 0.15

p = 0.50

n = 8

n = 11

n = 10

v4mmol/l and TD

r = 0.65

r = 0.37

r = 0.75*

p = 0.06

p = 0.26

p = 0.01

n = 10

n = 11

n = 10

v4mmol/l and vØ

r = 0.82*

r = 0.37

r = 0.75*

p = 0.01

p = 0.26

p = 0.01

n = 10

n = 11

n = 10

v4mmol/l and HIR

r = 0.32

r = 0.16

r = 0.74*

p = 0.41

p = 0.65

p = 0.01

n = 10

n = 11

n = 10

v4mmol/l and SP

r = −0.02

r = 0.01

r = 0.47

p = 0.95

p = 0.99

p = 0.17

n = 10

n = 11

n = 10

vIAS and TD

r = 0.52

r = 0.44

r = 0.72*

p = 0.12

p = 0.18

p = 0.02

n = 10

n = 11

n = 10

vIAS and vØ

r = 0.43

r = 0.46

r = 0.73*

p = 0.21

p = 0.16

p = 0.02

n = 10

n = 11

n = 10

vIAS and HIR

r = 0.17

r = 0.21

r = 0.75*

p = 0.64

p = 0.53

p = 0.01

n = 10

n = 11

n = 10

vIAS and SP

r = −0.08

r = 0.05

r = 0.46

p = 0.83

p = 0.87

p = 0.18

n = 10

n = 11

n = 10

Performance test parameters: v max  maximal velocity reached in incremental treadmill test, v 4mmol/l  fixed anaerobic threshold in incremental treadmill test, v IAS  individual anaerobic threshold in incremental treadmill test

Match-related physical performance parameters: TD total distance covered, v Ø  mean running speed, HIR number of high-intensity running, SP number of sprints

*Level of significance: p < 0.05; significant correlations are indicated by an asterisk

Table 3

Pearson correlations (r) between linear sprint test and match-related physical performance parameters for each season

Parameters

2013/2014

2014/2015

2015/2016

LS30m and TS

r = −0.79*

r = −0.74*

r = −0.86*

p < 0.01

p = 0.01

p < 0.01

n = 10

n = 10

n = 9

LS30m and HIR

r = −0.55

r = −0.24

r = 0.30

p = 0.13

p = 0.53

p = 0.43

n = 10

n = 10

n = 9

LS30m and SP

r = −0.34

r = −0.33

r = −0.24

p = 0.37

p = 0.39

p = 0.54

n = 10

n = 10

n = 9

Performance test parameter: LS 30m  30-m time in linear sprint test

Match-related physical performance parameters: TS top running speed, HIR number of high-intensity running, SP number of sprints

*Level of significance: p < 0.05; significant correlations are indicated by an asterisk

Table 4

Pearson correlations (r) between repeated-sprint test and match-related physical performance parameters for each season

Parameters

2013/2014

2014/2015

2015/2016

RST30m and TS

r = −0.71*

r = −0.57

r = −0.81*

p = 0.01

p = 0.06

p < 0.01

n = 9

n = 9

n = 9

RST30m and TD

r = −0.58

r = −0.21

r = 0.40

p = 0.10

p = 0.59

p = 0.28

n = 9

n = 9

n = 9

RST30m and vØ

r = −0.62

r = −0.22

r = 0.35

p = 0.08

p = 0.57

p = 0.36

n = 9

n = 9

n = 9

RST30m and HIR

r = −0.58

r = −0.46

r = −0.24

p = 0.10

p = 0.21

p = 0.54

n = 9

n = 9

n = 9

RST30m and SP

r = −0.21

r = −0.56

r = −0.65

p = 0.59

p = 0.12

p = 0.06

n = 9

n = 9

n = 9

RSTdecr and TS

r = −0.36

r = 0.15

r = 0.19

p = 0.22

p = 0.64

p = 0.61

n = 9

n = 9

n = 9

RSTdecr and TD

r = −0.65

r = −0.48

r = −0.71*

p = 0.06

p = 0.19

p = 0.03

n = 9

n = 9

n = 9

RSTdecr and vØ

r = −0.53

r = −0.39

r = −0.73*

p = 0.14

p = 0.31

p = 0.03

n = 9

n = 9

n = 9

RSTdecr and HIR

r = −0.30

r = −0.38

r = −0.82*

p = 0.44

p = 0.32

p = 0.01

n = 9

n = 9

n = 9

RSTdecr and SP

r = 0.14

r = −0.39

r = −0.54

p = 0.72

p = 0.30

p = 0.14

n = 9

n = 9

n = 9

Performance test parameters: RST 30m  30-m time in repeated-sprint test, RST decr  absolute performance decrement in repeated-sprint test

