Journal of Science and Medicine in Sport 23 (2020) 20–26

Contents lists available at ScienceDirect

Journal of Science and Medicine in Sport

journal homepage: www.elsevier.com/locate/jsams

Original research

Changes in performance markers and wellbeing in elite senior

professional rugby union players during a pre-season period:
Analysis of the differences across training phases
Adam Grainger a,∗ , Ross Neville a , Massimiliano Ditroilo a , Paul Comfort b
a
School of Public Health, Physiotherapy and Population Science, University College Dublin, Ireland
b
School of Health & Society, University of Salford, United Kingdom

a r t i c l e i n f o a b s t r a c t

Article history: Objectives: To assess the magnitude of change and association with variation in training load of two
Received 29 March 2019 performance markers and wellbeing, over three pre-season training blocks, in elite rugby union athletes.
Received in revised form 13 July 2019 Design: Observational.
Accepted 11 August 2019
Methods: Twenty-two professional players (age 25 ± 5 years; training age 6 ± 5 years; body mass,
Available online 17 August 2019
99 ± 13 kg; stature 186 ± 6 cm) participated in this study, with changes in lower (CMJ height) and upper
body (bench press mean speed) neuromuscular function and self-reported wellbeing (WB) assessed
Keywords:
during an 11-week period.
Bench press
Countermovement jump
Results: There was a small increase in CMJ height (0.27, ±0.17 – likely substantial; standardised effect
Self-reports size, ±95% confidence limits – magnitude-based inference) (p = 0.003), bench press speed (0.26, ±0.15 –
Periodisation likely substantial) (p = 0.001) and WB (0.26, ±0.12 – possibly substantial) (p < 0.0001) across the pre-season
Training load period. There was a substantial interaction in the effect of training load on these three variables across
the three training phases. A two-standard deviation (2SD) change in training load was associated with:
a small decrease in CMJ height during the power phase (−0.32, ±0.19 – likely substantial) (p = 0.001); a
small reduction in bench press speed during the hypertrophy phase (−0.40, ±0.32 – likely substantial)
(p = 0.02); and a small reduction in WB during the strength phase (−0.40, ±0.24 – very likely substantial)
(p < 0.0001). The effects of changes in training load across other phases were either likely trivial, only
possibly substantial, or unclear.
Conclusions: The effect of training load on performance can vary both according to the type of training
stimulus being administered and based on whether upper- or lower-body outcomes are being measured.
© 2019 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.

Practical implications 1. Introduction

• Assessment of CMJ height is resistant to changes in training load Rugby union players require both high levels of strength
during the hypertrophy phase, yet this lack of sensitivity could and power for the contact-wrestling activities of match-play
be explained by jump strategy employed. and advanced aerobic and anaerobic capacity for repeated sprint
• Assessment of bench press speed would appear to indicate that involvements.1,2 Although typically limited in duration (6–12
the likely range within which athletes can vary hypertrophy weeks depending upon availability),3 pre-season training periods
training loads is narrower than for strength and power. tend to focus on the development of many attributes, often incor-
• Programming to increase physical performance should be porating two to three distinct phases (hypertrophy, strength or
undertaken with consideration of psychological wellbeing main- power). As noted in prior rugby union research, phase duration may
tenance across the training cycle. vary depending upon player training age and desired outcome,4
yet typically lasts 3–6 weeks.5 Training volume within pre-season
periods is reported to be high (3500–4500 arbitrary units) with con-
current training regularly utilized,6,7 and focuses upon developing
the multifaceted aspects of strength and conditioning required for
∗ Corresponding author. modern professional rugby union.
E-mail address: [email protected] (A. Grainger).

https://doi.org/10.1016/j.jsams.2019.08.012
1440-2440/© 2019 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.
 A. Grainger et al. / Journal of Science and Medicine in Sport 23 (2020) 20–26 21

