Original Research

Dose-Response Relationship Between Internal

Training Load and Changes in Performance During
the Preseason in Youth Soccer Players
Diogo H. Figueiredo,1 Diego H. Figueiredo,1 Alexandre Moreira,2 Helcio R. Gonçalves,3 and
Antonio C. Dourado3
1
Department of Physical Education, State University of Maringa, Maringa, Brazil; 2Department of Sport, School of Physical Education
and Sport, University of Sao Paulo, Sao Paulo, Brazil; and 3Department of Physical Education, State University of Londrina, Londrina,
Brazil

Abstract
Figueiredo, DH, Figueiredo, DH, Moreira, A, Gonçalves, HR, and Dourado, AC. Dose-response relationship between internal
training load and changes in performance during the preseason in youth soccer players. J Strength Cond Res XX(X): 000–000,
2019—The aim of this study was to describe training intensity distribution based on the session rating of perceived exertion (sRPE)
and heart rate (HR) methods and examine the dose-response relation between internal training load (ITL) and change in perfor-
mance of 16 youth soccer players (mean 6 SD age: 18.75 6 0.68 years, height: 175.3 6 5.5 cm, body mass: 68.7 6 6.5 kg, and
body fat: 10.7 6 1.2%) belonging to a Brazilian first division team during a 3-week preseason. The sRPE and HR data were
registered daily to calculate the ITL and the training intensity distribution, in 3 intensity zones (low, moderate, and high). The Yo-yo
Intermittent Recovery Level 1 (Yo-yo IR1) was evaluated before and after experimental period. The total time spent in the low-
intensity zone (HR method) was greater (p , 0.01) compared with the moderate- and high-intensity zones. No difference was
observed between training intensity zones determined by the sRPE method (p . 0.05). Negative correlations were observed
between weekly mean sRPE-TL (r 5 20.69), Edward’s-TL (r 5 20.50), and change in Yo-yo IR1. Linear regression indicated that
weekly mean sRPE-TL (F1;14 5 13.3; p , 0.01) and Edward’s-TL (F1;14 5 4.8; p , 0.05) predicted 48.7 and 25.5% of the variance in
performance change, respectively. Stepwise linear regression revealed that these 2-predictor variables (F2;13 5 18.9; p , 0.001)
explained 74.5% of the variance in performance change. The results suggest that the sRPE and HR methods cannot be used
interchangeably to determine training intensity distribution. Moreover, sRPE-TL seems to be more effective than the HR-based TL
method to predict changes in performance in youth soccer players.
Key Words: session rating of perceived exertion, heart rate, team sports

Introduction Despite the widespread use of sRPE and HR-based methods to

monitor ITL in soccer (1,10,24), a key issue of debate regards how
The team-sport training process during the preseason aims to
the training intensity and daily training process should be orga-
develop and maximize some positive physiological adaptations in
nized to induce positive adaptations; moreover, it is imperative to
the athletes to optimize physical, technical, and tactical per-
advance the understanding on how these adaptations could in-
formances (9). The likelihood of achieving this aim may be in-
fluence the achievement of higher levels of sports performance
creased by a systematic and appropriate training design
(12,13). Thus, some researchers have examined the distribution
concerned with the balance between training load (TL) and re-
of training intensity in soccer using the sRPE and HR methods
covery (29,30,37). Indeed, this balance could aid in reducing the
(3,12,13,28,40). For instance, based on the sRPE method, Algrøy
risks of negative outcomes such as overuse injury, upper re-
et al. (3) reported that the daily training intensity distribution of
spiratory tract infections, signs and symptoms of nonfunctional
a professional soccer team was similar during preseason and in-
overreaching, or even overtraining syndrome (8,15,20,21). These
season training, with almost the same number of training sessions
training-induced adaptations are believed to be associated with
(TSs) performed in the 3 intensity zones (low-, moderate-, and
the internal TL (ITL), which consists of the actual physiological
high-intensity zones). Furthermore, Wrigley et al. (40) reported
stress imposed on the athletes’ organisms (39). It has been pro-
a different training intensity distribution for youth elite soccer
posed that monitoring ITL is imperative to assess athlete
players (U18). Using the HR method, the authors demonstrated
responses to a given training dose (26). Among a variety of ITL
that a higher proportion of the TSs were performed in the low-
monitoring methods, the session rating of perceived exertion
intensity zone (,70% of HRmax), during a 2-week training
(sRPE) and heart rate (HR)-based methods have been largely
period.
adopted for monitoring purposes in soccer (16,27).
A habitual characteristic of the soccer training process consists
of TSs aimed at improving player technical-tactical competence
Address correspondence to Diogo H. Figueiredo, diogohfigueiredo@hotmail.com. and physical performance (1,29,36). During these sessions, it is
Journal of Strength and Conditioning Research 00(00)/1–8 expected that high variability in training responses will be ob-
ª 2019 National Strength and Conditioning Association served between players who are undertaking a similar training

Copyright © 2019 National Strength and Conditioning Association. Unauthorized reproduction of this article is prohibited.
 Internal Training Load and Performance in Soccer Players (2019) 00:00
stimulus (2,27). This inherent condition emerges as a challenge
for monitoring the training process and for assessing the dose-
response relationship in team sports settings. Despite this diffi-
culty, an increasing number of studies have focused on analysis of
the dose-response relationship, in particular, taking into account
the association between the ITL (HR-based TL and sRPE-TL) and
changes in performance in professional (11–13,36) and young
soccer players (1,24,34,35). For instance, Los Arcos et al. (34)
showed negative correlations between both respiratory sRPE-TL
(sRPEres-TL) and muscular sRPE-TL (sRPEmus-TL) and changes
in dominant leg countermovement jump (CMJD) (r 5 20.23 and
r 5 20.54, respectively) and nondominant leg CMJ (CMJnD)
(r 5 20.30 and r 5 20.52, respectively) during the preseason
(9-week period) in Spanish second division young soccer players
(20.2 6 1.9 years).
