ORIGINAL RESEARCH

published: 21 March 2018

doi: 10.3389/fphys.2018.00247

RPE vs. Percentage 1RM Loading in

Periodized Programs Matched for
Sets and Repetitions
Eric R. Helms 1*, Ryan K. Byrnes 2 , Daniel M. Cooke 2 , Michael H. Haischer 1 ,
Joseph P. Carzoli 2 , Trevor K. Johnson 2 , Matthew R. Cross 1 , John B. Cronin 1,3 ,
Adam G. Storey 1 and Michael C. Zourdos 2
1
Sport Performance Research Institute New Zealand, Auckland University of Technology, Auckland, New Zealand, 2 Muscle
Physiology Laboratory, Department of Exercise Science and Health Promotion, Florida Atlantic University, Boca Raton, FL,
United States, 3 School of Exercise, Biomedical and Health Sciences, Edith Cowan University, Perth, WA, Australia

Purpose: To investigate differences between rating of perceived exertion (RPE) and

Edited by:
percentage one-repetition maximum (1RM) load assignment in resistance-trained males
Billy Sperlich, (19–35 years) performing protocols with matched sets and repetitions differentiated by
University of Würzburg, Germany
load-assignment.
Reviewed by:
Pedro Jiménez Reyes, Methods: Participants performed squats then bench press 3x/weeks in a daily
Universidad Católica San Antonio de undulating format over 8-weeks. Participants were counterbalanced by pre-test 1RM
Murcia, Spain
Tyler D. Williams,
then assigned to percentage 1RM (1RMG, n = 11); load-assignment via percentage
Samford University, United States 1RMs, or RPE groups (RPEG, n = 10); participant-selected loads to reach target RPE
*Correspondence: ranges. Ultrasonography determined pre and post-test pectoralis (PMT), and vastus
Eric R. Helms
lateralis muscle thickness at 50 (VLMT50) and 70% (VLMT70) femur-length.
[email protected]
Results: Bench press (1RMG +9.64 ± 5.36; RPEG + 10.70 ± 3.30 kg), squat (1RMG
Specialty section: + 13.91 ± 5.89; RPEG + 17.05 ± 5.44 kg) and their combined-total 1RMs (1RMG +
This article was submitted to
Exercise Physiology, 23.55 ± 10.38; RPEG + 27.75 ± 7.94 kg) increased (p < 0.05) in both groups as did PMT
a section of the journal (1RMG + 1.59 ± 1.33; RPEG +1.90 ± 1.91 mm), VLMT50 (1RMG +2.13 ± 1.95; RPEG
Frontiers in Physiology
+ 1.85 ± 1.97 mm) and VLMT70 (1RMG + 2.40 ± 2.22; RPEG + 2.31 ± 2.27 mm).
Received: 27 December 2017
Between-group differences were non-significant (p > 0.05). Magnitude-based inferences
Accepted: 06 March 2018
Published: 21 March 2018 revealed 79, 57, and 72% chances of mean small effect size (ES) advantages for squat;
Citation: ES 90% confidence limits (CL) = 0.50 ± 0.63, bench press; ES 90% CL = 0.28 ± 0.73,
Helms ER, Byrnes RK, Cooke DM, and combined-total; ES 90% CL = 0.48 ± 0.68 respectively, in RPEG. There were 4, 14,
Haischer MH, Carzoli JP, Johnson TK,
Cross MR, Cronin JB, Storey AG and
and 6% chances 1RMG had a strength advantage of the same magnitude, and 18, 29,
Zourdos MC (2018) RPE vs. and 22% chances, respectively of trivial differences between groups.
Percentage 1RM Loading in
Periodized Programs Matched for Conclusions: Both loading-types are effective. However, RPE-based loading may
Sets and Repetitions. provide a small 1RM strength advantage in a majority of individuals.
Front. Physiol. 9:247.
doi: 10.3389/fphys.2018.00247 Keywords: perceived exertion, resistance training, strength, autoregulation, powerlifting

Frontiers in Physiology | www.frontiersin.org 1 March 2018 | Volume 9 | Article 247

Helms et al. RPE vs. Percentage 1RM Loading

INTRODUCTION that the RPE group would increase strength to a greater extent
than the percentage-based group due to load progression aligning
The principle of individualization is paramount to consider in the more closely to each individual participant’s capabilities (Klemp
design of resistance training protocols to optimize adaptations et al., 2016).
(Borresen and Ian Lambert, 2009; Kiely, 2012). Indeed, evidence
exists demonstrating that training adaptation is improved when MATERIALS AND METHODS
program design is tailored to the athlete (Beaven et al., 2008a,b;
Jones et al., 2016). One method of individualizing resistance Participants
training is “autoregulating” load prescription through the use of A total of 24 males began this study. Three participants dropped
a rating of perceived exertion (RPE) (Helms et al., 2016). out, two due to minor injury (joint pain or muscular discomfort)
Recently, an iteration of the traditional RPE scale based on and one due to a family emergency; therefore, 21 participants
“repetitions in reserve” (RIR) prior to muscular failure at the completed the protocol (Table 1). Inclusion criteria were as
end of a set, was introduced to the literature (Zourdos et al., follows: (1) minimum resistance training experience of 2 years
2016). The RIR-based RPE scale may have more utility compared while also performing the back squat and bench press a minimum
to traditional Borg RPE, which has yielded submaximal scores of once per week for the last 6 months; (2) a minimum 1RM
(6.8–9.0) even when an individual performs a set to volitional back squat and bench press of 1.5x and 1.25x body mass,
failure (Shimano et al., 2006; Pritchett et al., 2009; Hackett et al., respectively; and (3) be free from injury/illness that would
2012). Therefore, it has been recently suggested RIR-based RPE contraindicate participation. Resistance training history was
is superior to traditional RPE for assessing intensity during determined by completing a questionnaire previously used with
resistance training (Helms et al., 2016). Additionally, researchers similar populations (Klemp et al., 2016). All participants were
have reported males and females to determine RIR accurately informed of potential risks and signed an informed consent
(within ∼1 repetition) during the leg and chest press exercises document prior to participation. Ethics approval was granted by
when sets are performed within 0–3 repetitions from failure the University Institutional Review Board.
(Hackett et al., 2017). In further support of the RIR-based RPE
approach, scores have been strongly and inversely correlated with Experimental Design
velocity for both the squat (r = −0.87, p < 0.001) and bench The aim of this study was to compare strength and hypertrophy
press (r = −0.79, p < 0.001; Helms et al., 2017b), the implication adaptations in trained individuals following a daily undulating
being that as movement velocity decreases with higher intensities, periodization model, differentiated only by load prescription
reported RPE increases (RIR decreases). (RPE or percentage of 1RM). Groups were counterbalanced to
Despite recent research regarding RIR-based RPE and ensure minimal differences (mean 1RMs as similar as possible
the importance of individualization in resistance training with as high a p-value as possible when comparing means) in
prescription, training load is commonly prescribed as a absolute and relative 1RM strength as measured by the Wilks
percentage of pre-test one-repetition maximum (1RM) (Fleck score (a validated method of measuring relative strength in
and Kraemer, 2014). However, if atypical performance occurred
during the pre-test 1RM or if there were testing administration
errors, loading based on percentage 1RM could then lead to an TABLE 1 | Descriptive characteristics of participants.
inappropriate stimulus during training (Zourdos et al., 2016).
Furthermore, the number of repetitions which can be performed Variable 1RMG RPEG Combined
at the same percentage of 1RM can differ substantially between (n = 11 males) (n = 10 males) (n = 21 males)