Fatigue from training stress has been frequently reported in elite week players were provided with an individually prescribed train-
level rugby union8–10 , with neuromuscular function (NMF) and ing programme, yet no testing or training load data was collected.
perceptual measures used as markers. NMF is the control of skele- Lower body NMF was assessed using CMJ height. Following
tal muscle action by spinal and supraspinal mechanisms, with the a standardised warm up including dynamic stretching and sub-
monitoring of NMF used as a method for detecting fatigue or neu- maximal CMJ’s, two scored CMJ’s were performed, with the best
romuscular impairment.11 The evidence also confirms that fatigue performance used in the analysis. Players were familiar with the
accumulates, or is compounded, over a training phase,6,12 high- CMJ, having performed it on a regular basis over preceding months.
lighting the need to better understand the association between All jumps were performed on an OptoJump optical measuring sys-
training load and changes in NMF and wellbeing (WB). Appropriate tem (Microgate, Bolzano, Italy). The OptoJump system has been
training loads that result in optimal adaptation, yet maintain play- reported to demonstrate excellent test re-test reliability (intraclass
ers in a positive state of readiness (that determines their ability to correlation coefficient [ICC] = 0.985) and low variability (2.7% coef-
effectively achieve their performance potential), are therefore key ficient of variation).21 Flight time was used to calculate jump height
and require continuous monitoring.13 using the following equation:
The monitoring of specific performance measures to assess NMF Jump Height = (9.81 m.s−2 × flight time2 )/8
in rugby union is a relatively recent area of research,14 with tests Upper body NMF was assessed using bench press speed. Follow-
typically involving objective NMF assessment9,12 and subjective ing a standardised warm up of dynamic stretching and plyometric
self-reported assessment of fatigue.8 A combination of lower- and push-ups, two bench press repetitions were performed on a Pre-
upper body NMF and some multi-dimensional measure of athlete cor smith machine (Surrey, UK). The best repetition (highest mean
WB is therefore routinely incorporated in most rugby union clubs.15 velocity) was used in the analysis. A 20 kg load (including bar
In an intermittent contact sport such as rugby, eccentric muscle weight) was selected based on prior research.22 As the reliabil-
actions and blunt muscle trauma are the likely contributors to a ity of ballistic bench press is noted as more reliable (mean power
decrement in NMF.8 The most common and easily implemented CV = 4.4%)23 than plyometric push up (mean power CV = 43.4%),10
form of NMF assessment is the countermovement jump (CMJ), with support for its use is clear. Further justification for the use of bench
the reliability and sensitivity of CMJ to detect changes in training press can be explained by players being more accustomed to the
load over time confirmed.9,12,16 The importance of upper body mea- bench press movement. Added to this, plyometric push ups are
sures in assessments of NMF has been identified,17 whereby the commonly calculated by flight time and from force-time data based
plyometric push-up appears sensitive to detecting change in NMF upon calculations used for jumping. Such measurement is erro-
in rugby league players.18 neous and as during a push up the players feet will remain in
Self-reported WB questionnaires are also commonly used as a contact with the floor and therefore the entire mass of the athlete
measure of how players are handling training load,10 with evidence is not accelerated, therefore further supporting the use of bench
supporting the use of such measures19 for monitoring acute and press instead. Players selected their own hand position and main-
chronic training loads20 and regularly used in elite rugby union tained this throughout the testing period. Subjects were instructed