Nevertheless, to the best of our knowledge, there is still a lack
of studies that analyzed the dose-response relationship between
both ITL (Edward’s-TL and sRPE-TL) methods and changes in
performance during an entire preseason in elite youth soccer
players (34). Therefore, the aim of this study was to describe the
training intensity distribution of youth soccer players, based on
the sRPE and HR methods, during an entire preseason phase, and
to examine the dose-response relationships between ITL measures
and changes in Yo-yo performance. It was hypothesized that
a similar training intensity distribution pattern to that reported in
a professional soccer team would be observed. It was also hy-
pothesized that the ITL would modulate the changes in Yo-yo
performance.
Methods
Experimental Approach to the Problem
This was a single-group observational study that was conducted
during an entire preseason (3-week period) for a Brazilian first
division state championship (U19 State Championship). During
the preseason, the sRPE-TL and Edward’s-TL methods were
monitored daily for each TS and friendly match (FM). Training
intensity distribution was quantified based on the HR and sRPE
methods using 3 predefined intensity zones: low (,75% HRmax
and sRPE ,4), moderate (75–85% HRmax and sRPE 4–7), and
high (.85% HRmax and sRPE .7) according to Lovell et al. (32)
for each TS and FM. To analyze the dose-response relationships,
data from the Yo-yo IR1 performance were assessed. The players
completed the Yo-yo IR1 twice; 1 day before, and 2 days after the
end of the experimental period. Players were largely familiarized
with the Yo-yo IR1 because this test was habitually used by the
coaching staff to evaluate player physical performance. To avoid
any potential effects of a previous TS on the Yo-yo IR1 perfor-
mance test, no TSs were performed on the day preceding the Yo-
yo IR1 test (24 hours before the evaluation).
Subjects
Eighteen youth male soccer players belonging to a Brazilian first
division team competing in the junior first division state cham-
pionship volunteered to participate in this investigation. For
a subject’s data to be included in the final analysis, the following
criteria were adopted: (a) completion of 100% of the TS in each
week of the assessed preseason; (b) completion of sRPE to de-
termine the ITL; (c) presenting all HR data during the TS and FM;
(d) participating for at least 45 minutes in each of the 2 evaluated
FM; and (e) not presenting any injuries during the experimental
Copyright © 2019 National Strength and Conditioning Association. Unauthorized reproduction of this article is prohibited.
period. Therefore, because of injuries, data from only 16 players
(mean 6 SD age: 18.75 6 0.68 years, height: 175.3 6 5.5 cm,
body mass: 68.7 6 6.5 kg, and body fat: 10.7 6 1.2%) were
retained for analysis. Subjects were informed that they were free
to withdraw from the study at any time. Goalkeeper data were
excluded from this study. All athletes and their parents were
briefed about procedures and signed an informed consent form
before participation. The study was approved by the State Uni-
versity of Londrina Human Research Ethics Committee for Re-
search involving Human Subjects (no 1.556.245/2016).
Procedures
Training Schedule. During the experimental period, all evaluated
players performed 5 TSs during the 1st week, 3 TSs in the 2nd
week, and 2 TSs in the 3rd week, with 1 TS per day. Two FMs (90
minutes) were performed on Wednesdays, in weeks 2 and 3. The
typical training program performed by soccer players during the
investigated period included technical-tactical training (technical
drills, small-sided games, and modified soccer games) and phys-
ical conditioning sessions, such as sprint training, speed-
endurance training, strength and power training, core exercises,
and high-intensity aerobic training. One strength TS per week
was performed, consisting of 8 exercises of a full-body workout
with 3 sets of 10 repetitions at 30% of 1 repetition maximum
loads. The training program was planned by the coach of the
team, and the researchers did not alter the original training pro-
gram. During the TS and FM, players were allowed to consume
isotonic sports drinks ad libitum (during recovery periods). All
TSs were performed in the afternoon (between 3:00 and 5:30 PM).
The 2 FMs were also played in the afternoon (between 3:00 and 5:
00 PM). The FMs were played against opponents with a lower
competitive level, and there was no fixed number of substitutions.
The player match participation was planned to allow a similar
match volume (min of participation) for all assessed players.
Training Intensity Distribution. Training intensity distribution
was quantified from the HRmax using predefined training in-
tensity zones (low intensity ,75%; moderate intensity 75–85%;
and high intensity .85%) (32). The software provided by the HR
monitor (Suunto Team Pod; Suunto Oy, Vantaa, Finland) was
used to determine the percentage of total training time spent in
each of the 3 training zones for each individual TS and FM. The
total training time spent (%) in each zone for all TSs and FMs was
then determined. The sRPE data were divided into 3 intensity
zones: low intensity (#4 arbitrary units [AUs]), moderate in-
tensity (.4 to ,7 AU), and high intensity ($7 AU) based on
a previous study (32). The athletes provided the sRPE score
(0–10) using the category‐ratio scale (CR‐10) scale (5) for each TS
and FM (22). Subsequently, the percentage of total training and
FM time spent (%) in each sRPE zones was calculated by adding
the duration of each session.
Internal Training Load. To determine the ITL for each player, 2
methods were adopted: sRPE-TL and HR-based methods
(Edward’s-TL).
Session Rating of Perceived Exertion. The internal TL was com-
puted using the sRPE method in accordance with previous studies
(22,26,38). Approximately 30 minutes after completion of every
TS and FM, the players were required to report the intensity of the
entire TS and FM using a CR-10 sliding scale (5), where 1 5 not
 Internal Training Load and Performance in Soccer Players (2019) 00:00
hard at all and 10 5 very, very hard. To determine ITL, expressed
in AUs, the product of the TS and FM duration (in minute) and
sRPE score (CR-10) rated by the player was used (22). The du-
ration of the TS and FM was recorded from the start (warm-up) to
the end of the session, including recovery periods but excluding
stretching exercises. The FM duration also excluded the half-time
rest period. An index of training monotony was also obtained by
dividing the mean daily TL by the SD (21).