athletes based on genetic differences and training background

Height (m) 1.75 ± 0.08 1.72 ± 0.06 1.74 ± 0.07
(Richens and Cleather, 2014). Thus, various issues exist when
Body Mass (kg) 80.2 ± 12.2 78.8 ± 9.72 79.5 ± 10.8
prescribing load solely with percentage of 1RM, whereas RPE
Body Fat (%) 10.8 ± 6.1 11.4 ± 5.1 11.1 ± 5.5
can account for individual differences in repetitions allowed
Age (yrs) 23.8 ± 4.2 20.9 ± 1.4 22.4 ± 3.4
and rate of adaptation. However, to our knowledge there is no
PMT (mm) 28.5 ± 6.4 30.6 ± 6.5 29.5 ± 6.4
study which has compared changes in strength and hypertrophy
VLMT50 (mm) 27.9 ± 3.6 27.3 ± 4.5 27.6 ± 4.0
over time between percentage based and RPE based training
VLMT70 (mm) 24.2 ± 3.1 23.8 ± 3.4 24.0 ± 3.2
programs.
Squat 1RM (kg) 139.2 ± 18.2 143.7 ± 24.9 141.3 ± 21.2
Therefore, the purpose of this study was to compare two
Bench Press 1RM (kg) 113.9 ± 18.7 120.9 ± 19.3 117.2 ± 18.8
resistance training protocols with matched repetitions, sets,
Squat Wilks Score 96.6 ± 15.0 99.4 ± 11.9 98.0 ± 13.3
exercises, and rest periods, but with differing methods of load
prescription; one group using percentage of pre-test 1RM and Bench Press Wilks 78.0 ± 7.9 83.8 ± 9.5 80.8 ± 8.9
Score
the other using the RIR-based RPE scale. We hypothesized the
method of load prescription would create minimal differences 1RMG, percentage 1RM load group; RPEG, RPE load group. Values are mean ± standard
in total volume (sets × repetitions × load) between groups and deviation. 1RM, one repetition maximum; PMT, pectoralis major muscle thickness;
VLMT50, vastus lateralis muscle thickness at 50% femur length; VLMT70, vastus lateralis
likewise, minimal differences in hypertrophy (Schoenfeld et al., muscle thickness at 50% femur length. Combined, mean values of all 21 participants. Wilks
2014). However, we hypothesized that intensity (both RPE and score (Vanderburgh and Batterham, 1999) is the scoring system used in powerlifting to
percentage of pre-test 1RM) would differ between groups, and determine strength relative to body mass.

Frontiers in Physiology | www.frontiersin.org 2 March 2018 | Volume 9 | Article 247

Helms et al. RPE vs. Percentage 1RM Loading

competitive powerlifting) (Vanderburgh and Batterham, 1999). were collected after the final warm-up and each 1RM attempt
Participants were assigned to either a percentage 1RM group (Zourdos et al., 2016). Previously researchers have identified
(1RMG, n = 11) with load assigned as percentages of pre-test that comparably trained lifters approached an ACV of ∼0.20
1RMs, or to an RPE group (RPEG, n = 10) with load selected m·s−1 and ∼0.15 m·s−1 on average for the squat and bench
by participants to reach target RPE ranges. press, respectively, at 1RM (Helms et al., 2017b; Ormsbee
A training duration of 8 weeks, while following a daily et al., 2017). Thus, the investigators made smaller increases
undulating periodized model, was selected as significant 1RM in load for 1RM attempts as velocity neared these thresholds.
and muscle thickness increases were recently reported in a study Additionally, during post-testing, the velocity at which pre-test
of this length on a similarly sized and trained population of males 1RMs were recorded was used to gauge when a participant
following similar progressions in volume and intensity (Klemp was approaching 1RM. Likewise, the proximity to this velocity
et al., 2016). Exercise selection, rest periods, and prescribed was used to aid 1RM post-test attempt selection. To provide a
set and repetitions were identical among groups. Both groups clear standard for the parameters of form, both exercises were
trained 3 times/week on non-consecutive days (i.e., Monday, performed in accordance with the standards of the International
Wednesday, and Friday) and performed the specified repetitions Powerlifting Federation (IPF, 2016) and a National Strength and
in a fixed, descending order each week. In a linear format, every Conditioning Association certified strength and conditioning
2 weeks (after the introductory week) the prescribed repetitions specialist with experience coaching powerlifters monitored all
decreased as load (either RPE or percentage of 1RM) increased testing and training sessions. Barbells and weight plates were
throughout. The final week consisted of a lowered volume taper calibrated (Eleiko Sport, Korsvägen, Halmstad, Sweden), and
leading into post-testing on the final day. The specific details of fractional plates (to the nearest 0.25 kg) were used to ensure
the programs’ structure are outlined in Table 2. loading precision in all testing sessions.