settings to optimise readiness.8 Saw et al.18 recommend that ath- to lower the barbell in a controlled manner until the bar made con-
letes report their objective well-being on a regular basis alongside tact with the chest, at which point they pressed up with maximal
other neuromuscular monitoring, as such tools have been shown intent, keeping the bar in their hands throughout. A linear encoder
to complement each other. (GymAware, Canberra, Australia) was used to assess mean velocity,
The majority of NMF and WB testing has been conducted post- with acceptable test-retest reliability reported in a recent study
match-play during competition phases and not across pre-season (ICC 0.78–0.87; SEM% 7.9–9.9%).24
training phases.10,12 Despite recent studies illustrating changes in WB was assessed using a self-report questionnaire. All players
lower body NMF of elite academy rugby union players during pre- completed a custom-made questionnaire upon waking in their own
season12 and subjective WB in a single training week in southern homes via an online player management tool (The Sports Office, UK).
hemisphere players,8 there is no research that assesses subjec- This five part questionnaire assessed subjective responses to ques-
tive measures of readiness and both upper body and lower body tions relating to sleep, muscle soreness, energy levels, mood and
objective measures of readiness during pre-season in elite senior appetite; based on prior recommendations.10 WB assessment was
level rugby union players. With this in mind, this study aimed to: scored out of 5 (arbitrary units), with 5 being the highest rating a
(1) assess the change in upper and lower body NMF and WB over player could provide for each part of the questionnaire and 1 being
three distinct pre-season training blocks; and (2) assess which per- the least.
formance marker is most sensitive to changes in accumulation of Total training load was quantified using the session rating of per-
pre-season training load in elite rugby players. ceived exertion method within 30 min of each session finishing. In
line with previous rugby research studies, values have been dis-
played in arbitrary units.8,12 A rate of perceived exertion was asked
2. Methods of each individual within thirty minutes of each training session
finishing, using the 10-point Borg scale. This approach to moni-
Twenty-two senior professional rugby union players (age 25 ± 5 toring training load25 has been validated as a reliable measure to
years; training age 6 ± 5 years; body mass, 99 ± 13 kg; stature assess across weekly training loads in elite rugby union26 and has
186 ± 6 cm) were recruited from an English Premiership rugby been regularly implemented in rugby union research.7 Daily train-
union club. Participants provided written informed consent and ing loads (Monday to Sunday) were combined to provide a total
the study was ethically approved by a Salford University Insti- for each week, thereby allowing for comparisons against the three
tutional Review Board. A repeated measures design was used to performance markers (CMJ height, bench press mean speed and
assess changes in lower and upper body NMF (CMJ height and bench self-reported WB) with weekly values presented in Fig. 1. A typical
press mean speed, respectively) and self-reported WB across a pre- training week is detailed in the schematic in Fig. 2, with measures
season period. Across the 11-week training period, 10 weeks were of NMF taken on the first and fourth day of each training week. Self-
programmed by the coaching staff of the rugby club, with week reported WB was assessed on the first, second, fourth, fifth and sixth
6 being a rest week. Weeks 1–3 were programmed to focus on day of each week.
hypertrophy-based resistance training; weeks 4, 5 and 7 focused Data were analysed using a linear mixed model procedure (Proc
on strength; and weeks 8–11 focused on power. During the rest Mixed) in the Statistical Analysis System (SAS Studio – University
22 A. Grainger et al. / Journal of Science and Medicine in Sport 23 (2020) 20–26