Edward’s Training Load. To evaluate ITL based on HR during
the TS and FM, the method proposed by Edwards (19) was
adopted. Internal TL was calculated for each TS and FM by
multiplying the accumulated duration in each HR zone (minute)
through a multiplier allocated to each zone (50–60% HRmax 5
1, .60–70% HRmax 5 2, .70–80% HRmax 5 3, .80–90%
HRmax 5 4, and .90–100% HRmax 5 5). The duration of the
TS and FM was recorded from the start (warm-up) to the end,
including recovery periods but excluding stretching exercises. The
FM duration also excluded the half-time rest period. Internal TL
was expressed in absolute values in AUs. To avoid missing data,
after each TS and FM, HR data from each player were immedi-
ately saved to a computer as an electronic spreadsheet (Excel;
Microsoft Corporation, USA), using Software from Suunto Team
Pod. The individual HRmax was considered as the highest value
reached for a player during the Yo-yo IR1 because the HRmax
reached during the TS or FM was lower than that found during
the Yo-yo IR1 test.
Performance Assessment: Yo-yo Intermittent Recovery Level 1
(Yo-yo IR1). The Yo-yo IR1 test consists of repeated 2 3 20-m
runs back and forth between the starting, turning, and finishing
lines at a progressively increasing speed controlled by audio
bleeps from a CD player (4). After each 2 3 20-m running bout,
the players were allowed a 10-second (2 3 5 m) active recovery
period. The test ended when the subjects failed twice to reach the
starting line (objective evaluation) or the subject felt unable to
complete another shuttle at the dictated speed (subjective evalu-
ation). Standard encouragements were provided to players at the
2-collection time-points by the same investigators. All athletes
were familiarized with the test and experimental procedures be-
fore the study. During the Yo-yo IR1, the HRmax of each player
was obtained using an HR monitor with a short-range radio te-
lemetry system (Suunto Team Pod; Suunto Oy). The test was
performed on a soccer field. The total distance covered during the
Yo-yo IR1 was considered as the test score. This method has been
shown to be valid and reliable in youth soccer players (18), with
an intraclass correlation and coefficient of variation for the Yo-yo
IR1 of 0.94 and 7.9%, respectively, for the total distance cov-
ered (17).
Statistical Analyses
Descriptive results are presented as mean 6 SD. The normality
assumption was verified using the Shapiro-Wilk test, and the
Levene’s test of homogeneity of variances was also applied. The
student’s paired t-test was used to determine any significant dif-
ference in Yo-yo IR1 performance from before training to after
training. One-way analysis of variance for repeated measures
followed by the Bonferroni post hoc test was used to evaluate
differences between weeks in the sRPE-TL, Edward’s-TL, and
ITL. The Friedman test and Dunn’s post hoc test were used to
identify possible differences between weekly training volume,
Copyright © 2019 National Strength and Conditioning Association. Unauthorized reproduction of this article is prohibited.
| www.nsca.com
monotony, and training intensity distribution determined by HR
and sRPE. To assess the magnitude of changes, the effect sizes
(ESs) were used according to Cohen (Cohen’s d) (14). The fol-
lowing criteria were adopted to interpret the magnitude of
change; ,0.2, trivial; 0.2–0.6, small; 0.6–1.2, moderate; 1.2–2.0,
large; and 2.0–4.0, very large (25). To examine the relationship
between weekly ITL measures and changes in the Yo-yo IR1
performance, Pearson product-moment correlations were per-
formed. Correlation coefficients (r) were interpreted using the
following qualitative descriptors: trivial (,0.1), small (,0.3),
moderate (0.3–0.5), large (0.5–0.7), very large (0.7–0.9), nearly
perfect (.0.9), and perfect (1.0) (25). Separate linear regression
models were fit to establish the relationship between each weekly
ITL measure (independent variables) and changes in Yo-yo IR1
performance (dependent variable). A multiple linear regression
was conducted to determine the relationship between the 2
weekly ITL measures (predictor variables: weekly mean sRPE,
weekly mean Edward’s-TL, and changes in Yo-yo IR1 perfor-
mance [dependent variable]) using a backwards stepwise re-
gression approach. The standardized difference (95% confidence
interval [CI]) in the sRPE-TL, Edward’s-TL, and ITL and stan-
dardized difference (99% CI) in each weekly ITL measure were
calculated. The level of significance was set at 5% (p # 0.05) for
all statistical analysis. The Statistical Package for the Social Sci-
ences (SPSS) (Version 20.0 for Windows; SPSS, Inc, Chicago, IL,
USA) was used to conduct the analysis.
Results
Table 1 shows the weekly average TS volume, sRPE-TL,
Edward’s-TL, and monotony during the 3 weeks. Higher ses-
sion volume in week 1 was observed compared with weeks 2 and
3 (p , 0.01), with no difference between weeks 2 and 3 (p .
0.05). A significant change in sRPE-TL was verified during the
training weeks (F2;30 5 270.2, p , 0.001). Week 1 showed higher
values compared with week 2 (95% CI [189.8–248.9]; p , 0.001)
and week 3 (95% CI [265.8–339.8]; p , 0.001), with a significant
decrease in week 3 compared with week 2 (95% CI [2124.7 to 2
42.2]; p , 0.001). There was a main effect for time for Edward’s-
TL (F1,4;21,2 5 37.0, p , 0.001), with higher values in week 1
compared with weeks 2 (95% CI [52.9–120.5]; p , 0.001) and 3
(95% CI [48.0–122.8]; p , 0.001), and no difference between
weeks 2 and 3 (95% CI [220.3 to 17.7]; p . 0.05). Regarding
monotony, no significant difference was observed between weeks
1 and 2 (p . 0.05); lower monotony was verified for week 3
compared with week 1 (p , 0.01).