One-Repetition Maximum (1RM) Testing Training Protocol

Participants were shown the resistance-training specific RPE While the RPEG self-selected their loads to reach the target RPE
scale while receiving verbal instruction on how scores are range, both groups provided RPE scores after their final warm-
determined (Zourdos et al., 2016). Briefly, this 1–10 rating up set and all working sets to allow RPE comparisons between
scale is based on the subjective determination of RIR prior to groups throughout the study. Percentage 1RM assignments in
reaching failure such that a score of 10 indicates the maximal 1RMG were based upon recently published loading relationships
number of repetitions for that load were performed, 9.5 indicates in an attempt to ensure that, on average, the assigned percentages
no further RIR but the same number of repetitions could be in 1RMG would fall within the corresponding RPE ranges
performed with a slightly heavier load, 9 indicates one RIR, assigned to RPEG. As an example, in our previously published
8.5 indicates 1–2 RIR, 8 indicates 2 RIR, etc. Following this work trained lifters reported a 7 RPE when performing eight
explanation, participants performed a standardized, bodyweight, repetitions with 70% of 1RM (Helms et al., 2016). So, in week
dynamic warm-up and then according to previously validated 2 day 1, when 1RMG performed eight repetitions with 70% of
procedures (Zourdos et al., 2016), the investigators proceeded to 1RM, the RPEG performed eight repetitions with a load that
test the 1RM of their back squat, followed by their bench press. resulted in a 6–8 RPE (Table 2). Participants reported to the
To aid the researchers in attempt selection, average concentric laboratory to perform monitored resistance training for a total
velocity (ACV) using a Tendo Weightlifting Analyzer (TENDO of 25 days over 8 consecutive weeks. Each training session took
Sports Machines, Trencin, Slovak Republic), and RPE scores place at the same time each day to account for any diurnal

TABLE 2 | Summary of training plans.

Percentage 1RM group (1RMG) RPE group (RPEG)

Week Monday Wednesday Friday Monday Wednesday Friday

0 x x 1RM Testing x x 1RM Testing

1 2 × 8 × 65% 2 × 6 × 70% 2 × 4 × 75% 2 × 8 × 5–7 RPE 2 × 6 × 5–7 RPE 2 × 4 × 5–7 RPE
2 3 × 8 × 70% 3 × 6 × 75% 3 × 4 × 80% 3 × 8 × 6–8 RPE 3 × 6 × 6–8 RPE 3 × 4 × 6–8 RPE
3 3 × 8 × 72.5%* 3 × 6 × 77.5%* 3 × 4 × 82.5%* 3 × 8 × 6–8 RPE 3 × 6 × 6–8 RPE 3 × 4 × 6–8 RPE
4 3 × 7 × 75% 3 × 5 × 80% 3 × 3 × 85% 3 × 7 × 7–9 RPE 3 × 5 × 7–9 RPE 3 × 3 × 7–9 RPE
5 3 × 7 × 77.5%* 3 × 5 × 82.5%* 3 × 3 × 87.5%* 3 × 7 × 7–9 RPE 3 × 5 × 7–9 RPE 3 × 3 × 7–9 RPE
6 3 × 6 × 80% 3 × 4 × 85% 3 × 2 × 90% 3 × 6 × 8–10 RPE 3 × 4 × 8–10 RPE 3 × 2 × 8–10 RPE
7 3 × 6 × 82.5%* 3 × 4 × 87.5%* 3 × 2 × 92.5%* 3 × 6 × 8–10 RPE 3 × 4 × 8–10 RPE 3 × 2 × 8–10 RPE
8 2 × 4 × 80% 2 × 3 × 85% 1RM Testing 2 × 4 × 6–8 RPE 2 × 3 × 6–8 RPE 1RM Testing

1RMG uses percentages of pre-test 1RM to assign loads while RPEG uses RPE based on repetitions in reserve. Values are displayed as sets × repetitions × load. *If all repetitions
were completed with previous week’s assigned loads, load was increased as listed. If any repetitions are missed, load remained the same as prior week. 1RM, one repetition maximum;
RPE, rating of perceived exertion.

Frontiers in Physiology | www.frontiersin.org 3 March 2018 | Volume 9 | Article 247

Helms et al. RPE vs. Percentage 1RM Loading

changes in strength. Pre- and post-testing for anthropometric TABLE 3 | Example RPE load adjustments.
measurements, muscle thickness, and 1RM strength took place
Actual RPE Assigned RPE range 6–8
48–72 h before week 1 and at the end of week 8, respectively. After
pre-testing, participants returned to the lab 48–72 h later to begin 1 Increase load by 20%
a lower volume and load introductory microcycle during week 1 2 Increase load by 16%
(Table 2). The “main training program” occurred in weeks 2–7, 3 Increase load by 12%
then during week 8 participants completed taper sessions on the 4 Increase load by 8%
first 2 days of training, and post-testing occurred on the final day 5 Increase load by 4%
of the taper and test week. 6 Participant choice
In 1RMG load was assigned as a percentage of pre-test 7 Participant choice
1RM and progressed in a linear fashion throughout the study.
7.5 Participant choice
However, on weeks 3, 5, and 7, load was only increased by
8 Participant choice
2.5% of 1RM if all sets and repetitions were completed on the
8.5 Decrease load by 2%
same day from the prior week. If any repetitions were not able
9 Decrease load by 4%
to be completed from the prior week, load remained the same
9.5 Decrease load by 6%
(Table 2). Within-week, if a participant was unable to complete
10 Decrease load by 8%
repetitions, the load was reduced 4% for every repetition missed
on the subsequent set for the same exercise. During week 1 for RPE, rating of perceived exertion.
the RPEG, the researchers selected loads for the participants
to ensure the goal of the introductory week was accomplished
performed prior to the bench press and a 10 min rest period
(acclimating the participants to the frequency and total volume
occurred after concluding the squat prior to initiating the bench
of training) and to aid in familiarizing the participants with RPE-
press.
based load selection. The researchers explained their rationale
for load selection to the participants during week 1 to better
familiarize the participants for weeks 2–8 where they self-selected Dietary Logs, Protein, and Amino Acid
load. Investigators selected the loads during week 1 based on Provision
the combined factors of the percentage of 1RM they expected To encourage consistent energy and food intake throughout
to fall within the RPE range, the RPE of the last warm-up set training and testing, a 3 consecutive-day food log was completed
and visual assessment of barbell speed. Additionally, researchers during the first week of training and then again during the final
conservatively estimated loads to land at the lower end of the week. In the interim period and prior to the final week food
target RPE range to prevent cumulative fatigue from pushing the log, participants were instructed to continue their normal dietary
subsequent set above the RPE range. habits. To control for the potential impact of nutrient timing
In weeks 2–8, RPEG participants were shown the record between groups, participants ingested branched chain amino
of their performance on the same day of the previous week acids (Xtend, Scivation, Burlington, N.C., USA) containing 3.5 g
to assist them in daily load selection. In all weeks, when the of leucine approximately 20 min prior to each training and testing
reported RPE score for a completed set fell outside of the target session (upon arriving at the lab, then they began their dynamic
RPE range, an automatic adjustment to the load was made for warm up 10 min after) and 30 g of whey protein (Scivation
the subsequent set. Based on previous research (Helms et al., Whey, Scivation, Burlington, N.C., USA) immediately after each
2017a), for every 0.5 RPE above or below the upper or lower session. Both whey protein and branched-chain amino acids
RPE threshold, respectively, load was decreased or increased by were provided because of their ability to enhance muscle protein
2% in an attempt to bring the subsequent set’s RPE closer to synthesis (Tipton et al., 2001; Moore et al., 2009).
the assigned range. An example of how this load adjustment
protocol was implemented for an RPE range of 6–8 is displayed Muscle Thickness Testing
in Table 3. When the load fell within the assigned RPE range, Pectoralis major muscle thickness (PMT) and 50% (VLMT50)
the participant (or the researchers in the case of week 1) had and 70% vastus lateralis muscle thickness (VLMT70) were
the choice to modify load as desired so long as they believed it assessed via ultrasonography (Bodymetrix Pro System,
would still fall within the target RPE range. If a participant missed Intelemetrix Inc., Livermore, Calif., USA) prior to 1RM pre
assigned repetitions, for example completing 7 repetitions when and post-testing. This method of testing was previously used
8 were assigned, the set was considered a 10 RPE and each missed to assess the growth response to resistance training (Schoenfeld
repetition was considered a full RPE point for load-adjustment et al., 2014) and was validated with magnetic resonance imaging
purposes (i.e., if 5 repetitions at a 7–9 RPE was assigned, and 4 (Reeves et al., 2004). Scans were performed prior to 1RM
repetitions were completed, the load on the subsequent set would assessment on the right side of the body during pre- and
be reduced by 8%; 4% for being a full RPE point above the upper post-testing. Sites were scanned lateral to medial with the
threshold of the range and an additional 4% for being 1 repetition transducer perpendicular to the skin. Sites were scanned twice
short of the target). In both groups, 5–7 min rest periods were and an average of the two scans was recorded. However, if
administered between working sets and after the final warm the difference between the two scans was >2 mm, a third was
up set before the first working set. Additionally, the squat was performed and the two values within 2 mm were averaged. The