Fig. 1. Shows the total training load in arbitrary units by week and phase across the 11-week preseason training period. Data are raw means and standard deviations.

Fig. 2. Schematic representation of the research design Hyp = hypertrophy-based resistance training; Str = strength-based resistance training; Pow = power-based resistance
training; CMJ = counter movement jump; BP = bench press; both these tests took place Monday and Thursday each week; WB = well being; the questionnaire was administered
Monday, Tuesday, Thursday, Friday, Saturday of each week.

Edition). A mixed model was used to estimate: (i) individual players’ and (ii) in almost identical models. An additional benefit of using
changes in NMF and WB across the three distinct training phases mixed modelling (over, say, a two-way repeated measures ANOVA)
(hypertrophy, strength and power; and (ii) the effect of weekly was that we could use least squares means estimate statements
training load on NMF and WB within each of the three training (for (i)) and estimate statements (for (ii)) to derive the between-
phases. That is, we could nest the weekly repeated measurements within subjects and moderating effects from these almost identical
within players, specify the week of testing as a between-subjects models.
random factor, and estimate the effect of training phase as a fixed A dummy variable was specified to account for the effect of the
factor.27 For the final analysis, a two-way reliability model was players having a week off in week six. A clear effect for the dummy
specified with player ID and the week of testing as two separate ran- variable was found for CMJ height only (F1, 149 = 8.35, p = 0.004).
dom factors. The first random factor allowed for different intercepts Therefore the magnitudes of the changes in mean bench press
(starting points) and slopes (changes over time) to be specified speed and WB were analysed without incorporation of a dummy
between individual players. The second random factor allowed for variable in the statistical model. Changes in NMF and WB were eval-
us to account for the typical difference in each parameter from week uated as fixed effects by deriving the mean changes between each
to week. Training phase was specified in the model as a fixed factor of the distinct training phases. The fixed effect of training load was
to estimate changes over time between phases (i), above. Training evaluated as a linear numeric predictor; that is, as the change in the
phase was interacted with an additional fixed factor summarising two NMF markers and WB associated with a two standard devia-
weekly training load to estimate the effect of training load on NMF tion (2SD) change in training load. A 2SD change was used because
and WB within each of the three training phases (ii). Mixed mod- this enabled us to estimate differences in the dependent variables
elling was appropriate because it enabled us to answer both (i) between players with below (mean – 1SD) and above (mean +1SD)
 A. Grainger et al. / Journal of Science and Medicine in Sport 23 (2020) 20–26 23