Match-related physical performance parameters: TS top running speed, TD total distance covered, v Ø  mean running speed, HIR number of high-intensity running, SP number of sprints

*Level of significance: p < 0.05; significant correlations are indicated by an asterisk

Table 5

Pearson correlations (r) between vertical jump test and match-related physical performance parameters for each season

Parameters

2013/2014

2014/2015

2015/2016

CMJ and TS

r = 0.63*

r = 0.41

r = 0.66*

p = 0.02

p = 0.17

p = 0.03

n = 9

n = 10

n = 10

CMJ and AD+

r = −0.74*

r = −0.43

r = 0.37

p = 0.02

p = 0.22

p = 0.29

n = 9

n = 10

n = 10

Performance test parameter: CMJ countermovement jump height

Match-related physical performance parameters: TS top running speed, AD+ aerial duels won

*Level of significance: p < 0.05; significant correlations are indicated by an asterisk

Large correlations (r ≥ 0.5) across two seasons were also found between v4mmol/l and vIAS on the one hand and TD on the other hand. The same was further obtained between vmax and HIR, RST30m and SP, RSTdecr and both TD and vØ, as well as CMJ and TS.

All remaining performance test and match-related physical performance parameters showed large relationships (r ≥ 0.5) in only one or in none of the three seasons.

Descriptive and correlation analyses were also run for all seasons as a whole. The respective results can be found in Supplementary Tables 1 and 2.

Discussion

The purpose of this study was to examine the relationship between performance test and physical match performance parameters in professional soccer players across three consecutive seasons. The main findings were:
  • The majority of performance test parameters showed large correlations to one or more selected match parameters in at least one of the three evaluated seasons.

  • Between vmax and TD, vmax and vØ, LS30m and TS as well as between RST30m and TS large to very large correlations were found across all three seasons.

  • Nonetheless, variable correlations across the three seasons were observed, depending on the parameters investigated.

Key parameters of aerobic endurance, as indicated by the outcomes of the incremental treadmill test (vmax, v4mmol/l, vIAS) were moderately to very largely related to TD and vØ during matches. This finding supports previous research examining the relationship between aerobic endurance and match running data in male (Bradley et al., 2011b; Buchheit et al., 2010; Castagna et al., 2009; Castagna et al., 2010; Fernandes-da-Silva, Castagna, Teixeira, Carminatti, & Guglielmo, 2016; Krustrup et al., 2003; Rampinini et al., 2007a) and female soccer players (Bradley et al., 2014). In the present study, vmax was the most stable aerobic endurance parameter in terms of relationships to TD and vØ compared to v4mmol/l and vIAS. This might be explained by vmax representing a more thorough indicator of endurance, targeting not only aerobic (below the anaerobic threshold), but to some extent also anaerobic (above the anaerobic threshold) components of endurance capacity.

In addition, vmax was largely correlated with HIR over two seasons. This result supports current evidence showing that peak speed reached during incremental field tests is moderately to very largely associated with the ability to cover greater distances at high or very high running speeds during matches (Castagna et al., 2009; Castagna et al., 2010; Fernandes-da-Silva et al., 2016; Rampinini et al., 2007a; Rebelo et al., 2014). Based on the findings of Balsom, Ekblom, and Sjödin (1994), Rampinini et al. (2007a) suggested that a high aerobic endurance capacity contributes to rapid recovery for athletes during high-intensity intermittent exercise.

