Quantification of Training Load Relative to Match Load of Youth National Team Soccer Players

Abstract

Background:

Previous studies have examined the training load relative to match load in club settings. The aims of this study were to (1) quantify the external training load relative to match load in days before a subsequent international game and (2) examine the cumulative training load in relation to match load of U-17 national team field soccer players.

Hypothesis:

Volume and intensity load parameters will vary between trainings; the farthermost trainings have the highest load gradually decreasing toward the match.

Study Design:

Prospective cohort study.

Level of Evidence:

Level 4.

Methods:

External training load data were collected from 84 youth national team players using global positioning technology between 2016 and 2020. In the national team setting, training load data were obtained from 3 days before the actual match day (MD-3, MD-2, MD-1 days) and analyzed with regard to the number of days up to the game. Volume and intensity parameters were calculated as a percentage of the subsequent match load.

Results:

Significant differences were found between MD-1 and MD-2, as well as between MD-1 and MD-3 for most volume parameters (P < 0.01; effect sizes [ESs] 0.68-0.99) and high-intensity distance (P < 0.002; ES 0.67 and 0.73) and maximum velocity (P < 0.002; ES 0.82) as intensity parameters. Most cumulative values were significantly different from total duration (P < 0.001, common language ES 0.80-0.96).

Conclusion:

The training volume gradually decreased as match day approached, with the highest volume occurring on MD-3. Intensity variables, such as maximum velocity, high-intensity accelerations, and meterage per minute were larger in MD-1 training relative to match load. Training volume was lowest in MD-1 trainings and highest in MD-3 trainings; intensity however varies between training days.

Clinical Relevance:

The findings of this study may help to understand the special preparational demands of international matches, highlighting the role of decreased training volume and increased intensity.

In recent years, training and match load monitoring has become popular in modern soccer science. Objective measures of internal and external load data allow coaches and practitioners to plan the training and recovery strategies employed during periods of congested schedules. ⁹ Effective monitoring techniques can also aid in reducing injury risk and optimizing performance. Consequently, in the club environment, training and match load monitoring is commonly applied in the training periodization process. ¹²

Global positioning system (GPS) technology allows strength and conditioning practitioners to precisely monitor the locomotor activity of elite players.^{4,8,11,19,20,33,36} Match activity profiles, characterized by distance covered in high-intensity running and sprinting, number of accelerations and decelerations, and changes of directions have markedly increased in recent years. ⁶ Elite players are occasionally required to play highly demanding games every 3 days, making recovery between matches extremely important. This is particularly evident for national teams when players are required to compete in friendly or official games every 48 to 72 hours. National team duty also comprises a higher injury risk for players required to compete in additional tournaments in an already busy playing schedule. During a 5-season period (2009-2014), over one-third of national team squad members sustained a time-loss injury due to match-play. ¹³

Several studies^{18,21,29,34,37} have documented the external load of elite soccer players during matches. Weekly training load has also been the focus of research, and contemporary studies^2,3,15,17 have highlighted the physical load of soccer players during in-season training days and its effects on long-term match performance. Most studies^23,24,30,31 examined the absolute load of players represented by traditional GPS parameter, such as total distance, high-intensity distance, and sprint distance. However, this method does not take into account the individual- and position-specific training load in relation to the actual match demands. To improve the specificity of load placed on players, the relative training load has been used recently, which compares the training load with the match load of elite-level players and found very small differences in tapering strategies toward the game.^5,16,28,35 Most published studies demonstrate progressively decreased training load from MD-4 to MD-1 (4 days before match—MD-4, 3 days before match—MD-3, 2 days before match—MD-2, and 1 day before match—MD-1) in preparation for the match, where only MD-1 differed significantly from other training days in several internal and external load variables. ²⁶ Although, absolute training load values are useful in planning the microcycle, differences exist between nations and leagues because of their special approach to soccer training. Individually quantified training load relative to match load was previously used as a possible approach to prepare players for the physical demands of the game.^5,16,28,35