Frontiers in Physiology | www.frontiersin.org 4 March 2018 | Volume 9 | Article 247

Helms et al. RPE vs. Percentage 1RM Loading

site for the chest was designated as half the distance between of the 1RMG. Thus, data is presented as the probability of
the nipple and the anterior axillary line. Vastus lateralis scans an advantage of RPEG, 1RMG or a trivial difference between
were performed in the supine position. Sites were marked and groups.
measured at 50 and 70%, respectively, of the distance from Finally, differences between groups for the mean total across
the greater trochanter to the lateral epicondyle of the femur the 8-week study and at each time point (weeks 1–8) for the
(Abe et al., 1994, 1998). All scans were performed by the same average weekly RPE, relative volume load (sets × repetitions
investigator. × percentage 1RM), relative intensity per repetition (average
percentage 1RM per repetition for the week), change in PRS
Readiness Questionnaires and change in DALDA scores were determined by 2 tailed
Prior to beginning warm up sets, participants completed part A independent T-tests with an alpha of 0.05. Analyses were
and B of the daily analysis of life demands for athletes (DALDA) performed using a statistical software package (IBM SPSS
questionnaire and recorded a 1–10 perceived recovery status Statistics 21, SPSS Inc., Chicago, IL).
(PRS) score by hand. The DALDA is a two-part questionnaire
consisting of an A, B, or C Likert scale in which users record
whether they (A), feel worse than normal, (B), feel normal, or RESULTS
(C), feel better than normal. Part A consists of 9 broad categories Participant Adherence
in which stress can be assessed and part B consists of a list of
Participants were required to complete at least 90% of all sessions
25 questions pertaining to specific sources of stress (Rushall,
to be included (no more than two missed sessions and no missed
1990). The PRS scale is a simple 0–10 scoring system where the
sessions during the taper). The 1RMG as a whole completed
higher the score indicates greater recovery and the more likely the
98% of all sessions. The RPEG as a whole completed 97% of the
individual would expect improved performance (Laurent et al.,
squat portion and 96.5% of the bench portion of the sessions (in
2011).
one instance a participant did not bench press as a precaution
due to shoulder discomfort, which subsided by the next training
Statistical Analysis session).
To assess within group pre to post changes in muscle thickness
and strength, we performed independent paired T-tests set
at an alpha of 0.05. Despite relative homogeneity due to One-Repetition Maximum (1RM) Strength
counterbalancing, there was still some variation between groups and Muscle Thickness
in 1RM strength and muscle thickness. Thus, to analyse Both 1RMG and RPEG significantly increased back squat,
differences between groups we utilized analyses of covariance bench press, and combined 1RM strength relative to baseline
with pre-test scores as covariates. This is the preferred method (p < 0.001). Specifically, squat 1RM increased in 1RMG by
of analysis to account for the fact that participants with low pre- 13.9 ± 5.9 kg and in RPEG by 17.1 ± 5.4 kg while bench press
test scores generally improve more than those with high pre-test 1RM increased by 9.6 ± 5.4 kg and 10.7 ± 3.3 kg in 1RMG and
scores (Vickers and Altman, 2001). RPEG, respectively. Post-test 1RM back squat was 153.1 ± 16.6
To supplement null hypothesis testing, we calculated between and 160.7 ± 28.4 and post-test 1RM bench press was 113.9 ± 18.7
group effect size (ES) values such that each groups’ change score and 131.6 ± 19.5 in 1RMG and RPEG, respectively. Combined
(post-test–pre-test) was divided by the pooled standard deviation squat and bench press 1RM increased by 23.6 ± 10.4 kg in 1RMG
(SD) of both groups’ change scores (Morris and DeShon, 2002; and by 27.8 ± 7.9 kg in RPEG.
Page, 2014; Dankel et al., 2017). Thresholds for ES were based Additionally, muscle thickness significantly increased at
on Hopkins’ scale such that an ES of <0.20 was considered all measurement sites in both groups relative to baseline.
trivial, and threshold values of 0.20, 0.60, 1.20, and 2.00 were Specifically, PMT increased in 1RMG by 1.6 ± 1.3 mm
used to represent small (and the smallest worthwhile effect), (p < 0.001) and in RPEG by 1.9 ± 1.9 mm (p < 0.001). Post-
moderate, large, and very large effects (Batterham and Hopkins, test PMT was 30.1 ± 6.7 and 32.5 ± 6.8 in 1RMG and RPEG,
2006; Hopkins et al., 2009). Additionally, we calculated the 90% respectively. Likewise, VLMT50 increased by 2.1 ± 2.0 mm
confidence limits (CL) of each ES, using the small sample size bias (p = 0.004) and 1.9 ± 2.0 mm (p = 0.01) in 1RMG and
adjustment of the SD outlined by Becker (Becker, 1988; Morris, RPEG, respectively. Post-test VLMT50 was 30.0 ± 4.2 and
2008), to determine the probability that there was a positive 29.1 ± 4.9 in 1RMG and RPEG, respectively. Finally, VLMT70
(≥ 0.20 ES), trivial (0.19 to −0.19 ES), or negative (≤ −0.20 ES) increased in 1RMG by 2.4 ± 2.2 mm (p = 0.004) and
effect of the “intervention” (RPEG). Based on the same rationale in RPEG by 2.3 ± 2.3 mm (p = 0.02). Post-test VLMT70
for utilizing an analysis of covariance, we used the Hopkins was 26.6 ± 4.0 and 26.1 ± 2.7 in 1RMG and RPEG,
spreadsheet “analysis of a pre-post parallel-groups controlled respectively.
trial with adjustment for a predictor” (Hopkins, 2006) with the Overall, there were no significant differences observed
pre-test values as the covariate for the above calculations. For between groups for 1RM or muscle thickness. However, there
clarity of interpretation, rather than presenting the likelihood were small between group ES for squat, bench press, and
of a negative effect of the “intervention” (RPEG) relative to combined 1RM which all favored RPEG. Exact p values and
the “control” (1RMG) with negative ES values, we removed the ES 90% CL, along with probabilities of advantage or trivial
the sign and presented this as the probability of an advantage difference are displayed in Table 4.