Figure 1 presents the ITL during the 3-week preseason period.
There was a significant main effect for time (F2;30 5 682.2; p ,
0.001). The post hoc analysis showed that ITL was significantly
higher in week 1 compared with week 2 (95% CI
[1,255.6–1,531.6]; ES 5 26.4; p , 0.001) and week 3 (95% CI
[1,785.2–2,096.3]; ES 5 29.5; p , 0.001). A significant decrease
in ITL was observed during week 3 compared with week 2 (95%
CI [2690.9 to 2403.4]; ES 5 23.22; p , 0.001).
Figure 2 shows the training time (%) spent in the 3 intensity
zones (black bars—HR zones and white bars—sRPE zones) for all
training weeks. The total time spent in the HR low-intensity zone
(61.8%) was greater than in the moderate- (20.9%) and higher-
intensity zones (17.3%) during the 3-week training period (p ,
0.01 and p , 0.001, respectively). However, there were no sig-
nificant differences reported between the total time spent in the 3
intensity sRPE zones (p . 0.05).
 Internal Training Load and Performance in Soccer Players (2019) 00:00
Table 1
Weekly average training session volume, sRPE-TL, Edward’s-TL, and monotony during the investigation (mean and SD).*
Weeks
Measure Week 1 Week 2
Weekly average session 78.8 6 16.8 55.7 6 12.7†
volume (min)
sRPE-TL (AU) 516 6 49.0 297 6 46.0†
Edward’s-TL (AU) 217 6 53.0 131 6 23.0†
Monotony (AU) 3.6 6 0.7 3.2 6 0.9
*sRPE 5 session rating of perceived exertion; TL 5 training load; AU 5 arbitrary unit.
†Significantly different to week 1 (p , 0.001).
‡Significantly different to week 1 (p , 0.01).
§Significantly different to week 2 (p , 0.001).
Figure 3 display the results of the Yo-yo IR1 (m) before and
after the 3-week training period. A significant increase in the Yo-
yo IR1 performance was detected from before training to after
training (t 5 22.6; ES 5 0.83; p , 0.05).
Figure 4 shows the relationship between each independent
variable and the change in Yo-yo IR1 performance. Both weekly
mean sRPE-TL (r 5 20.69; 99% CI [20.85 to 20.08]) and
weekly mean Edward’s-TL (r 5 20.50; 99% CI [20.93 to 0.14])
showed a large and negative correlation with change in Yo-yo IR1
performance. Linear regression analysis indicated that weekly
mean sRPE-TL (F1;14 5 13.3; R2 5 0.487; p , 0.01) and weekly
mean Edward’s-TL (F1;14 5 4.8; R2 5 0.255; p , 0.05) signifi-
cantly predicted 48.7 and 25.5% of the variance in the change in
Yo-yo IR1 performance, respectively.
The stepwise linear regression analysis for the component factor
models for percentage change in the Yo-yo IR1 performance is
presented in Table 2. The single best predictor for change in Yo-yo
IR1 performance was the weekly mean sRPE-TL, explaining
48.7% of the performance. The final model included 2 predictors
that collectively explained 74.5% of the variance in change in the
Yo-yo IR1 performance (F2;13 5 18.9; R2 5 0.745; p , 0.001).
Weekly mean Edward’s-TL (b 5 20.508; t 5 23.624; p , 0.01)
and weekly mean sRPE-TL (b 5 20.700; t 5 24.995; p , 0.001)
are predictors of the change in the Yo-yo IR1 performance. The
relationship between these 2 variables and change in the Yo-yo IR1
performance can be represented by the equation: % change in Yo-
yo IR1 performance: (20.471 3 weekly mean sRPE-TL) 1 (2
0.397 3 weekly mean Edward’s-TL) 1 256.37, where weekly
mean sRPE-TL and Edward’s-TL are expressed in AUs.
Discussion
The purpose of this study was to describe the training intensity
distribution during the preseason, based on the sRPE and HR
methods, and to examine the dose-response relationships between
ITL (sRPE-TL and Edward’s-TL) and change in Yo-yo IR1 per-
formance in youth soccer players. The main findings of the study
were twofold: The total training time spent in the low-intensity
HR zone was greater than that in the moderate- and high-
intensity zones; only a few TSs were performed in moderate- and
high-intensity HR zones. However, there were no differences for
the total time spent in the different intensity zones when consid-
ering the intensity distribution based on sRPE. In addition, both
ITL measures (sRPE-TL and Edward’s-TL) significantly predicted
(48.7 and 25.5%, respectively) the variance in change in the Yo-
yo IR1 performance. The final model including these 2 predictors
explained 74.5% of the variance in Yo-yo IR1 performance
change.
Copyright © 2019 National Strength and Conditioning Association. Unauthorized reproduction of this article is prohibited.
Effect size (ES)
Week 3 Week 1 vs. Week 2 Week 1 vs. Week 3 Week 2 vs. Week 3
49.2 6 11.2† 21.55 22.07 20.54
213 6 51.0†§ 24.61 26.06 21.73
132 6 43.0† 22.11 21.76 0.03
2.5 6 2.2‡ 20.50 20.67 20.42
The present results show greater ITL and volume (minute)
performed during week 1 compared with weeks 2 and 3 (Table 1
and Figure 1). These findings were expected because of the higher
number of sessions performed in week 1 (5 vs. 4 and 3 in weeks 2
and 3, respectively); moreover, this higher ITL for week 1 con-
firms the sensitivity of the sRPE and HR-based methods to esti-
mate ITL in soccer, while providing new evidence for the
usefulness of adopting these methods to monitor ITL in youth
soccer players. It should be highlighted that the sensitivity of the
sRPE to monitor ITL in youth soccer players is in agreement with
reports by Freitas et al. (23). The authors conducted a study
designed with a 2-week overloading training phase followed by
a 2-week tapering phase during the preseason of elite youth
Brazilian soccer players (U16). The reported results demonstrated
that the ITL assessed by means of the sRPE method was aligned
with the planned overloading and tapering training weeks. In-
deed, Freitas et al. (23) reported a significant reduction in ITL for
weeks 3 and 4 (i.e., tapering phase) compared with weeks 1 and 2
(i.e., overloading phase), and a large ES for the magnitude of the
difference in ITL between week 1 and weeks 3 and 4 (2.82 and 2
3.87, respectively); and also between week 2 and weeks 3 (23.72)
and 4 (25.57). In conjunction, the data from this study and those
presented by Freitas et al. (23) reinforce the usefulness of adopting
the sRPE and HR-based methods for monitoring purposes in
youth elite soccer players.