Although the match and training load in club teams has been extensively investigated, very few researchers have examined the training load of players in national team settings. Of the 2 studies that have examined the weekly periodization of training load in youth athletes, both have been limited to club-level players.^17,39 Therefore, the first aim of this study was to quantify the training load of youth national team soccer players in relation to match load. We hypothesized that training volume would be lower in training days closer to match day, whereas training intensity would increase on MD-3 and MD-1 days, when conditioning and activation can take place. The second aim of our study was to provide insight into the cumulative training load of youth national team players. Physical demands of the game changes rapidly, which might be presented in the preparation process of the national teams before international matches. Volume and even more intensity of the training sessions before a match might represent the style and philosophy of the head coach of the team, leading to differences between teams in that regard. To the best of our knowledge, no other studies have investigated, however, the training load relative to match load in youth national team players, where high demands are placed on players in a very short amount of time because of the congested schedule and international opponents. We hypothesized that training load relative to match load on different training days before training days will vary substantially and volume will decrease toward the actual match.

Methods

Design

A total of 64 matches of 4 successive U-17 national teams (players born in 2000: 20 matches; 2001: 13; 2002: 20; 2003: 11) were recorded between 2016 and 2019. The 4 U-17 teams were coached by 3 different head coaches, with 1 coach working with 2 teams. GPS technology was used to monitor absolute and relative external match day load and training load of the 3 days (MD-1, MD-2, MD-3) leading up to the international match. Periodization in soccer is commonly used so, where MD indicates match day, MD-1 (match day–1) is the last training day before the subsequent match and MD-3 is the first day in the actual microcycle in a national team setting. Training durations showed large variances for MD-1, MD-2, and MD-3 trainings (55.92 ± 15.87, 71.99 ± 31.84, and 75.62 ± 27.34 minutes, respectively), with minor differences between different age-groups. On 17 occasions on MD-3 were previous matches, the remaining days were trainings.

National team training camps were organized in 3 different ways. Most were 5 days long and consisted of 2 international matches (Figure 1A), others were 7 to 10 days long and included 3 international friendly matches with 1 day in between (Figure 1B). European Qualifications and European Championship camps involved 3 to 5 international competitive matches with 2 recovery days between matches (Figure 1C). Altogether 179 different training days were recorded (2000, 55 days; 2001, 32 days; 2002, 60 days; 2003, 32 days). Only outfield players with full match participation were included in the study. Until 2018, the UEFA official game time for the U-17 age-group was 80 minutes. This was changed to 90 minutes from 2018, making the absolute load data incomparable across the different seasons. For this reason, the actual match values were used as a reference for each training microcycle before the game.

Figure 1.

Typical schedule of the U-17 youth national teams. (A) Most of the training camps consist of 4 training sessions and 2 international matches. (B) International preparatory tournaments are generally longer and include 3 international matches. (C) European qualification tournaments and European Championships are organized in a way that teams play international matches every 72 hours.

Subjects

Eighty-four elite U-17 outfield soccer players born between 2000 and 2003 (age 16.8 ± 0.3 years; mass 70.9 ± 6.2 kg; height 179.7 ± 5.5 cm) participated in this study. All the players were members of the U-17 National Team at the time and had at least 5 years of soccer experience. The study was approved by the ethical review board of the University of Physical Education (Budapest) and conformed to the Code of Ethics of the World Medical Association (Declaration of Helsinki). Players and their parents/guardians provided written consent for the use of training and match data.