Frontiers in Physiology | www.frontiersin.org 5 March 2018 | Volume 9 | Article 247

Helms et al. RPE vs. Percentage 1RM Loading

TABLE 4 | Strength and muscle thickness changes.

Variable P-value Size of effect Chance of RPE-loading Chance of trivial difference Chance of %1RM-loading
(mean ± 90% CL) advantage (≥0.20 ES) (%) (−0.19 to 0.19 ES) (%) advantage (≥0.20 ES) (%)

Squat 1RM 0.32 0.50 ± 0.63 79 18 4

Bench 1RM 0.52 0.28 ± 0.73 57 29 14
Combined 1RM 0.38 0.48 ± 0.68 72 22 6
PMT 0.66 0.15 ± 0.79 46 32 22
VLMT50 0.76 −0.13 ± 0.76 23 33 44
VLMT70 0.79 −0.06 ± 0.68 25 38 37

Between group differences in strength and muscle thickness. CL, confidence limit; RPE, rating of perceived exertion; ES, effect size; 1RM, one repetition maximum; PMT, pectoralis
major muscle thickness; VLMT50, vastus lateralis muscle thickness at 50% femur length; VLMT70, vastus lateralis muscle thickness at 50% femur length.

Training RPE, Volume, and Intensity approached significance (p = 0.08). Likewise, the change from
For the squat, RPE was significantly higher in RPEG vs. 1RMG in week 7 to 8 in RPEG (1.1 ± 1.1) vs. 1RMG (0.3 ± 1.0) also
weeks 4, 6, 7, and 8, and the difference approached significance approached significance (p = 0.09).
(p = 0.09) in week 5. Likewise, RPE was higher for the bench press
in RPEG during weeks 2–8 compared to 1RMG. Figure 1 displays
the weekly average RPE scores for both groups, for both lifts, DISCUSSION
throughout the study. Average squat RPE for the entire 8-week
period also significantly differed (p = 0.04) with higher values The goal of this study was to compare two resistance training
in RPEG (7.2 ± 0.3) compared to 1RMG (6.5 ± 1.0). Likewise, protocols differentiated only by loading strategy to determine if
average bench press RPE for the 8-week period was significantly they would produce different muscle thickness, psychometric,
(p < 0.001) higher in RPEG (7.3 ± 0.3) compared to 1RMG and performance outcomes. Our first hypothesis that greater
(5.8 ± 1.0). strength gains would be achieved by individualizing load
Similarly, weekly average relative intensity per repetition assignment via RPE was partially supported. Null hypothesis
(defined as the load used in training divided by the pre-test testing did not reveal a significant difference between groups.
1RM) diverged with significantly higher values in RPEG at However, small (0.28–0.50) between group ES differences were
weeks 6–8 and 2–8 in the squat and bench press respectively, found with probabilities favoring RPEG. Our second hypothesis,
compared to 1RMG. As 1RMG had a pre-planned load, changes that muscle thickness changes would be similar between groups
in relative intensity per repetition illustrate how RPEG increased was supported as there were no significant differences between
training loads throughout the study, comparatively. Lastly, 1RMG and RPEG for any muscle thickness measurement.
relative volume load differed significantly between groups with Furthermore, between group ESs were trivial and probabilities
RPEG performing more volume than 1RMG at weeks 7 and 8 and were unclear.
weeks 3 and 8 for the squat and bench press, respectively. The Since the recent introduction of the RIR-based RPE scale to
relative intensity per repetition and relative volume load values the literature (Zourdos et al., 2016), researchers have postulated
for both groups, for both lifts, throughout the study are displayed that greater performance could be achieved by using the scale
in Figure 2. to “autoregulate” load (Helms et al., 2016, 2017b; Zourdos
Average relative intensity for the entire 8-week period was et al., 2016; Ormsbee et al., 2017). To our knowledge, this
not significantly different between 1RMG (78.73 ± 0.20%) is the first study that has addressed and provided initial
and RPEG (79.73 ± 4.44%) for squat (p = 0.49). Likewise, support for this claim. With that said, strength differences
average relative volume load for the entire 8-week period was not between groups were small and variable enough to fall short
significantly different between 1RMG (10.49 ± 0.21) and RPEG of statistical significance. This may indicate that while some
(10.39 ± 0.67) for squat (p = 0.66). However, average relative individuals could benefit from using RPE as a loading strategy,
intensity for the entire 8-week period was significantly greater in for others, the choice between using percentage 1RM- or RPE-
RPEG (84.14 ± 2.02%) compared to 1RMG (78.70 ± 0.18%) for based loading is inconsequential (at least in the short term).
bench press (p < 0.001). Additionally, average relative volume However, at the group level the RPEG trained at a higher
load for the entire 8-week period was also significantly greater average RPE than 1RMG. Specifically, RPE diverged at week
in RPEG (10.84 ± 0.41) compared to 1RMG (10.49 ± 0.21) for 4 for squat and week 2 for bench press and then remained
bench press (p = 0.03). higher in RPEG throughout the rest of the study. Interestingly,
significant increases in strength and hypertrophy occurred in
Perceived Readiness both groups, despite the majority of training occurring ∼3–
Week to week changes in DALDA part A, part B and PRS scores 4 repetitions from failure (RPE ∼6–7). This provides further
were not significantly different between groups at any time point evidence that training to failure at all times is not necessary to
(data not shown). However, the change in average PRS score make significant gains in hypertrophy (Sampson and Groeller,
from week 6 to 7 in RPEG (−0.6 ± 0.5) vs. 1RMG (−0.1 ± 0.8) 2016) or strength (Izquierdo et al., 2006; Davies et al., 2016), at