However, the aerobic endurance performance parameters assessed in the present study were only small to moderately related to SP. This is in contrast to previous studies demonstrating large to very large relationships between incremental field tests and total sprint distance during matches (Castagna et al., 2010; Fernandes-da-Silva et al., 2016; Krustrup et al., 2003). A possible explanation might be that the abovementioned studies used total sprint distance rather than the number of sprints for analysis, and that different testing protocols (e.g., shuttle runs) were used. Taken together, the results of the present study suggest that none of the aerobic endurance parameters assessed during the incremental treadmill test (vmax, v4mmol/l, and vIAS) can be used as indicators of SP.

Very large correlations were found between LS30m and TS during matches across all three seasons. Although Rampinini et al. (2007a) did not find any relationship between the best trial during a RST and TS, more recent studies (Massard, Eggers, & Lovell, 2017; Mendez-Villanueva et al., 2011) reported similar results to those in the present investigation. Specifically, Mendez-Villanueva et al. (2011) stated that faster professional youth players reached higher absolute peak running speeds in games compared to their slower counterparts regardless of the playing position, with large to very large effect sizes.

Similar to LS30m, large to very large correlations over the three seasons were evident between RST30m and TS. The magnitude of correlations are higher than reported by Buchheit et al. (2010), who found moderate correlations between these parameters in young soccer players. These findings suggest that both LS30m and RST30m are valid and reliable indicators of TS during match play. In addition, RST30m and SP showed large correlations over two seasons. Accordingly, improving RST30m performance may contribute to maintaining sprinting performance throughout a soccer match (Rampinini et al., 2007a). Furthermore, RSTdecr was largely to very largely correlated to both TD and vØ across two seasons, while there were only small to moderate relationships to TS and small to large relationships to SP. RSTdecr solely takes into account to what extent the performance of a player decreases over the course of the test, regardless of absolute sprint times (Dawson, 2012). This might explain the relationships to more endurance-related match parameters such as TD and vØ. Ultimately, it seems reasonable to draw conclusions about speed and endurance-related match performance parameters of players based on RST, depending on the specific RST parameter.

Interestingly, although vmax and RSTdecr have been shown to not correlate with each other (Pyne, Saunders, Montgomery, Hewitt, & Sheehan, 2008), in this study both tests were associated with TD. Hence, vmax and RSTdecr seem to contribute to TD in different ways. Therefore, determining both parameters during performance testing seems justified.

CMJ was largely related to TS in two seasons, contradicting the results of Rampinini et al. (2007a). However, the relationship in the present investigation seems plausible as vertical jump height has been reported as a strong indicator of sprinting performance in elite male soccer players (Wisløff, Castagna, Helgerud, Jones, & Hoff, 2004). In contrast, CMJ showed no consistent and partially contradictory correlation to AD+, even when controlling for body height (data not shown). This indicates that the ability to win a high percentage of aerial duels not only depends on jump height but also on several additional factors, such as timing, anticipation, and body placement. Beyond that, there was also no relationship between CMJ and the number of aerial duels (data not shown).

A novelty, and the main strength of this study, is that the players were tracked over three consecutive seasons. The team played in the same division and was trained by the same coach using a similar tactical system during the three seasons, while many players systematically performed in the same positional role. Therefore, there was continuity over the time of the investigation. This is important as it reduced the potential confounding effects of playing style and tactical organization of teams that inherently influence match variability (Bush, Archer, Hogg, & Bradley, 2015). Concomitantly, variability was further reduced by including the average of up to 10 matches relating to the match performance outcomes, whereas in most other studies on this topic only 1 to 4 games were considered (e.g., Aquino et al., 2017; Castagna et al., 2010; Fernandes-da-Silva et al., 2016). In addition, based on performance tests including aerobic endurance, sprint, and power-related assessments, a wider picture of associations between physical capacity and match-related physical performance parameters can be provided through this study. These circumstances enabled us to investigate to what extent results persist over time. On the one hand, some results were stable across all seasons. Among other relationships, this was evident between LS30m and TS (very large correlations) or LS30m and SP (small to moderate correlations). On the other hand, variable relationships were evident for most of the parameters. Correlations were considered as variable across the three seasons if they varied between at least three classifications of correlation coefficients according to Hopkins (2002), e.g., from small to large. Among others, high variations in relationships were detected with regards to HIR. There were only small to moderate relationships to v4mmol/l, vIAS, and RSTdecr in the first two seasons. In contrast, very large correlations were shown in the third season. An even greater variability was evident regarding the relationship between CMJ and AD+ (varying between moderate positive and very large negative relationships). Hence, the present results suggest that the correlations are variable across the three seasons, depending on the parameters investigated. A possible explanation is that some match-related physical performance parameters are more resistant to the multitude of influencing factors in soccer than others. Moreover, some performance tests may reflect the demands of the match consistently, while others do not.