Procedures

National team training settings are much different from club environment as players do not have as many opportunities to play full games. Therefore, in our study, each of the U-17 international game load values were used as a reference for the training load in the preceding 3 days. The physical activity of the players during the data collection period were monitored during each friendly and official match and training session using a portable 10 Hz GPS unit with a 100 Hz accelerometer (Catapult S5 between June 2016 and June 2019 and Catapult S7 between June 2019 and March 2020, Catapult Sports). Each unit was worn in custom-made vests between the shoulder blades, allowing unrestricted movements of the upper limbs and torso. Each player used the same device during the study period. ¹⁴ The mean velocity difference of the S5 (https://football-technology.fifa.com/media/172169/catapultgps-s5-epts-report-nov2018.pdf) and S7 (https://football-technology.fifa.com/media/172128/oct-2019-catapult-vectorgps-fifa-epts-report.pdf) units were validated against the gold standard Vicon system for several velocity zones during multiple soccer-specific drills. External load variables such as total duration (minutes), total distance (m), meterage per minute (m/min), total player load (arbitrary units, AU), ³⁸ high-intensity running distance (m > 19.8 km/h), sprint distance (m > 25.2 km/h), maximum sprint speed distance (m > 30 km/h), explosive distance (m > 2 m/s²), and number of high-intensity accelerations (number > 3 m/s²) and decelerations (number < −3 m/s²) were used in this study. The aforementioned speed^1,32 and intensity ²² thresholds were established based on previous studies. Besides the commonly used GPS variables, inertial movement analysis (IMA) data were also used. ²⁷ The total number of IMA events comprised the sum of accelerations, decelerations, and changes of directions. ²⁵ Intensity parameters were calculated by normalizing the load parameters by the duration of the match or training session. The mean value of each training session was expressed relative to the mean external load registered during the match for players, who fully participated in both the reference match and all the training sessions (MD-1, MD-2, MD-3) before the game itself. The relative load on training days has been calculated as a percentage of the players’ match load in each subsequent game. Following this analogy, as an example, a 30% training load on MD-1 in total running distance would mean 3600 m, if in the upcoming match the player completed 12,000 m. For days when teams had 2 training (2 occasions) sessions, the volume parameters were regarded as a sum and intensity parameters as an average value. During the statistical analysis, we have compared the mean values for each training day of those players, who fully participated in the subsequent international match.

Statistical Analyses

Statistical analyses were conducted using SPSS for Windows 16.0 (IBM Corp). Data were tested for normality using a Shapiro-Wilk test. When data were normally distributed, dependent t tests were used to quantify the possible differences between the equal-sized samples. Otherwise, the nonparametric Wilcoxon test was performed. Samples were considered as related, given that data were collected from multiple days of the same players. Two types of effect sizes (ESs) were used to quantify the magnitude of the test results: Cohen d value for the dependent t-test results and the common language effect size (CLES) for the results of the Wilcoxon and the dependentt test. Thus, with utilizing the CLES values, the power of the results between the 2 tests became comparable. Cohen d values were classified as trivial (<0.2), small (>0.2-0.6), moderate (>0.61-1.2), large (>1.21-2.0), and very large (>2.0). ⁷ Only moderate-large ESs were considered for further analysis. Descriptive values are presented as means ± standard deviations. After application of the Bonferroni correction, the significance level was set at P < 0.002 for the volume and intensity variables, and P < 0.001 for the cumulative values.

Results

Relative Volume Training Load Variables

Figure 2 presents the volume parameters for the relative external load obtained from MD-1, MD-2, and MD-3 trainings before international matches. Statistical analyses revealed significant differences with moderate ESs between MD-1 and MD-2, as well as MD-1 and MD-3 training days in total volume of high-intensity distance, total player load, explosive distance, total distance, total duration, sprint distance, total number of IMA events, and number of high-intensity decelerations (P < 0.01; ES 0.68-0.99).

Figure 2.

Training volume variables from training days before matches expressed as a percentage of match values. The black dotted line represents the full match load as 100% of each variable. Different coloured bars indicate the match day (MD) and training days (MD-1, MD-2, MD-3), as shown in the notation. Total duration was recorded as minutes, total player load as AU (arbitrary units), meterage per minute as meter/minute, total distance, high-intensity distance (>19.8 km/h), sprint distance (>25.1 km/h), distance >30 km/h, and explosive distance (>2.0 m/s²) as meter and number of IMA events, number of accelerations (>3.0 m/s²), and number of decelerations (>−3.0 m/s²) as count. IMA, inertial movement analysis.

Relative Intensity Training Load Variables

Figure 3 presents the intensity parameter values for the relative external load obtained from MD-1, MD-2, and MD-3 trainings before international matches. Regarding the intensity load of the training sessions, we found a similar declining pattern as in the case of the volume load variables, however, to a lesser extent. Moderate but significant differences were observed between MD-1 and MD-2/MD-3 days for the high-intensity distance (P < 0.002; ES 0.67 and 0.73). Similar reductions were seen in maximum velocity between MD-1 and MD-2 (P < 0.002; ES 0.82).