Fig. 3. Shows changes in counter movement jump height across the distinct training phases. The smallest worthwhile change was derived by means of standardisation.
Asterisks indicate changes that are clear at the 95% level and likelihood that the true value of the effect is substantial, as follows: *possibly, **likely, ***very likely, ****most
likely.

Fig. 4. Shows changes in bench press velocity across three distinct training phases. The smallest worthwhile change was derived by means of standardisation. Asterisks
indicate changes that are clear at the 95% level and likelihood that the true value of the effect is substantial, as follows: *possibly, **likely, ***very likely, ****most likely.

values for weekly training load.27 The magnitudes of these effects 0.6–1.2, moderate; 1.2–2.0, large; >2.0, very large.28 To estimate
were assessed by means of standardisation (by dividing the effect the importance and uncertainty in the effects, probabilities for the
by the appropriate between-subjects SD at baseline) and evalu- true magnitudes of the effects were evaluated using the following
ated according to the following scale: <0.2, trivial; 0.2–0.6, small; scale: <0.5%, most unlikely; 0.5–5%, very unlikely; 5–25%, unlikely;
24 A. Grainger et al. / Journal of Science and Medicine in Sport 23 (2020) 20–26

Fig. 5. Shows changes in wellbeing across three distinct training phases. The smallest worthwhile change was derived by means of standardisation. Asterisks indicate changes
that are clear at the 95% level and likelihood that the true value of the effect is substantial, as follows: *possibly, **likely, ***very likely, ****most likely.

25–75%, possibly; 75–95%, likely; 95–99.5%, very likely; >99.5%, most

likely.27 Bonferroni adjustments were applied for multiple com-
parisons, and uncertainty in the estimate was expressed as 95%
confidence limits. The resulting effects were deemed clear when
the chances of a positive and negative magnitude was possibly
likely and very unlikely, respectively.27 Owing to recent criticisms
about both magnitude-based29 and statistical significance-based30
approaches to inference, in the results, we present p values along-
side the effect magnitudes and their accompanying qualitative
probabilities inferences.

3. Results

There were clear changes in CMJ height (F3, 34.6 = 345.93, p <
0.0001), mean bench press speed (F3, 41.3 = 752.77, p < 0.0001),
and WB (F3, 40.2 = 898.40, p < 0.0001) over the pre-season training
period in total. As for the effect sizes, there was a small increase in
CMJ height (Fig. 3) (0.27, ±0.17 – likely substantial; standardized
effect size, ±95% confidence limits – magnitude-based inference)
(p = 0.003), bench press speed (Fig. 4) (0.26, ±0.15 – likely substan- Fig. 6. Shows the relationship between training load, two performance markers
tial) (p = 0.001), and WB (Fig. 5) (0.45, ±0.21 – very likely substantial) and wellbeing across three distinct training phases. Data are standardised effects
(p < 0.0001). Whilst CMJ height and bench press speed increased and represent changes in the outcome variable associated with 2SD difference in
somewhat linearly over the pre-season period – i.e., the most training load. Asterisks indicate effects that are clear at the 95% level and likelihood
that the true value of the effect is substantial, as follows: *possibly, **likely, ***very
meaningful changes observed for CMJ height (0.27, ±0.17 – likely likely, ****most likely.
substantial) (p = 0.003) and bench press speed (0.26, ±0.15 – likely
substantial) (p = 0.001) were based on the comparison of training
block three (power) and one (hypertrophy) – differences in the (F3, 187 = 7.49, p < 0.0001), indicating that the effect of training load
changes in WB across phases suggests that there was a slightly more varied within each phase (Fig. 6).
meaningful increase during the final power phase (0.35, ±0.28 – Whilst a 2SD change in training load was associated with a
likely substantial) (p < 0.0001) – i.e. comparing block three (power) small decrease in CMJ height during the power phase (−0.32, ±0.19
and two (strength). – likely substantial) (p = 0.001), the effect of training load on CMJ
Preliminary analyses showed a clear interaction effect between height during the strength (-0.27, ±0.46 – possibly substantial)
training load and training phase for CMJ height (F3, 153 = 4.08, (p = 0.25) and hypertrophy (0.04, ±0.33 – unclear) (p = 0.82) phases
p < 0.01), bench press speed (F3, 174 = 2.78, p = 0.042), and WB was only possibly substantial and unclear, respectively.
 A. Grainger et al. / Journal of Science and Medicine in Sport 23 (2020) 20–26 25