Besides considering each season separately, the seasons can also be analyzed as a whole. In this context, supplementary Table 2 presents the correlations between performance test and match-related physical performance parameters for all seasons taken together. Here, magnitudes of correlations in general reflect the mean values of correlations when considering each season separately. However, no conclusions can be drawn about the variability across the three seasons on this basis.

In this context, it is important to acknowledge that additional factors such as team tactics (Bradley, Carling et al., 2011a; Carling, 2011), quality of opponents (Rampinini, Coutts, Castagna, Sassi, & Impellizzeri, 2007b), ball possession (Bradley et al., 2013; Carling, 2010; Di Salvo, Gregson, Atkinson, Tordoff, & Drust, 2009), match status (Redwood-Brown, O’Donoghue, Robinson, & Neilson, 2017), and environmental factors (Mohr, Nybo, Grantham, & Racinais, 2012; Trewin et al., 2017) may have a strong influence on the variability in match-to-match performance. Consequently, it is difficult to evaluate the effectiveness of a training intervention program or the contribution of several underlying capabilities on the basis of match-related parameters (Mackenzie & Cushion, 2013). This issue highlights the importance of performance testing in order to monitor training interventions.

There are also some limitations of the present study that should be taken into account when interpreting the findings. A major issue in comparing the findings of this study with the results of previous studies is that a different battery of physical tests was used to measure physical capacity. The same applies for the match-related physical performance parameters, as previous studies used different match analysis systems and definitions of the parameters used.

Another potential limitation of the present study is the variability of sample sizes (from 8 to 11 players), depending on the parameters and seasons evaluated, which might have influenced results. As new players joined the team and other players were transferred to other clubs, team composition changed over the course of the three seasons investigated. However, transferring players is common practice in soccer and therefore cannot be avoided. A further limitation is that not all common key parameters of aerobic endurance were assessed, specifically VO2max and running economy (Hoppe et al., 2013). In the same context, the introduction of vIAS as a more individual parameter of the anaerobic threshold is a novelty, as this parameter has not been accounted for in comparable studies (e.g., Krustrup et al., 2003).

Further investigations with larger sample sizes are needed to confirm our findings. Larger sample sizes would also enable position-dependent analysis. In addition, it would be interesting if there is a relationship between potential changes in players’ physical capacities and a change in physical performance in actual games (Buchheit, Simpson, & Mendez-Villanueva, 2013). Thus, minimum requirements for physical capacity in soccer could be identified on both a team and an individual level.

Conclusions

This study demonstrates the relationship between several performance test parameters and match-related physical performance in professional soccer players, highlighting the presence or absence of the test parameters’ criterion validity.

The present study thereby extends the current knowledge on this topic by revealing parameter-specific variability of correlations over the course of three seasons, suggesting that vmax, LS30m, and RST30m seem to be the most specific and consistent parameters associated with TD and vØ (vmax), and TS (LS30m and RST30m), respectively. Nonetheless, because of the multifactorial and complex nature of soccer performance itself (Svensson & Drust, 2005), the use of field tests and laboratory assessments to predict on-field match performance should be applied with caution. However, if a coach makes specific demands on a player in a specific position, selected performance parameters may serve as appropriate indicators of the associated physical match performance. As further practical applications, these parameters can be used for monitoring the training status of the players or the efficacy of training interventions.

Notes

Acknowledgements

The authors would like to thank OPTA Sports for providing the match data used in this study.

Compliance with ethical guidelines

Conflict of interest

S. Altmann, M. Kuberczyk, S. Ringhof. Neumann and A. Woll declare that they have no competing interests.

The study was approved by the institutional review board and was conducted in accordance with the Declaration of Helsinki. All subjects gave their written informed consent prior to participation.