Frontiers in Physiology | www.frontiersin.org 6 March 2018 | Volume 9 | Article 247

Helms et al. RPE vs. Percentage 1RM Loading

FIGURE 1 | Weekly average RPE values for (A) for squat and (B) bench press. *p< 0.05, **p < 0.01, and ***p < 0.001. RPE, rating of perceived exertion.

least when training with moderate to heavy loads (Ogasawara of 1RM were too low or the progression rate was too slow
et al., 2013). in 1RMG, whereas participants in RPEG were able to progress
Mirroring this divergence in RPE, relative intensity per at an individualized rate. While we made an effort to assign
repetition was also higher beginning at week 6 for squats and percentages of 1RM which should yield similar RPE to the
week 2 for bench press in RPEG compared to 1RMG. Thus, range prescribed in RPEG (Helms et al., 2016), greater total
for a large part of the study RPEG trained at a higher RPE and volume (p = 0.03) at a higher average intensity (p < 0.001)
percentage of pre-test 1RM than 1RMG, which may explain the was performed by RPEG for the bench press. However, the
higher probability of enhanced 1RM in RPEG. Differences in relevance of this difference is questionable, as there were not
relative volume load were not expected since we matched sets significant differences between groups for the squat in total
and repetitions. Nonetheless, likely due to the higher relative volume (p = 0.49) or average intensity per repetition (p = 0.66),
intensity (as relative volume is sets × repetitions × percentage yet the squat had the highest probability of greater strength gain
1RM), RPEG performed more bench press volume overall and due to RPE-based loading. Alternatively, if this is a limitation
more volume at two time points for the squat (weeks 7 and 8) and of the study, it might also be a limitation of percentage 1RM-
bench press (weeks 3 and 8). Related to the volume performed, based loading in general, as the number of “repetitions allowed”
our second hypothesis that muscle thickness changes would be at a given percentage of 1RM and rates of adaptation differ
similar between groups, was supported. As stated, while there substantially between individuals (Richens and Cleather, 2014).
were some differences in volume performed between groups, it In summary, both 1RMG and RPEG increased 1RM squat
was not substantial enough to generate greater hypertrophy in and bench press (p < 0.001) along with both upper and lower
the short-term. body muscle thickness (p < 0.05) over the course of 8 weeks.
Interestingly, the PRS changes between groups approached Although no statistically significant differences between groups
significance (p = 0.08–0.09) at weeks 7 and 8. The RPEG had a existed, there were small between-group ESs in favor of RPEG for
larger decrease in PRS from week 6 to 7 and then a larger increase 1RM squat (0.50) and bench press (0.28), which when analyzed
in PRS from week 7 to 8, compared to 1RMG. This might indicate probabilistically, translated to 79 and 57% greater odds for
that in the final week prior to the taper where load was the highest strength gain in favor of RPEG, respectively. Moreover, there
(week 7), RPEG overreached to a greater extent than 1RMG and were various points throughout the study where average RPE
that the taper was more effective for RPEG, as their PRS score per set, relative volume and relative intensity per repetition were
rebounded to a greater degree during week 8. This PRS score higher in RPEG vs. 1RMG, possibly explaining the likelihood of
pattern provides some insight into how RPE-based loading may a small advantage in favor of RPEG for strength improvement.
help to ensure the temporal goals of a mesocycle are adhered to. Practically speaking, although RPEG may have provided a
On the other hand, changes in DALDA scores between groups slight benefit in the present study for strength, this does not mean
were non-significant at all time points. However, based on our that RPE and percentage of 1RM should be seen as mutually
anecdotal observation of the participants, as time went on the exclusive for load prescription. For example, RPE accuracy may
DALDA forms were completed more quickly, with less effort vary by individual; thus, a lifter who is inaccurate with RPE may
and with less attention to detail. This might highlight a potential not be advised to use solely RPE for load prescription. In this
advantage of the PRS compared to the DALDA, in that it takes situation, a conservative percentage of 1RM can be assigned for
less effort and time to evaluate readiness using a singular 1–10 a set number of repetitions for the initial set. However, a “goal”
scale compared to a 34 item, 3 point Likert-scale questionnaire. RPE range could also be established (i.e., 4 sets of 8 at 70% of
A limitation of this study is that strength improvement may 1RM with goal RPE of 6–8), and the individual could adjust the
have been greater in RPEG because the prescribed percentages subsequent sets if the first set RPE is out of the goal range. The

Frontiers in Physiology | www.frontiersin.org 7 March 2018 | Volume 9 | Article 247

Helms et al. RPE vs. Percentage 1RM Loading

FIGURE 2 | Weekly average values for (A) intensity relative to pre-test 1RM per repetition for squat, (B) bench press and (C) volume load relative to pre-test 1RM for
squat and (D) bench press. *p < 0.05, **p < 0.01, and ***p < 0.001. 1RM, one repetition maximum.