Changes in training load were associated with a small decrease ferences make comparisons across studies difficult, some apparent
in bench press speed during the hypertrophy phase (-0.40, ±0.32 – similarities are worth noting. Gathercole et al.16 argued that ath-
likely substantial) (p = 0.02). There was also a clear effect of training letes adapt their CMJ strategy, by increasing squat depth, which
load on bench press speed during the strength phase (0.27, ±0.46 – increases the duration for the application of force and therefore
likely substantial) (p = 0.25). Meanwhile, the effect of training load maintains impulse, to avoid any decreased jump height value being
on bench press speed during the power phase was likely trivial presented, despite NMF existing. Within this study, it could there-
(−0.09, ±0.24 – likely trivial) (p = 0.42). fore be argued that the lack of sensitivity shown in jump height
Changes in training load had a small negative effect on WB could be explained by the movement strategies of subjects not
during the strength phase (−0.40, ±0.24 – very likely substantial) having been assessed. Future research should perhaps look at drop
(p < 0.0001). The effects during the hypertrophy (−0.12, ±0.11 – jump varieties that involve less ability to adopt jump strategy and
very likely trivial) (p = 0.03) and power (0.04, ±0.09 – most likely more complex kinematics,30 with increased sensitivity noted com-
trivial) (p = 0.39) phases were both clear, but clearly trivial. pared to CMJ in elite soccer.31 Similarly, single leg landing and
balance measures have been noted as sensitive to detecting residual
fatigue post professional rugby union match-play.32
4. Discussion Differences in the recovery rates of upper- and lower-body
musculature33 might help to explain the differences in the effect of
Our results show that there were small increases in upper-body training load on CMJ and bench press velocity across phases. Fig. 6
NMF and WB across the pre-season period, and that there were indicates that the players could have benefitted from additional
differences in the effect of training load on upper- and lower- load during the strength phase, though there appear to be individ-
body NMF within each of the three different training phases. Visual ual responses here too (given the width of the confidence limits),
inspection of Fig. 6 indicates that the effects of training load on and it is unclear as to whether such an increase in training load
CMJ height were greater during the strength and during the power would have been detrimental to the peaking that clearly occurred
phases than during the hypertrophy phase. The effects of training between the strength and final power phase of the pre-season pro-
load on bench press velocity during the hypertrophy and during gramme (Fig. 4). What is clear is that substantial changes in training
the strength phases was greater than during the power phase. In load during the hypertrophy phase would have negatively affected
fact, Fig. 6 shows that changes in training load affected bench press bench press speed. This suggests that the likely range within which
velocity in opposite directions, negatively and positively, during the athletes can vary upper-body NMF hypertrophy training loads is
hypertrophy and strength phases respectively. The effect of training narrower than for strength and power. Again, translated into raw
load on WB was greater during the strength phase than during the units, our data suggests that an approximate 1300 AU change in
hypertrophy or power phases. Our results suggest, therefore, that training load would be required during the hypertrophy phase to
the effect of training load on NMF and WB varies both according lead to a decrease in bench press velocity by the smallest worth-
to the type training stimulus being administered and the outcome while change (i.e., by 0.20 standardized units, or, in this sample,
being measured. approximately 0.04 m per second).
Where CMJ was negatively affected by an increase in train- Hypertrophy training is by its nature very demanding, with both
ing load, we believe that this could possibly be explained by the mechanical and metabolic stress accumulated over time as a result
nature of the training phase stimulus. CMJ height is a proxy measure of higher repetition work and increased time under tension. The
for relative power output, and power training involves more cen- typical stress response of hypertrophy training may therefore be
tral nervous system involvement and resultant fatigue. Therefore, too sensitive for alterations in training load to become beneficial.
all things being equal (namely, jump strategy, which we consider Our results indicate that programming was appropriate to main-
below), it should be unsurprising that CMJ height decreased as a tain, and even enhance, WB. Fig. 5 shows that players’ WB clearly
power phase training load increased. Translated into raw units, peaked in the final phase of the pre-season period. The main find-
which can make this analysis even more relevant to practitioners, ing worth discussing here is the extent to which an increase in
our data suggests that an approximate 1500 AU change in training training load during the strength phase negatively affected play-
load would be required during the power phase to lead to a decrease ers’ WB. We estimate that as a little as a 500 AU change in training
in CMJ height by the smallest worthwhile change (i.e., by 0.20 stan- load would be required during the hypertrophy phase to nega-
dardized units, or approximately 1.3 cm). A different study design tively affect WB by the smallest worthwhile change. Given ongoing
than was possible here – i.e., an experimental study where training debates about the quality of player welfare in contact field sports
load is directly manipulated rather than observationally recorded such as rugby union,34 it is critical from a practitioner’s perspec-
– is required to answer this question more directly. tive to consider why the effect of the training load on WB would
That the effect of training load on CMJ height during the strength vary according to the training stimulus being administered. It
phase was only possibly small (i.e., meaning that is was also pos- could be argued that strength training is more psychologically
sibly trivial) suggests that there were individual differences in the taxing than hypertrophy or power training, as one is required to
effect of the strength training stimulus. The lower and upper con- work at near-maximal exertion levels with an attentional focus
fidence limits for the effects during the strength phase appear to required to complete movements in a safe and efficient man-
support this claim, with Fig. 6 showing that these CLs are substan- ner. Added to this, to increase strength, it stands to reason that
tially wider than for the effects during the hypertrophy and power elite athletes need (much more than general population samples)
phases. Following research by Argus et al.4 and Roe et al.,12 we sus- to push beyond physical limits,35 which is both psychologically
pect that chronological and training age are possible moderating demanding and requires a high level of alignment between one’s
factors that could explain individual responses to increased load physical self-concept and actual physical strength.36 It is also well
during a strength training phase. Considering reliability research known that the maximal training at increased frequencies that is
by Roe et al.10 assessing well-being questionnaires in rugby union, required to produce strength adaptations in elite athletes can lead
player motivation could also be considered a factor within the indi- to overtraining,37 which also has psychological consequences that
vidual responses collected. are known to affect performance.37
Previous research has also focused on the effect of training Research by Argus et al.4 and Roe et al.12 suggest that ath-
load on lower body performance. This is from both academy rugby lete conditioning (i.e. fitness) and preparedness (i.e. maturity) are
union9 and senior level rugby union.8 Whilst methodological dif- major factors in the assessment of performance change relative
26 A. Grainger et al. / Journal of Science and Medicine in Sport 23 (2020) 20–26