Supplementary material

12662_2018_519_MOESM1_ESM.docx (34 kb)
Supplementary Table 1. Mean values (±SD) of the performance test and match-related physical performance parameters for all seasons. Supplementary Table 2. Pearson correlations (r) between performance test and match-related physical performance parameters for all seasons.

References

  1. Altmann, S., Hoffmann, M., Kurz, G., Neumann, R., Woll, A., & Haertel, S. (2015). Different starting distances affect 5‑m sprint times. Journal of strength and conditioning research, 29(8), 2361–2366.  https://doi.org/10.1519/JSC.0000000000000865.CrossRefPubMedGoogle Scholar
  2. Altmann, S., Spielmann, M., Engel, F. A., Neumann, R., Ringhof, S., Oriwol, D., & Haertel, S. (2017). Validity of single-beam timing lights at different heights. Journal of strength and conditioning research, 31(7), 1994–1999.  https://doi.org/10.1519/JSC.0000000000001889.CrossRefPubMedGoogle Scholar
  3. Aquino, R., Palucci Vieira, L. H., de Paula Oliveira, L., Cruz Goncalves, L. G., & Pereira Santiago, P. R. (2017). Relationship between field tests and match running performance in high-level young brazilian soccer players. The journal of sports medicine and physical fitness.  https://doi.org/10.23736/S0022-4707.17.06651-8.Google Scholar
  4. Balsom, P. D., Ekblom, B., & Sjödin, B. (1994). Enhanced oxygen availability during high intensity intermittent exercise decreases anaerobic metabolite concentrations in blood. Acta physiologica Scandinavica, 150(4), 455–456.  https://doi.org/10.1111/j.1748-1716.1994.tb09711.x.CrossRefPubMedGoogle Scholar
  5. Bradley, P. S., Sheldon, W., Wooster, B., Olsen, P., Boanas, P., & Krustrup, P. (2009). High-intensity running in English FA Premier League soccer matches. Journal of sports sciences, 27(2), 159–168.  https://doi.org/10.1080/02640410802512775.CrossRefPubMedGoogle Scholar
  6. Bradley, P. S., Carling, C., Archer, D., Roberts, J., Dodds, A., Di Mascio, M., Krustrup, P., et al. (2011a). The effect of playing formation on high-intensity running and technical profiles in English FA Premier League soccer matches. Journal of sports sciences, 29(8), 821–830.  https://doi.org/10.1080/02640414.2011.561868.CrossRefPubMedGoogle Scholar
  7. Bradley, P. S., Mohr, M., Bendiksen, M., Randers, M. B., Flindt, M., Barnes, C., Krustrup, P., et al. (2011b). Sub-maximal and maximal Yo-Yo intermittent endurance test level 2: heart rate response, reproducibility and application to elite soccer. European journal of applied physiology, 111(6), 969–978.  https://doi.org/10.1007/s00421-010-1721-2.CrossRefPubMedGoogle Scholar
  8. Bradley, P. S., Carling, C., Gomez Diaz, A., Hood, P., Barnes, C., Ade, J., Mohr, M., et al. (2013). Match performance and physical capacity of players in the top three competitive standards of English professional soccer. Human movement science, 32(4), 808–821.  https://doi.org/10.1016/j.humov.2013.06.002.CrossRefPubMedGoogle Scholar
  9. Bradley, P. S., Bendiksen, M., Dellal, A., Mohr, M., Wilkie, A., Datson, N., Krustrup, P., et al. (2014). The application of the Yo-Yo intermittent endurance level 2 test to elite female soccer populations. Scandinavian journal of medicine & science in sports, 24(1), 43–54.  https://doi.org/10.1111/j.1600-0838.2012.01483.x.CrossRefGoogle Scholar
  10. Buchheit, M., Mendez-Villanueva, A., Simpson, B. M., & Bourdon, P. C. (2010). Match running performance and fitness in youth soccer. International journal of sports medicine, 31(11), 818–825.  https://doi.org/10.1055/s-0030-1262838.CrossRefPubMedGoogle Scholar