Like most high-level team sports coaches, many soccer coaches
propose elevated volumes of high-intensity TLs in the preseason
to improve athletes’ performance and thus increase their chances
of success during the competitive period (31). This approach
might be, in part, associated with high pressure from the
Figure 1. Overall ITL assessed by sRPE across the 3-week
training period (mean and SD). *Significantly different from
week 1 (p , 0.001); †Significantly different from week 2 (p ,
0.001). AU 5 arbitrary unit; ITL 5 internal training load; sRPE
5 session rating of perceived exertion.
 Internal Training Load and Performance in Soccer Players (2019) 00:00 | www.nsca.com

Figure 2. Training time (%) spent in the 3 intensity zones (low, moderate, and higher) when
determined by heart rate (HR) and session rating of perceived exertion (sRPE) during (A) 5 week
1, (B) 5 week 2, (C) 5 week 3, and (D) 5 total of weeks. a Different from low–intensity zone based
on HRmax (p , 0.01). b Different from low‐intensity zone based on sRPE (p # 0.05). c Different
from moderate‐intensity zone based on sRPE (p # 0.05).

environment, including media attention and fans, to achieve pos- a 1-week training in preseason and in-season periods, the authors
itive results from the beginning of the season. However, to date, reported that 65 and 81% of the training intensity distribution
limited studies have investigated the training intensity distribution were completed in the low-intensity HR zone (,70% of HRmax),
in team sports, particularly in youth soccer teams (3,12,28,40), to 16 and 13% at moderate intensity (70–80% of HRmax), and 18
verify whether such beliefs may actually occur through empirical and 6% at high intensity (.80% HRmax), respectively. The
observations. The current results demonstrate that 31 and 62% of findings from Jeong et al. (28) suggest that across all preseason
the total training time spent during the preseason were performed sessions, the time distributions in HR zones between 80 and
in the low-intensity zone (sRPE #4 and ,75% of HRmax, re- 100% of HRmax were significantly higher compared with those
spectively), whereas 41 and 21% were performed at moderate in- observed during the in-season training period. Therefore, the
tensity (sRPE between 4.5 and 6.5 and 75–85% of HRmax, physiological demands of this training phase are greater than at
respectively), and 28 and 17% at higher intensities (sRPE $7 and other times in the annual plan. This greater physiological stress
.85% HRmax, respectively). This finding is in agreement with
reports in professional soccer players (3). Algrøy et al. (3) also
adopted the sRPE method to examine the training intensity distri-
bution, and the authors demonstrated that professional soccer
players completed 35 6 2% of TSs at lower intensity (RPE ,4 AU),
38 6 2% at moderate intensity (RPE 4.5 and 6.5 AU), and 27 6
4% at higher intensities (RPE .7 AU) during a 4-week preseason
period. Interestingly, they reported almost the same number of TSs
performed in the 3 intensity zones. Based on the data provided by
Algrøy et al. (3) and those from this study, a clear pattern of the
training intensity distribution seems to exist in elite soccer, for both
professional and youth elite soccer. However, a greater number of
studies on this issue with larger samples and conducted in different
countries could aid in advancing this knowledge while corrobo-
rating or refuting the present findings.
Concerning the analysis of the training intensity distribution
based on the HR method, it should be mentioned that the results Figure 3. Mean values 6 SD of the Yo-yo Intermittent Re-
of this study were similar to those reported by Jeong et al. (28), covery Test Level 1 (Yo-yo IR1) (m) of pre-training and post-
who conducted a study to quantify the physiological loads in training during the 3-week preseason. *Significant difference
compared with pre-training (p , 0.05).
Korean professional soccer players (24 6 3 years). During

Copyright © 2019 National Strength and Conditioning Association. Unauthorized reproduction of this article is prohibited.
 Internal Training Load and Performance in Soccer Players (2019) 00:00

Figure 4. Linear regression between (A) weekly mean sRPE-TL and (B) weekly mean Edward’s-
TL with the % change before training and after training achieved during the Yo-yo Intermittent
Recovery Test Level 1 (Yo-yo IR1) in meters for the 16 subjects. AU 5 arbitrary unit; sRPE 5
session rating of perceived exertion; TL 5 training load.

completed during preseason might be attributed to the need for
players to complete the high-intensity exercise sessions required
to achieve optimal fitness for the competitive season.
In line with the results of this study, Wrigley et al. (40) reported
that U14 soccer players completed 65, 30, and 9% of the total
time spent in the lowest (,70% of HRmax), moderate (71–80%
of HRmax), and highest (.81% of HRmax) intensity HR zones,
and U16 completed 64, 20, and 15% in the lowest, moderate, and
highest HR zones, respectively. The authors also reported that
U18 completed 67, 16, and 17% of the total time spent in the
lowest, moderate, and highest HR zones, respectively during
a 2-week training period (i.e., competitive season). These results
suggest that in the highest intensity zone, a main effect over these
chronological age groups was noted (F2,21 5 6.62; p 5 0.006)
with significantly higher values in the U18 group compared with
the other younger groups. This increased high-intensity training
activity in the U18 players might reflect a change in focus of the
training stimulus administered by the coaches and sports science
team along with the enhanced physical development which
accompanies normal growth and maturation (40).