proposed strategy could also be used in a sports team setting accordance with the Declaration of Helsinki. The protocol was
where athletes with different training backgrounds and muscle approved by the The Florida Atlantic University Institutional
characteristics may perform substantially different repetitions at Review Board.
the same percentage of 1RM (Richens and Cleather, 2014); thus,
athletes could use the goal RPE range to adjust load accordingly. AUTHOR CONTRIBUTIONS
Furthermore, the strategy of using percentage 1RM and RPE
in conjunction also accounts for daily readiness with a baseline EH with the assistance of JBC, AS, and MZ developed the
structure, in that the individual has a pre-determined load, yet concept and overall design of the study. EH, RB, DC, MH, JPC,
can adjust in accordance with the goal RPE if recovery between TJ, and MZ developed the specific data collection procedures,
sessions was inadequate. recruited the participants, and collected the data. RB conducted
For future research, we recommend that inter-individual all ultrasound scans and conducted the raw ultrasound data
differences be explored. It has already been established that analysis. RB, DC, and EH coordinated data collection sessions,
training age may impact the ability to accurately rate RPE times and schedules and contacted participants throughout the
(Zourdos et al., 2016). However, other characteristics such as study. EH and MC performed the statistical analysis of the data.
temperament or social attitudes toward resistance training may EH created the tables and figures and wrote the initial draft of the
influence RPE ratings and therefore could be used to predict manuscript with the assistance of MC on the statistical analysis
which individuals might respond better to an RPE-based loading section. All authors revised and edited the manuscript.
strategy.
ACKNOWLEDGMENTS
ETHICS STATEMENT
This article reuses parts of EH thesis using the Repetitions in
This study was carried out in accordance with the Reserve-based Rating of Perceived Exertion Scale to Autoregulate
recommendations of The Florida Atlantic University Powerlifting Training, available at https://aut.researchgateway.ac.
Institutional Review Board with written informed consent nz/handle/10292/10829. We would like to thank the lab assistants
from all subjects. All subjects gave written informed consent in and the research participants who made this study possible.

Frontiers in Physiology | www.frontiersin.org 8 March 2018 | Volume 9 | Article 247

Helms et al. RPE vs. Percentage 1RM Loading

REFERENCES Jones, N., Kiely, J., Suraci, B., Collins, D. J., de Lorenzo, D., Pickering, C., et al.
(2016). A genetic-based algorithm for personalized resistance training. Biol.
Abe, T., Brechue, W. F., Fujita, S., and Brown, J. B. (1998). Gender Sport 33, 117–126. doi: 10.5604/20831862.1198210
differences in FFM accumulation and architectural characteristics of muscle. Kiely, J. (2012). Periodization paradigms in the 21st century: evidence-
Med. Sci. Sports Exerc. 30, 1066–1070. doi: 10.1097/00005768-199807000- led or tradition-driven? Int. J. Sports Physiol. Perform. 7, 242–250.
00007 doi: 10.1123/ijspp.7.3.242
Abe, T., Kondo, M., Kawakami, Y., and Fukunaga, T. (1994). Prediction equations Klemp, A., Dolan, C., Quiles, J. M., Blanco, R., Zoeller, R. F., Graves,
for body composition of Japanese adults by B-mode ultrasound. Am. J. Hum. B. S., et al. (2016). Volume-equated high- and low-repetition daily
Biol. 6, 161–170. doi: 10.1002/ajhb.1310060204 undulating programming strategies produce similar hypertrophy and strength
Batterham, A. M., and Hopkins, W. G. (2006). Making meaningful adaptations. Appl. Physiol. Nutr. Metab. 41, 699–705. doi: 10.1139/apnm-2015-
inferences about magnitudes. Int. J. Sports Physiol. Perform. 1, 50–57. 0707
doi: 10.1123/ijspp.1.1.50 Laurent, C. M., Green, J. M., Bishop, P. A., Sjökvist, J., Schumacker,
Beaven, C. M., Cook, C. J., and Gill, N. D. (2008a). Significant strength gains R. E., Richardson, M. T., et al. (2011). A practical approach to
observed in rugby players after specific resistance exercise protocols based on monitoring recovery: development of a perceived recovery status
individual salivary testosterone responses. J. Strength Cond. Res. 22, 419–425. scale. J. Strength Cond. Res. 25, 620–628. doi: 10.1519/JSC.0b013e3181c
doi: 10.1519/JSC.0b013e31816357d4 69ec6
Beaven, C. M., Gill, N. D., and Cook, C. J. (2008b). Salivary testosterone Moore, D. R., Robinson, M. J., Fry, J. L., Tang, J. E., Glover, E. I., Wilkinson, S.
and cortisol responses in professional rugby players after four B., et al. (2009). Ingested protein dose response of muscle and albumin protein
resistance exercise protocols. J. Strength Cond. Res. 22, 426–432. synthesis after resistance exercise in young men. Am. J. Clin. Nutr. 89, 161–168.
doi: 10.1519/JSC.0b013e3181635843 doi: 10.3945/ajcn.2008.26401
Becker, B. J. (1988). Synthesizing standardized mean-change measures. Morris, S. B. (2008). Estimating effect sizes from pretest-posttest-control
Br. J. Math. Stat. Psychol. 41, 257–278. doi: 10.1111/j.2044-8317.1988. group designs. Organ Res. Meth. 11, 364–386. doi: 10.1177/10944281062
tb00901.x 91059
Borresen, J., and Lambert, M., I. (2009). The quantification of training load, the Morris, S. B., and DeShon, R. P. (2002). Combining effect size estimates in meta-
training response and the effect on performance. Sports Med. 39, 779–795. analysis with repeated measures and independent-groups designs. Psychol.
doi: 10.2165/11317780-000000000-00000 Methods 7, 105–125. doi: 10.1037/1082-989X.7.1.105
Dankel, S. J., Mouser, J. G., Mattocks, K. T., Counts, B. R., Jessee, M. B., Buckner, Ogasawara, R., Loenneke, J. P., Thiebaud, R. S., and Abe, T. (2013). Low-load
S. L., et al. (2017). The widespread misuse of effect sizes. J. Sci. Med. Sport 20, bench press training to fatigue results in muscle hypertrophy similar to high-
446–450. doi: 10.1016/j.jsams.2016.10.003 load bench press training. Int. J. Clin. Med. 4, 114–121. doi: 10.4236/ijcm.2013.
Davies, T., Orr, R., Halaki, M., and Hackett, D. (2016). Erratum to: effect 42022
of training leading to repetition failure on muscular strength: a systematic Ormsbee, M. J., Carzoli, J. P., Klemp, A., Allman, B. R., Zourdos, M. C., Kim, J.
review and meta-analysis. Sports Med. 46, 605–610. doi: 10.1007/s40279-016- S., et al. (2017). Efficacy of the repetitions in reserve-based rating of perceived
0509-x exertion for the bench press in experienced and novice benchers. J. Strength
Fleck, S. J., and Kraemer, W. (2014). Designing Resistance Training Programs, 4Edn. Cond. Res. doi: 10.1519/JSC.0000000000001901. [Epub ahead of print].
Champaign, IL: Human Kinetics. Page, P. (2014). Beyond statistical significance: clinical interpretation of
Hackett, D. A., Cobley, S., Davies, T., Michael, S., and Halaki, M. (2017). Accuracy rehabilitation research literature. Int. J. Sports Phys. Ther. 9, 726–736.
in estimating repetitions to failure during resistance exercise. J. Strength Cond. Pritchett, R. C., Green, J. M., Wickwire, P. J., and Kovacs, M. (2009).
Res. 38, 2162–2168. doi: 10.1519/JSC.0000000000001683 Acute and session RPE responses during resistance training: Bouts to
Hackett, D. A., Johnson, N. A., Halaki, M., and Chow, C. M. (2012). A failure at 60% and 90% of 1RM. South Afr. J. Sports Med. 21, 23–26.
novel scale to assess resistance-exercise effort. J. Sports Sci. 30, 1405–1413. doi: 10.17159/2078-516X/2009/v21i1a304
doi: 10.1080/02640414.2012.710757 Reeves, N. D., Maganaris, C. N., and Narici, M. V. (2004). Ultrasonographic
Helms, E. R., Cronin, J., Storey, A., and Zourdos, M. C. (2016). Application assessment of human skeletal muscle size. Eur. J. Appl. Physiol. 91, 116–118.
of the Repetitions in reserve-based rating of perceived exertion scale for doi: 10.1007/s00421-003-0961-9
resistance training. Strength Cond. J. 38, 42–49. doi: 10.1519/SSC.00000000000 Richens, B., and Cleather, D. J. (2014). The relationship between the number
00218 of repetitions performed at given intensities is different in endurance and
Helms, E. R., Cross, M. R., Brown, S. R., Storey, A., Cronin, J., and strength trained athletes. Biol. Sport 31, 157–161. doi: 10.5604/20831862.10
Zourdos, M. C. (2017a). Rating of perceived exertion as a method of 99047
volume autoregulation within a periodized program. J. Strength Cond. Res. Rushall, B. S. (1990). A tool for measuring stress tolerance in elite athletes. J. Appl.
doi: 10.1519/JSC.0000000000002032. [Epub ahead of print]. Sport Psychol. 2, 51–66. doi: 10.1080/10413209008406420
Helms, E. R., Storey, A., Cross, M. R., Brown, S. R., Lenetsky, S., Ramsay, Sampson, J. A., and Groeller, H. (2016). Is repetition failure critical for the
H., et al. (2017b). RPE and velocity relationships for the back squat, bench development of muscle hypertrophy and strength? Scand. J. Med. Sci. Sports
press, and deadlift in powerlifters. J. Strength Cond. Res. 31, 292–297. 26, 375–383. doi: 10.1111/sms.12445
doi: 10.1519/JSC.0000000000001517 Schoenfeld, B. J., Ratamess, N. A., Peterson, M. D., Contreras, B., Tiryaki-
Hopkins, W. G. (2006). Spreadsheets for analysis of controlled trials, with Sonmez, G., and Alvar, B. A. (2014). Effects of different volume-equated
adjustment for a subject characteristic. Sportscience 10, 46–50. Available online resistance training loading strategies on muscular adaptations in well-trained
at: http://sportsci.org/2006/wghcontrial.htm men. J. Strength Cond. Res. 29, 2909–2918. doi: 10.1519/JSC.00000000000
Hopkins, W. G., Marshall, S. W., Batterham, A. M., and Hanin, J. (2009). 00480
Progressive statistics for studies in sports medicine and exercise Shimano, T., Kraemer, W. J., Spiering, B. A., Volek, J. S., Hatfield, D. L., Silvestre,
science. Med. Sci. Sports Exerc. 41, 3–13. doi: 10.1249/MSS.0b013e31818 R., et al. (2006). Relationship between the number of repetitions and selected
cb278 percentages of one repetition maximum in free weight exercises in trained
IPF (2016). International Powerlifting Federation Technical Rules Book. Available and untrained men. J. Strength Cond. Res. 20, 819–823. doi: 10.1519/R-
online at: http://www.powerlifting-ipf.com/rules/technical-rules.html. 18195.1
Izquierdo, M., Ibanez, J., Gonzalez-Badillo, J. J., Hakkinen, K., Ratamess, Tipton, K. D., Rasmussen, B. B., Miller, S. L., Wolf, S. E., Owens-Stovall,
N. A., Kraemer, W. J., et al. (2006). Differential effects of strength S. K., Petrini, B. E., et al. (2001). Timing of amino acid-carbohydrate
training leading to failure versus not to failure on hormonal responses, ingestion alters anabolic response of muscle to resistance exercise. Am. J.
strength, and muscle power gains. J. Appl. Physiol. (1985) 100, 1647–1656. Physiol. Endocrinol. Metab. 281, E197–E206. doi: 10.1152/ajpendo.2001.281.2.
doi: 10.1152/japplphysiol.01400.2005 E197