  11. Buchheit, M., Simpson, B. M., & Mendez-Villanueva, A. (2013). Repeated high-speed activities during youth soccer games in relation to changes in maximal sprinting and aerobic speeds. International journal of sports medicine, 34(1), 40–48.  https://doi.org/10.1055/s-0032-1316363.PubMedGoogle Scholar
  12. Bush, M. D., Archer, D. T., Hogg, R., & Bradley, P. S. (2015). Factors influencing physical and technical variability in the English Premier League. International journal of sports physiology and performance, 10(7), 865–872.  https://doi.org/10.1123/ijspp.2014-0484.CrossRefPubMedGoogle Scholar
  13. Carling, C. (2010). Analysis of physical activity profiles when running with the ball in a professional soccer team. Journal of sports sciences, 28(3), 319–326.  https://doi.org/10.1080/02640410903473851.CrossRefPubMedGoogle Scholar
  14. Carling, C. (2011). Influence of opposition team formation on physical and skill-related performance in a professional soccer team. European journal of sport science, 11(3), 155–164.  https://doi.org/10.1080/17461391.2010.499972.CrossRefGoogle Scholar
  15. Castagna, C., Impellizzeri, F., Cecchini, E., Rampinini, E., & Alvarez, J. C. B. (2009). Effects of intermittent-endurance fitness on match performance in young male soccer players. Journal of strength and conditioning research, 23(7), 1954–1959.  https://doi.org/10.1519/JSC.0b013e3181b7f743.CrossRefPubMedGoogle Scholar
  16. Castagna, C., Manzi, V., Impellizzeri, F., Weston, M., & Barbero Alvarez, J. C. (2010). Relationship between endurance field tests and match performance in young soccer players. Journal of strength and conditioning research, 24(12), 3227–3233.  https://doi.org/10.1519/JSC.0b013e3181e72709.CrossRefPubMedGoogle Scholar
  17. Chamari, K., Hachana, Y., Ahmed, Y. B., Galy, O., Sghaïer, F., Chatard, J.-C., & Wisløff, U. (2004). Field and laboratory testing in young elite soccer players. British journal of sports medicine, 38(2), 191–196.CrossRefPubMedPubMedCentralGoogle Scholar
  18. Chyron Hego (2018). TRACAB technology description: DFL Google Scholar
  19. Currell, K., & Jeukendrup, A. E. (2008). Validity, reliability and sensitivity of measures of sporting performance. Sports medicine, 38(4), 297–316.  https://doi.org/10.2165/00007256-200838040-00003.CrossRefPubMedGoogle Scholar
  20. Dawson, B. (2012). Repeated-sprint ability: Where are we? International journal of sports physiology and performance, 7(3), 285–289.CrossRefPubMedGoogle Scholar
  21. DFL (2014). Definitionskatalog Offizielle Spieldaten version 3.0: [Definition catalogue official match data version 3.0] Google Scholar
  22. Faude, O., Schlumberg, A., Fritsche, T., Treff, G., & Meyer, T. (2010). Leistungsdiagnostische Testverfahren im Fußball – methodische Standards [Performance diagnosis in football – methodological standards]. Deutsche Zeitschrift für Sportmedizin, 61(6), 129–133.Google Scholar
  23. Fernandes-da-Silva, J., Castagna, C., Teixeira, A. S., Carminatti, L. J., & Guglielmo, L. G. A. (2016). The peak velocity derived from the Carminatti Test is related to physical match performance in young soccer players. Journal of sports sciences, 34(24), 2238–2245.  https://doi.org/10.1080/02640414.2016.1209307.CrossRefPubMedGoogle Scholar
  24. Al Haddad, H., Simpson, B. M., & Buchheit, M. (2015). Monitoring changes in jump and sprint performance: best or average values? International journal of sports physiology and performance, 10(7), 931–934.  https://doi.org/10.1123/ijspp.2014-0540.CrossRefPubMedGoogle Scholar
  25. Halson, S. L. (2014). Monitoring training load to understand fatigue in athletes. Sports medicine, 44(Suppl 2), S139–S147.  https://doi.org/10.1007/s40279-014-0253-z.CrossRefPubMedGoogle Scholar