Based on the results of the aforementioned studies, it could be
inferred that both professional and elite youth soccer players tend
to present a similar pattern of training intensity distribution. It is
possible that the soccer TSs in which coaches try to mimic the real
demands of the match-play might induce higher training intensity
during the session, thus elevating the proportion of the TSs per-
formed at moderate and high intensity, regardless of the level of the
players (professional vs. youth players). Speculatively, this result
could be explained by the higher amount of mixed TSs performed
by soccer players. Moreover, it could be assumed that as a conse-
quence of soccer-specific TSs (i.e., physical- and technical/tactical-
specific TSs), professionals and elite youth players perform rela-
tively hard workouts, especially in the preseason when the soccer
players are undertaking TSs with the aim of enabling them to fulfill
the physiological demands of the competitive season (28). It is
important to highlight that in this study, the exclusively low-
intensity sessions were mainly conducted in the form of recovery
exercise, notably during the day after the FM.
Owing to the present results concerning ITL and changes in
Yo-yo IR1, it is possible to infer about the dose-response re-
lationship during the training process. These findings add im-
portant and novel information to the literature regarding TL
manipulation and physical performance responses in elite youth
soccer players. The results reported in the current study suggest
that a high mean weekly sRPE-TL and mean weekly Edward’s-TL
associated with TS and FM (Figure 4) can impair several physical
fitness factors believed to be relevant for on-field soccer perfor-
mance, showing the large variability in individual response to TS
and FM. However, the results concerning the increase in physical
performance (Yo-yo IR1 performance) (Figure 3) might therefore
be associated with a positive training adaptation mainly because
of the appropriate balance in training stress and recovery.
Therefore, it is worth mentioning that the result of the present
investigation seems to be aligned with the earlier proposed ta-
pering strategy to optimize the outcomes (6). Data from the meta-
analysis conducted by Bosquet et al. (6) showed that the most
efficient tapering strategy for maximizing physiological adapta-
tions and consequently performance gains was adopting a 2-week
taper period, with an exponential reduction in training volume
without any modification in either training intensity or frequency;
it is important to highlight that the present results add novel ev-
idence for an optimal tapering strategy in soccer because most
studies on this issue have been conducted with endurance
sports (6).
The relation between dose and response is imperative to opti-
mize the training process. Coaches must be aware of the factors
that could affect this relation to appropriately manipulate the
training stimuli. The results of this study showed a linear dose-
response relation between mean weekly sRPE-TL and mean
weekly Edward’s-TL in the preseason and improvements in Yo-
yo IR1 performance. These findings are similar to those reported
in youth soccer players (1,24,33) and professional soccer players
(11,34,36) which showed some linear dose-response
Table 2
Stepwise linear regression model summary for components
assuming % change in Yo-yo IR1 (m) as a dependent variable.*†
Model R R2 Adjusted R2 SEE
1 0.698 0.487 0.450 19.07
2 0.863 0.745 0.706 13.96
*sRPE 5 session rating of perceived exertion; TL 5 training load.
†1. Predictors: (constant), weekly mean sRPE-TL. 2. Predictors: (constant), weekly mean sRPE-TL,
weekly mean Edward’s-TL.

Copyright © 2019 National Strength and Conditioning Association. Unauthorized reproduction of this article is prohibited.
 Internal Training Load and Performance in Soccer Players (2019) 00:00
relationships between sRPE-TL and HR-derived TL methods
with changes in physical fitness performance (r 5 20.64 to 0.78).
Interestingly, a recent study found a strong positive relation
between Spanish second division professional soccer players
(27.7 6 4.3 years) sRPE-TL (accumulated) (r 5 0.70; R2 5 0.48)
and the sum of RPE (r 5 0.75; R2 5 0.57) and improvements in
VIFT (11), whereas there was no significant relationship between
the Edward’s-TL (accumulated) (r 5 0.25; R2 5 0.06) and
improvements in VIFT during preseason training (i.e., a 4 week
period) (11). This result could suggest therefore that sRPE-TL is
a valid measure to monitoring the TL in professional soccer
players (11). This positive linear dose-response relationship
results also suggest that a higher TL is more likely to be beneficial
for improvement in aerobic performance in soccer players. Nev-
ertheless, this should be considered with caution because of the
large and negative associations found between sRPE-TL (re-
spiratory and muscular sRPE) and changes in physical fitness
performance (r 5 20.20 to 20.54) during the preseason period
(i.e., 9 weeks) in Spanish second division young soccer players
(20.2 6 1.9 years), suggesting that higher TLs could also be
detrimental to physical performance development (34). In addi-
tion, the authors also found negative correlation between the
change in CMJnD and sRPEmus-TL (r 5 20.52). Therefore, the
players who accumulated higher perceived TL were more likely to
decreases in CMJnD performance (34). It is worth mentioning
that the authors of this study (34) evaluated jumping performance
(i.e., CMJ, arm swing CMJ, and single-leg CMJ), sprinting
(i.e., 5- and 15-m times) and the aerobic fitness running test
(i.e., lactate thresholds) to assess physical fitness.
However, some studies showed no significant associations
between sRPE-TL and training-induced changes in markers of
aerobic fitness in youth soccer players (1,7). Based on the results
of Los Arcos et al. (34) and the aforementioned studies, a pos-
sible explanation for the differences between the results may
therefore be the deconstruction of the overall perceived exertion,
into sRPEres-TL and sRPEmus-TL. For example, Los Arcos
et al. (34) showed that the impairments in physical fitness per-
formance were associated with higher training volume and
feelings of strain in the leg muscles because of the accumulated
perceived (leg) muscular exertion. Owing to differences between
the construct of different types of physical tests, further studies
could assess longitudinal changes in a variety of physical per-
formance tests, while observing the dose-response relation for
each physical attribute in youth soccer players, to advance the
knowledge on this issue in the soccer science field. To the best of
our knowledge, this is the first study to demonstrate a negative
substantial dose-response relationship between both ITLs
(sRPE-TL and Edward’s-TL) and changes in intermittent per-
formance in youth soccer players.