Frontiers in Physiology | www.frontiersin.org 9 March 2018 | Volume 9 | Article 247

Helms et al. RPE vs. Percentage 1RM Loading

Vanderburgh, P. M., and Batterham, A. M. (1999). Validation of the Conflict of Interest Statement: The authors declare that the research was
Wilks powerlifting formula. Med. Sci. Sports Exerc. 31, 1869–1875. conducted in the absence of any commercial or financial relationships that could
doi: 10.1097/00005768-199912000-00027 be construed as a potential conflict of interest.
Vickers, A. J., and Altman, D. G. (2001). Statistics notes: analysing controlled
trials with baseline and follow up measurements. BMJ 323, 1123–1124. Copyright © 2018 Helms, Byrnes, Cooke, Haischer, Carzoli, Johnson, Cross, Cronin,
doi: 10.1136/bmj.323.7321.1123 Storey and Zourdos. This is an open-access article distributed under the terms
Zourdos, M. C., Klemp, A., Dolan, C., Quiles, J. M., Schau, K. of the Creative Commons Attribution License (CC BY). The use, distribution or
A., Jo, E., et al. (2016). Novel resistance training-specific rating reproduction in other forums is permitted, provided the original author(s) and the
of perceived exertion scale measuring repetitions in reserve. J. copyright owner are credited and that the original publication in this journal is cited,
Strength Cond. Res. 30, 267–275. doi: 10.1519/JSC.00000000000 in accordance with accepted academic practice. No use, distribution or reproduction
01049 is permitted which does not comply with these terms.

Frontiers in Physiology | www.frontiersin.org 10 March 2018 | Volume 9 | Article 247