to altered training load. These factors were not addressed in this 9. Kennedy RA, Drake D. The effect of acute fatigue on countermovement jump
study and therefore ought to be considered as our primary limi- performance in rugby union players during preseason. J Sports Med Phys Fitness
2017; 57(10):1261–1266.
tations. Future research should therefore account for the possible 10. Roe G, Darrall-Jones J, Till K et al. Between-days reliability and sensitivity of
moderating effects of these factors (e.g. athlete conditioning and common fatigue measures in rugby players. Int J Sports Physiol Perform 2016;
athlete chronological and training age) NMF, WB, training stim- 11(5):581–586.
11. Noakes TD. Physiological models to understand exercise fatigue and the adap-
ulus, and training load. Sampling variation is also a limitation to tations that predict or enhance athletic performance. Scand J of Med Sci Sports
be noted when research has been conducted in a single setting. 2000; 10(3):123–145.
Despite the ecological validity of our study design and its findings, 12. Roe GAB, Darrall-Jones JD, Till K et al. Preseason changes in markers of lower
body fatigue and performance in young professional rugby union players. Eur J
future research would benefit from data collection across multiple
Sports Sci 2016; 16(8):981–988.
teams. This would allow for a more sophisticated, multilevel, mod- 13. Gallo TF, Cormack SJ, Gabbett TJ et al. Pre-training perceived wellness
elling of outcomes and therefore enhance generalizability.38 Future impacts training output in Australian football players. J Sports Sci 2016;
34(15):1445–1451.
studies could also enhance the external validity of our model by
14. Tavares F, Smith TB, Driller MW. Fatigue and recovery in rugby: a review. Sports
addressing the relationship between NMF, WB, training stimulus Med 2017; 47(8):1515–1530.
and training load across field sports. Lastly, as a result of these find- 15. Starling LT, Lambert MI. Monitoring rugby players for fitness and fatigue: what
ings, practitioners and researchers are advised to use more sensitive do coaches want? Int J Sports Physiol Perform 2018; 13(6):777–782.
16. Gathercole R, Sporer B, Stellingwerff T et al. Alternative countermovement-jump
measures and meaningful variables to monitor NMF. An example analysis to quantify acute neuromuscular fatigue. Int J Sports Physiol Perform
of this would be the analysis of force time data using force plates, 2015; 10(1):84–92.
which permit the comparison of changes in jump strategy during a 17. Johnston RD, Gibson NV, Twist C et al. Physiological responses to an intensified
period of rugby league competition. J Strength Cond Res 2013; 27(3):643–654.
CMJ. 18. Saw AE, Main LC, Gastin PB. Monitoring the athlete training response: subjective
self-reported measures trump commonly used objective measures: a systematic
review. Br J Sports Med 2016; 50(5):281-.
5. Conclusions 19. Gastin PB, Meyer D, Robinson D. Perceptions of wellness to monitor adaptive
responses to training and competition in elite australian football. J Strength Cond
As a result of this research, practitioners are advised to be par- Res 2013; 27(9):2518–2526.
20. Glatthorn JF, Gouge S, Nussbaumer S et al. Validity and reliability of optojump
ticularly attuned to programming of training load and required photoelectric cells for estimating vertical jump height. J Strength Cond Res 2011;
recovery time course during training phases. Evidence presented 25(2):556–560.
would suggest that the effect of training load on NMF and well- 21. Ostrowski KJ, Wilson GJ, Weatherby RP et al. The effect of weight training volume
on hormonal output and muscular size and function. J Strength Cond Res 1997;
being varies both according to the type of training stimulus being
11:148–154.
administered and upon the outcome being measured. Specifically, 22. García-Ramos A, Haff GG, Padial P et al. Reliability of power and velocity variables
the effect of changes in training load on CMJ was greater during collected during the traditional and ballistic bench press exercise. Sports Biomech
2018; 17(1):117–130.
strength and power phases, the effect of changes in training load on
23. Orange ST, Metcalfe JW, Marshall P et al. Test-retest reliability of a commer-
bench press velocity was greater during hypertrophy and strength cial linear position transducer (GymAware PowerTool) to measure velocity and
phases, and wellbeing was only substantially affected by changes power in the back squat and bench press. J Strength Cond Res 2018. http://dx.