Overall, the magnitude of the association indicates that ;50%
and only ;25% of the variance in performance change can be
explained by weekly sRPE-TL or mean weekly Edward’s-TL,
respectively; the final model included these 2-predictor variables
which explained ;74% of the variance in the Yo-yo IR1 per-
formance change. This result is similar to that presented by
Campos-Vasquez et al. (11), who showed that the ITL measure
(accumulated sRPE-TL and the sum of RPE) could explain about
50%, and the Edwards’-TL only ;6.3% of the variation in in-
termittent performance after a preseason in professional soccer
players.
A limitation of this study is that ITL (sRPE-TL and Edward’s-
TL) was quantified in the absence of external TL measures,
which could aid in explaining the dose-response relationship
Copyright © 2019 National Strength and Conditioning Association. Unauthorized reproduction of this article is prohibited.
| www.nsca.com
and changes in physical performance. External TL measures
should be included (i.e., time motion-analysis and global po-
sitioning system measures) in future studies concerned with the
topic of the dose-response relationship in soccer players.
Moreover, it is worth mentioning that this study was conducted
over a relatively short preseason period, and using only 1 soccer
team; it is possible that different results might be observed in
other soccer teams because of the training philosophy of the
coaches, types of TSs performed, and characteristics of the
training periodization; future work should be conducted to
extend the results of this study, to corroborate or refute the
present findings; indeed, it is important to conduct a similar
research design over longer training periods or during in-season
training.
In conclusion, the present results show that the training intensity
distribution in the 3 intensity zones (low, moderate, and high) across
all weeks were similar based on the sRPE method. Furthermore, the
large and negative associations between weekly mean ITL measures
and change in Yo-yo IR1 performance suggest that a higher ITL
across the 3 weeks of training could lead to worse Yo-yo IR1 per-
formance. The present results also showed that ITLs (weekly mean
sRPE-TL and weekly mean Edward’s-TL) explain about 50 and
25% of the variation in the Yo-yo IR1 performance changes after the
preseason, suggesting that the sRPE is useful as a global indicator of
individual training response in youth soccer players.
Practical Applications
The sRPE-TL seems to be a better method of monitoring TL
compared with the Edward’s-TL because it explains a larger
percentage of the variance in Yo-yo IR1 performance changes
in youth soccer players. Taking into account the advantages of
adopting the sRPE-TL, it could be possible for coaches to
easily evaluate the daily TL and intensity of TSs of youth
soccer players, providing staff with tools to quantify the dose-
response relationship during different types of sessions and
training periods, to optimize player training adaptations, im-
proving the likelihood of achieving higher levels of
performance.
References
1. Akubat I, Patel E, Barrett S, and Abt G. Methods of monitoring the
training and match load and their relationship to changes in fitness in
professional youth soccer players. J Sports Sci 30: 1473–1480, 2012.
2. Alexiou H and Coutts AJ. A comparison of methods used for quantifying
internal training load in women soccer players. Int J Sports Physiol Per-
form 3: 320–330, 2008.
3. Algrøy EA, Hetlelid KJ, Selier S, and Pedersen JIS. Quantifying training
intensity distribution in a group of Norwegian professional soccer players.
Int J Sports Physiol Perform 6: 70–81, 2011.
4. Bangsbo J, Iaia FM, and Krustrup P. The Yo-yo intermittent recovery test:
A useful tool for evaluation of physical performance in intermitent sports.
Sports Med 38: 37–51, 2008.
5. Borg GA. Psychophysical bases of perceived exertion. Med Sci Sports
Exerc 14: 377–381, 1982.
6. Bosquet L, Montpetit J, Arvisais D, and Mujika I. Effects of tapering on
performance: A meta-analysis. Med Sci Sports Exerc 39: 1358–1365,
2007.
7. Brink MS, Nederhof E, Visscher C, Schmikli SL, and Lemmink KA.
Monitoring load, recovery, and performance in young elite soccer players.
J Strength Cond Res 24: 597–603, 2010.
8. Brink MS, Visscher C, Arends S, Zwerver J, Post WJ, and Lemmink KA.
Monitoring stress and recovery: New insights for the prevention of injuries
and illnesses in elite youth soccer players. Br J Sports Med 44: 809–815,
2010.
 Internal Training Load and Performance in Soccer Players (2019) 00:00
9. Buchheit M, Racinais S, Bilsborough JC, Bourdon PC, Voss SC, Hocking J,
et al. Monitoring fitness, fatigue and running performance during a pre-season
training camp in elite football players. J Sci Med Sport 16: 550–555, 2013.
10. Campos-Vazquez MA, Mendez-Villanueva A, Gonzalez-Jurado JA, Leon-
Prados JA, Santalla A, and Suarez-Arrones L. Relationships between
rating-of-perceived-exertion- and heart-rate-derived internal training load
in professional soccer players: A comparison of on-field integrated train-
ing sessions. Int J Sports Physiol Perform 10: 587–592, 2015.
11. Campos-Vazquez MA, Toscano-Bendala FJ, Mora-Ferrera JC, and
Suarez-Arrones L. Relationship between internal load indicators and
changes on intermittent performance after the preseason in professional
soccer players. J Strength Cond Res 31: 1477–1485, 2017.
12. Castagna C, Impellizzeri FM, Chaouachi A, Bordon C, and Manzi V.
Effect of training intensity distribution on aerobic fitness variables in elite
soccer players: A case study. J Strength Cond Res 25: 66–71, 2011.
13. Castagna C, Impellizzeri FM, Chaouachi A, and Manzi V. Preseason
variations in aerobic fitness and performance in elite-standard soccer
players: A team study. J Strength Cond Res 27: 2959–2965, 2013.