Figure 3.

Training intensity variables from training days before matches expressed as a percentage of match values. The black dotted line represents the full match load as 100% of each variable. Different coloured bars indicate the match day (MD) and training days (MD-1, MD-2, MD-3), as shown in the notation. Maximum velocity was recorded as highest velocity reached during the session in km/h, meterage per minute as m/min, total player load as arbitrary units per minute (AU/min), total distance, high-intensity distance (>19.8 km/h), sprint distance (>25.1 km/h), distance >30 km/h, and explosive distance (>2.0 m/s²) as m/min and number of IMA events, number of accelerations (>3.0 m/s²), and number of decelerations (>−3.0 m/s²) as count/min. IMA, inertial movement analysis.

Relative Cumulative Training Load

Figure 4 shows the 3-day cumulative training load before an international game for the selected variables. Cumulative load of total duration (2.10 ± 0.25 matches) and number of high-intensity accelerations (2.84 ± 0.77 matches) were above the 2 match values. Conversely, high-intensity running distance was below 1 match value (0.95 ± 0.03 matches). When summarized in the 3 days before the match, all other analyzed variables were between 1 and 2 matches load. Differences were found in cumulative data between the total duration and other parameters, such as total distance, explosive distance, sprint distance, decelerations, and high-intensity distance (P < 0.001, CLES 0.80-0.96). Similar differences were revealed between total player load and explosive distance, decelerations (P < 0.001, CLES 0.82-0.87), and even more interestingly between accelerations and decelerations (P < 0.001, CLES 0.83).

Figure 4.

Cumulative average external training load of the different age-groups expressed as a ratio of training load values per match load values of the 3 training days before an international match. The black-colored bars demonstrate the match values, meanwhile the other coloured bars represent the cumulative volume of the 3 training days for each variable. Total duration was recorded as minutes, total player load as AU (arbitrary units), total distance, high-intensity distance (>19.8 km/h), sprint distance (>25.1 km/h), distance >30 km/h, and explosive distance (>2.0 m/s²) as meters and number of IMA events, number of accelerations (>3.0 m/s²), and number of decelerations (>−3.0 m/s²) as count. IMA, inertial movement analysis.

Discussion

The aims of this study were to compare the external training load of U-17 national team soccer players in relation to international matches and to provide insight into the cumulative training load of the prematch training days. In conjunction with previous studies,^28,35 we have found that in general, training load decreased toward match day. As expected, the average external training load was highest on MD-3, as the farthest training day from the upcoming game. This could be because that (1) in international qualification tournaments, the matches are organized every 72 hours, which means that MD-3 is a previous match and (2) because this day of training provides the last opportunity for coaches to apply high training load in tactical situations before the game. This is in contrast with other studies conducted in club settings, where the highest training load was experienced on MD-4. ²⁸ However, in the national team environment, MD-4 is not included in the strictly considered microcycle. Although mostly 1-week microcycles are used to plan the periodization programming of club teams, in national team camps and tournaments, players are expected to play games every 48 or 72 hours. The result is that most of the MD-3 training days are either matches or compensation trainings (if the players have not participated in the required amount of playing minutes to satisfy the physiological demands). Accordingly, microcylces are 4 days long, including MD, MD-1, MD-2, and MD-3, as opposed to the widely accepted 1-week cycle employed in the club soccer environment.

In standard national team settings, the MD-1 day is similar to club trainings before a game and includes physical activation, tactical refinements, and standard situation practices. ²⁸ In most studies where weekly training load has been examined, researchers have found that the training load was lowest on the day before the match in absolute^2,3,26 and relative values. ³⁵ In the national team setting, MD-2 days can vary; they are either low-impact recovery days in preparation for a qualifier tournament to allow regeneration from the previous match or can also include another match in the case of a special preparational training camp (Figure 1). From a physiological and physical perspective, the MD-3 days provide the only opportunity for loading. Occasionally, these days might include 2 training sessions, but in official tournaments, MD-3 often represents another game day. For this reason, MD-3 days have a large variance with respect to training load. Although we would have liked to compare MD-3 days when matches were played or double sessions were performed, a limited dataset prevented this further analysis.