in load during the strength phase. The importance of the place- doi.org/10.1519/JSC.0000000000002715. Published ahead of print.
24. Foster C, Florhaug JA, Franklin J et al. A new approach to monitoring exercise
ment of strength and power phases are emphasized given that CMJ training. J Strength Cond Res 2001; 15(1):109–115.
peak occurred in the middle of pre-season. However, the aforemen- 25. Comyns T, Flanagan EP. Applications of the session rating of perceived exertion
tioned lack of sensitivity of some measures and variables needs system in professional rugby union. Strength Cond J 2013; 35(6):78–85.
26. Bell BA, Ene M, Smiley W et al. A multilevel model primer using SAS PROC MIXED,
consideration within practices implemented. In: Paper presented at: SAS Global Forum., 2013.
27. Hopkins WG, Batterham AM, Marshall SW et al. Progressive statistics.
Sportscience 2009; 13(39):55–70.
Acknowledgements 28. Sainani KL. The Problem with “Magnitude-based Inference”. Med Sci Sports Exerc
2018; 50(10):2166–2176.
We would like to thank the staff and players at Sale Sharks Rugby 29. Amrhein V, Greenland S, McShane B. Scientists rise up against statistical signif-
icance. Nature 2019; 567:305–307.
Club for their involvement in this research. The research was not 30. Hamilton D. Drop jumps as an indicator of neuromuscular fatigue and recovery
supported by any external financial support. in elite youth soccer athletes following tournament match play. J Austrailian
Strength Cond 2009; 17(4):3–8.
31. Fitzpatrick JF, Akenhead R, Russell M et al. Sensitivity and reproducibility of a
References fatigue response in elite youth football players. Sci Med Football 2019. http://dx.
doi.org/10.1080/24733938.2019.1571685. Published ahead of print.
1. Darrall-Jones JD, Jones B, Till K. Anthropometric, sprint and high intensity run- 32. Troester JC, Duffield R. Monitoring residual 36 h post-match neuromuscular
ning profiles of english academy rugby union players by position. J Strength Cond fatigue in rugby union; a role for postural control? Eur J Sport Sci 2019. http://
Res 2015; 30(5):1348–1358. dx.doi.org/10.1080/17461391.2019.1606941. Published ahead of print.
2. Barr MJ, Sheppard JM, Gabbett TJ et al. Long-term training-induced changes in 33. Hoffman JR, Kraemer WJ, Fry AC et al. The effects of self-selection for frequency
sprinting speed and sprint momentum in elite rugby union players. J Strength of training in a winter conditioning program for football. J Strength Cond Res
Cond Res 2014; 28(10):2724–2731. 1990; 4(3):76–82.
3. Cross MJ, Williams S, Trewartha G et al. The influence of in-season training loads 34. Quarrie KL, Raftery M, Blackie J et al. Managing player load in professional
on injury risk in professional rugby union. Int J Sports Physiol Perform 2015; rugby union: a review of current knowledge and practices. Br J Sports Med 2016;
11(3):350–355. 51(5):421–427.
4. Argus CK, Gill ND, Keogh JWL et al. Changes in strength, power, and steroid 35. Soligard T, Schwellnus M, Alonso JM et al. How much is too much? (Part 1)
hormones during a professional rugby union competition. J Strength Cond Res International Olympic Committee consensus statement on load in sport and
2009; 23(5):1583–1592. risk of injury. Br J Sports Med 2016; 50(17):1030–1041.
5. Gamble P. Periodization of training for team sports athletes. Strength Cond J 2006; 36. Baker J, Horton S, Robertson-Wilson J et al. Nurturing sport expertise: factors
28(5):56–66. influencing the development of elite athlete. J Sports Sci Med 2003; 2(1):1–9.
6. Argus CK, Gill N, Keogh J et al. Effects of a short-term pre-season training pro- 37. Marsh HW, Perry C, Horsely C, Roche L. Multidimensional self-concepts of elite
gramme on the body composition and anaerobic performance of professional athletes: How do they differ from the general population? J Sport Exerc Psychol
rugby union players. J Sport Sci 2010; 28(6):679–686. 1995; 17(1):70–83.
7. Bradley WJ, Cavanagh BP, Douglas W et al. Quantification of training load, 38. Hopkins WG. New view of statistics, 2002. Available at: http://sportsci.org/
energy intake, and physiological adaptations during a rugby preseason: a case resource/stats. Accessed March 20, 2019.
study from an elite European rugby union squad. J Strength Cond Res 2015;
29(2):534–544.
8. Tavares F, Healey P, Smith T et al. The effect of training load on neuro-
muscular performance, muscle soreness and wellness during an in-season
non-competitive week in elite rugby athletes. J Sports Med Phys Fitness 2017;
58(11):1565–1571.
Reproduced with permission of copyright owner. Further reproduction
prohibited without permission.