  26. Hopkins, W. G. (2002). A scale of magnitudes for effect statistics. A new view of statistics, 502 Google Scholar
  27. Hopkins, W. G., Schabort, E. J., & Hawley, J. A. (2001). Reliability of power in physical performance tests. Sports medicine, 31(3), 211–234.CrossRefPubMedGoogle Scholar
  28. Hoppe, M. W., Baumgart, C., Sperlich, B., Ibrahim, H., Jansen, C., Willis, S. J., & Freiwald, J. (2013). Comparison between three different endurance tests in professional soccer players. Journal of strength and conditioning research, 27(1), 31–37.  https://doi.org/10.1519/JSC.0b013e31824e1711.CrossRefPubMedGoogle Scholar
  29. Krustrup, P., Mohr, M., Amstrup, T., Rysgaard, T., Johansen, J., Steensberg, A., Bangsbo, J., et al. (2003). The yo-yo intermittent recovery test: physiological response, reliability, and validity. Medicine and science in sports and exercise, 35(4), 697–705.  https://doi.org/10.1249/01.MSS.0000058441.94520.32.CrossRefPubMedGoogle Scholar
  30. Liu, H., Hopkins, W., Gómez, A. M., & Molinuevo, S. J. (2013). Inter-operator reliability of live football match statistics from OPTA sportsdata. International Journal of Performance Analysis in Sport, 13(3), 803–821.  https://doi.org/10.1080/24748668.2013.11868690.CrossRefGoogle Scholar
  31. Mackenzie, R., & Cushion, C. (2013). Performance analysis in football: a critical review and implications for future research. Journal of sports sciences, 31(6), 639–676.  https://doi.org/10.1080/02640414.2012.746720.CrossRefPubMedGoogle Scholar
  32. Massard, T., Eggers, T., & Lovell, R. (2017). Peak speed determination in football: is sprint testing necessary? Science and medicine in football, 10, 1–4.  https://doi.org/10.1080/24733938.2017.1398409.CrossRefGoogle Scholar
  33. Mendez-Villanueva, A., & Buchheit, M. (2011). Physical capacity-match physical performance relationships in soccer: simply, more complex. European journal of applied physiology, 111(9), 2387–2389.  https://doi.org/10.1007/s00421-011-1868-5.CrossRefPubMedGoogle Scholar
  34. Mendez-Villanueva, A., Buchheit, M., Simpson, B., Peltola, E., & Bourdon, P. (2011). Does on-field sprinting performance in young soccer players depend on how fast they can run or how fast they do run? Journal of strength and conditioning research, 25(9), 2634–2638.  https://doi.org/10.1519/JSC.0b013e318201c281.CrossRefPubMedGoogle Scholar
  35. Mohr, M., Krustrup, P., & Bangsbo, J. (2003). Match performance of high-standard soccer players with special reference to development of fatigue. Journal of sports sciences, 21(7), 519–528.  https://doi.org/10.1080/0264041031000071182.CrossRefPubMedGoogle Scholar
  36. Mohr, M., Nybo, L., Grantham, J., & Racinais, S. (2012). Physiological responses and physical performance during football in the heat. PLoS ONE, 7(6), e39202.  https://doi.org/10.1371/journal.pone.0039202.CrossRefPubMedPubMedCentralGoogle Scholar
  37. Oliver, J. L. (2009). Is a fatigue index a worthwhile measure of repeated sprint ability? Journal of science and medicine in sport, 12(1), 20–23.  https://doi.org/10.1016/j.jsams.2007.10.010.CrossRefPubMedGoogle Scholar
  38. Pekas, D., Trajković, N., & Krističević, T. (2016). Relation between fitness tests and match performance in junior soccer players. Sport science, 9(suppl 2), 88–92.Google Scholar
  39. Pyne, D. B., Saunders, P. U., Montgomery, P. G., Hewitt, A. J., & Sheehan, K. (2008). Relationships between repeated sprint testing, speed, and endurance. Journal of strength and conditioning research, 22(5), 1633–1637.  https://doi.org/10.1519/JSC.0b013e318181fe7a.CrossRefPubMedGoogle Scholar