To date, most practitioners have used an average or maximum of volume values from a known number of club matches to create a match reference, to which training load values can be compared.^28,35 For this, players need to have a large number of match values of full or nearly full participation in matches, since partial participation makes assessment inaccurate, both in terms of volume and intensity of match load, partly because of the pacing strategies adopted by the players relative to their field time. ¹⁰ National team players do not have the opportunity to play as many matches as they do in their clubs. In the present study, U-17 national team players fully participated between 1 and 14 matches (4.04 ± 3.28) in their entire international season. Consistent with previous studies,^15,35 to precisely quantify the training load in relation to match load, we used each game value as a reference for the preceding training sessions. Although retrospective in nature, this approach provides more recent and accurate results as a reference for further training load investigation, in contrast to collecting data for a longitudinal period, which is then used for a full season.

In our study, we represented the cumulative load of the 3 training days for the volume variables in the ratio of match load. Because matches are played in every 72 hours in national team settings, microcycles are well suited for 3 days. However, to allow comparison with previous studies, the match load was included in the microcycle cumulative load. All included variables, except high-intensity running distance reached on average the 1 match value in the 3 days of training. We have found that for certain age-groups, the cumulative load of accelerations in relation to match load is much higher than shown in previous studies,^15,35 although the number of decelerations barely reached the 1 match value. This also shows a discrepancy in the loading of mechanical variables in the sample teams. All other recorded variables remained between the 1 and 2 match values. Despite having mainly regenerative and activation training sessions between matches, accumulated values can be higher because of the fact that MD-3 and occasionally even MD-2 are previous matches, which in the case of this study will be included in the cumulative load of the chosen variable.

Although most of the studies^15,28,35 on the subject have used volume variables, when assessing the weekly training load relative to match load, the intensity of the drills and sessions is equally important, especially, when such a short amount of time is available for game preparation, such is the case between international matches. A similar approach was applied, when peak 5-minute periods were used to represent training intensity in relation to match load. ⁵ In our study, when normalized for training time, training intensity showed similar patterns to training volume, except for meterage per minute and number of high-intensity accelerations, which were even above match intensity on the MD-1 and MD-3 days. The higher intensity in the 3 days before the match can be explained by the load of the previous match, or the high load prescribed by the head coach. On the other hand, in MD-1, high-intensity drills were used in short periods to activate players for the match on the next day. We found a large difference in the intensity of mechanical load variables, such as accelerations and decelerations, corresponding with the volume load results. However, the mechanical intensity, represented by the number of accelerations and decelerations, was similar or even higher than match values in the days leading into the locomotive intensity, represented by distance covered or sprint and high-intensity distance, which is relatively low compared with match values. This loading difference must be further investigated and optimized to perfectly prepare players to meet international game demands.

Practical Applications

The results presented in our study highlight the importance of monitoring training load in the periodization process to adequately prepare players for an international game. Providing the optimal loading stimulus to players in such a short period as that experienced in international tournaments or 2 matches per week scenarios is crucial to reach peak performance and minimize the potential risk of injury. The congested schedule of the national team setting provides a narrow window of opportunity to monitor training load before matches. Therefore, gaining adequate quantity and quality of training data are paramount; practitioners can use this data to periodize the players loading and recovering strategies. Cumulative load is an important part of training monitoring, as it can provide insight into continuous volume load within the microcycle. MD-3 provides an opportunity for a larger training volume, when high speed and sprint running distance might be high. However, MD-2 is mainly used for recovery and recuperation. MD-1 ought to be used for activation and tactical refinement by keeping the volume low and intensity similarly high as the match values. We also suggest practitioners to collect training intensity data besides the commonly used volume parameters, as it plays a key role in the pre-match trainings. As part of a tapering strategy, during the training days before international matches, volume will normally be kept low relative to match reference values, while intensity might be higher, preparing the players for the high demands of the game against international opponents.