14. Cohen J. Statistical Power Analysis for the Behavioral Science (2nd ed.).
Hillsdale, NJ: Lawrence Erlbaum, 1988.
15. Coutts A, Reaburn P, Piva TJ, and Murphy A. Changes in selected bio-
chemical, muscular strength, power, and endurance measures during de-
liberate overreaching and tapering in rugby league players. Int J Sports
Med 28: 116–124, 2007.
16. Coutts AJ, Rampinini E, Marcora SM, Castagna C, and Impellizzeri FM.
Heart rate and blood lactate correlates of perceived exertion during small-
sided soccer games. J Sci Med Sport 12: 79–84, 2009.
17. Deprez D, Coutts AJ, Lenoir M, Fransen J, Pion J, Philippaerts R, et al.
Reliability and validity of the Yo-Yo intermittent recovery test level 1 in
young soccer players. J Sports Sci 32: 903–910, 2014.
18. Deprez D, Fransen J, Lenoir M, Philippaerts RM, and Vaeyens R. The Yo-
Yo intermitente recovery test level 1 is reliable in Young high-level soccer
players. Biol Sport 32: 65–70, 2015.
19. Edwards S. High performance training and racing. In: The Heart Rate
Monitor Book. Edwards S, ed. Sacramento, CA: Feet Fleet Press, 1993. pp.
113–123.
20. Fahlman MM and Engels HJ. Mucosal IgA and URTI in American college
football players: A year longitudinal study. Med Sci Sports Exerc 37:
374–380, 2005.
21. Foster C. Monitoring training in athletes with reference to overtraining
syndrome. Med Sci Sports Exerc 30: 1164–1168, 1998.
22. Foster C, Florhaug JA, Franklin J, Gottschall L, Hrovatin LA, Parker S,
et al. A new approach to monitoring exercise training. J Strength Cond Res
15: 109–115, 2001.
23. Freitas CG, Aoki MS, Franciscon CA, Arruda AFS, Carling C, and
Moreira A. Psychophysiological responses to overloading and tapering
phases in elite young soccer players. Pediatr Exerc Sci 26: 195–202, 2014.
24. Gil-Rey E, Lezaun A, and Los Arcos A. Quantification of the perceived
training load and its relationship with changes in physical fitness perfor-
mance in junior soccer players. J Sports Sci 33: 2125–2132, 2015.
Copyright © 2019 National Strength and Conditioning Association. Unauthorized reproduction of this article is prohibited.
25. Hopkins WG, Marshall SW, Batterham AM, and Hanin J. Progressive
statistics for studies in sports medicine and exercise science. Med Sci Sports
Exerc 41: 3–13, 2009.
26. Impellizzeri FM, Rampinini E, Coutts AJ, Sassi A, and Marcora SM. Use
of RPE-based training load in soccer. Med Sci Sports Exerc 36:
1042–1047, 2004.
27. Impellizzeri FM, Rampinini E, and Marcora SM. Physiological assessment
of aerobic training in soccer. J Sports Sci 23: 583–592, 2005.
28. Jeong TS, Reilly T, Morton J, Bae SW, and Drust B. Quantification of the
physiological loading of one week of “pre-season” and one week of “in-
season” training in professional soccer players. J Sports Sci 29:
1161–1166, 2011.
29. Kellmann M, Bertollo M, Bosquet L, Brink M, Coutts AJ, Duffield R, et al.
Recovery and performance in sport: Consensus statement. Int J Sports
Physiol Perform 13: 240–245, 2018.
30. Kenttä G and Hassmén P. Overtraining and recovery. A conceptual
model. Sports Med 26: 1–16, 1998.
31. Lago-Penas C and Sampaio J. Just how important is a good season start?
Overall team performance and financial budget of elite soccer clubs.
J Sports Sci 33: 1214–1218, 2015.
32. Lovell TWJ, Sirotic AD, Impellizzeri FM, and Coutts AJ. Factors affecting
perception of effort (session rating of perceived exertion) during rugby
league training. Int J Sports Physiol Perform 8: 62–69, 2013.
33. Los Arcos A, Martı́nez-Santos R, Yancil J, and Mendez-Villanueva A.
Monitoring perceived respiratory and muscular exertions and physical
fitness in young professional soccer players during a 32-week period.
Kinesiol 49: 1–8, 2017.
34. Los Arcos A, Martı́nez-Santos R, Yanci J, Mendiguchia J, and Méndez-
Villanueva A. Negative associations between perceived training load,
volume and changes in physical fitness in professional soccer players.
J Sports Sci Med 14: 394–401, 2015.
35. Los Arcos A, Yanci J, Mendiguchia J, and Gorostiaga EM. Rating of
muscular and respiratory perceived exertion in professional soccer play-
ers. J Strength Cond Res 28: 3280–3288, 2014.
36. Manzi V, Bovenzi A, Impellizzeri FM, Carminati I, and Castagna C. In-
dividual training-load and aerobic-fitness variables in premiership soccer
players during the precompetitive season. J Strength Cond Res 27:
631–636, 2013.
37. Manzi V, D’Ottavio S, Impallizzeri FM, Chaouachi A, Chamari K, and
Castagna C. Profile of weekly training load in elite male professional
basketball players. J Strength Cond Res 24: 1399–1406, 2010.
38. Milanez VF, Pedro RE, Moreira A, Boullosa DA, Salle-Neto F, and
Nakamura FY. The role of aerobic fitness on session rating of per-
ceived exertion in futsal players. Int J Sports Physiol Perform 6:
358–366, 2011.
39. Viru A and Viru M. Nature of training effects. In: Exercise and Sport
Science. Garret WE and Kirkendall DT, eds. Philadelphia, PA: Lippincott
Williams and Wilkins, 2000. pp. 67–95.
40. Wrigley R, Drust B, Stratton G, Scott M, and Gregson W. Internal training
load and performance in soccer players. J Sports Sci 30: 1573–1580, 2012.