  40. Rampinini, E., Bishop, D., Marcora, S. M., Ferrari Bravo, D., Sassi, R., & Impellizzeri, F. M. (2007a). Validity of simple field tests as indicators of match-related physical performance in top-level professional soccer players. International journal of sports medicine, 28(3), 228–235.  https://doi.org/10.1055/s-2006-924340.CrossRefPubMedGoogle Scholar
  41. Rampinini, E., Coutts, A. J., Castagna, C., Sassi, R., & Impellizzeri, F. M. (2007b). Variation in top level soccer match performance. International journal of sports medicine, 28(12), 1018–1024.  https://doi.org/10.1055/s-2007-965158.CrossRefPubMedGoogle Scholar
  42. Rebelo, A., Brito, J., Seabra, A., Oliveira, J., & Krustrup, P. (2014). Physical match performance of youth football players in relation to physical capacity. European journal of sport science, 14(suppl 1), S148–S156.  https://doi.org/10.1080/17461391.2012.664171.CrossRefPubMedGoogle Scholar
  43. Redwood-Brown, A., O’Donoghue, P., Robinson, G., & Neilson, P. (2017). The effect of score-line on work-rate in English FA Premier League soccer. International Journal of Performance Analysis in Sport, 12(2), 258–271.  https://doi.org/10.1080/24748668.2012.11868598.CrossRefGoogle Scholar
  44. Roecker, K., Striegel, H., & Dickhuth, H.-H. (2003). Heart-rate recommendations: transfer between running and cycling exercise? International journal of sports medicine, 24(3), 173–178.  https://doi.org/10.1055/s-2003-39087.CrossRefPubMedGoogle Scholar
  45. Di Salvo, V., Gregson, W., Atkinson, G., Tordoff, P., & Drust, B. (2009). Analysis of high intensity activity in Premier League soccer. International journal of sports medicine, 30(3), 205–212.  https://doi.org/10.1055/s-0028-1105950.CrossRefPubMedGoogle Scholar
  46. Stølen, T., Chamari, K., Castagna, C., & Wisløff, U. (2005). Physiology of soccer: an update. Sports medicine, 35(6), 501–536.CrossRefPubMedGoogle Scholar
  47. Svensson, M., & Drust, B. (2005). Testing soccer players. Journal of sports sciences, 23(6), 601–618.  https://doi.org/10.1080/02640410400021294.CrossRefPubMedGoogle Scholar
  48. Trewin, J., Meylan, C., Varley, M. C., & Cronin, J. (2017). The influence of situational and environmental factors on match-running in soccer: a systematic review. Science and medicine in football, 1(2), 183–194.  https://doi.org/10.1080/24733938.2017.1329589.CrossRefGoogle Scholar
  49. Turner, A., Walker, S., Stembridge, M., Coneyworth, P., Reed, G., Birdsey, L., & Moody, J. (2011). A testing battery for the assessment of fitness in soccer players. Strength and conditioning journal, 33(5), 29–39.  https://doi.org/10.1519/SSC.0b013e31822fc80a.CrossRefGoogle Scholar
  50. Wasserman, K., Hansen, J. E., Sue, D. Y., Casaburi, R., & Whipp, B. J. (1999). Principles of exercise testing and interpretation. Baltimore: Lippincott Williams & Wilkins.Google Scholar
  51. Wisløff, U., Castagna, C., Helgerud, J., Jones, R., & Hoff, J. (2004). Strong correlation of maximal squat strength with sprint performance and vertical jump height in elite soccer players. British journal of sports medicine, 38(3), 285–288.CrossRefPubMedPubMedCentralGoogle Scholar

Copyright information

© Springer-Verlag GmbH Deutschland, ein Teil von Springer Nature 2018

Authors and Affiliations

  1. 1.Department for Performance Analysis, Institute of Sports and Sports ScienceKarlsruhe Institute of TechnologyKarlsruheGermany
  2. 2.Karlsruher Sport-ClubKarlsruheGermany
  3. 3.BioMotion Center, Institute of Sports and Sports ScienceKarlsruhe Institute of TechnologyKarlsruheGermany
  4. 4.Department for Social Sciences, Institute of Sports and Sports ScienceKarlsruhe Institute of TechnologyKarlsruheGermany

Personalised recommendations