Content uploaded by Andrew Harrison
Author content
All content in this area was uploaded by Andrew Harrison on Apr 16, 2018
Content may be subject to copyright.
1 23
Sports Medicine
ISSN 0112-1642
Sports Med
DOI 10.1007/s40279-017-0730-2
The Impact of Resistance Training on
Swimming Performance: A Systematic
Review
Emmet Crowley, Andrew J.Harrison &
Mark Lyons
1 23
Your article is protected by copyright and
all rights are held exclusively by Springer
International Publishing Switzerland. This e-
offprint is for personal use only and shall not
be self-archived in electronic repositories. If
you wish to self-archive your article, please
use the accepted manuscript version for
posting on your own website. You may
further deposit the accepted manuscript
version in any repository, provided it is only
made publicly available 12 months after
official publication or later and provided
acknowledgement is given to the original
source of publication and a link is inserted
to the published article on Springer's
website. The link must be accompanied by
the following text: "The final publication is
available at link.springer.com”.
SYSTEMATIC REVIEW
The Impact of Resistance Training on Swimming Performance:
A Systematic Review
Emmet Crowley
1
•Andrew J. Harrison
1
•Mark Lyons
1
ÓSpringer International Publishing Switzerland 2017
Abstract
Background The majority of propulsive forces in swim-
ming are produced from the upper body, with strong cor-
relations between upper body strength and sprint
performance. There are significant gaps in the literature
relating to the impact of resistance training on swimming
performance, specifically the transfer to swimming
performance.
Objective The aims of this systematic literature review are
to (1) explore the transfer of resistance-training modalities
to swimming performance, and (2) examine the effects of
resistance training on technical aspects of swimming.
Methods Four online databases were searched with the
following inclusion criteria: (1) journal articles with out-
come measures related to swimming performance, and (2)
competitive swimmers participating in a structured resis-
tance-training programme. Exclusion criteria were (1)
participants with a mean age \16 years; (2) untrained,
novice, masters and paraplegic swimmers; (3) triathletes
and waterpolo players; (4) swimmers with injuries or ill-
ness; and (5) studies of starts and turns specifically. Data
were extracted using the Preferred Reporting Items for
Systematic Reviews and Meta-Analyses (PRISMA)
guidelines, and the Physiotherapy Evidence Database
(PEDro) scale was applied.
Results For optimal transfer, specific, low-volume, high-
velocity/force resistance-training programmes are optimal.
Stroke length is best achieved through resistance training
with low repetitions at a high velocity/force. Resisted
swims are the most appropriate training modality for
improving stroke rate.
Conclusion Future research is needed with respect to the
effects of long-term resistance-training interventions on
both technical parameters of swimming and overall
swimming performance. The results of such work will be
highly informative for the scientific community, coaches
and athletes.
Key Points
The current literature shows that resistance training
can improve swimming performance in response to
various types of resistance-training programmes.
For optimal transfer, low-volume, high-
velocity/force resistance-training programmes are
recommended.
Trends identified in the literature suggest that for
improving stroke length, low repetitions at high
intensities are needed and resisted swims could
present a viable training modality for increasing
stroke rate.
&Emmet Crowley
emmet.crowley@ul.ie
Andrew J. Harrison
drew.harrison@ul.ie
Mark Lyons
mark.lyons@ul.ie
1
Biomechanics Research Unit, Department of Physical
Education and Sport Sciences, University of Limerick,
Limerick, Ireland
123
Sports Med
DOI 10.1007/s40279-017-0730-2
Author's personal copy
1 Introduction
Swimming performance is influenced by a complex inter-
action of physiological, morphological, neuromuscular,
biomechanical and technical factors. More specifically,
swimming velocity is a product of stroke rate and stroke
length [1], and increasing stroke rate [2] or stroke length
[3] independently has been found to improve swimming
performance. It has been suggested that improvement in
swimming velocity requires programmes including ele-
ments of high frequency, duration and intensity, resulting
in high overall training volumes [4]. It is not well-estab-
lished in the existing literature whether resistance training
improves stroke rate, stroke length and overall swimming
performance.
Resistance training can be defined as the ability of a
given muscle or group of muscles to generate muscular
force under specific conditions. The benefits of resistance
training in swimmers have been previously reviewed [5,6]
and have been questioned by coaches due to concerns
about increases in muscle mass (hypertrophy) or decreases
in flexibility which could increase drag forces and nega-
tively affect swimming performance. Despite these con-
cerns, resistance-training programmes are commonly
implemented by swimmers [7,8]. Resistance-training
modalities are intended to overload the muscles used in
swimming and increase maximal power output. The pro-
posed physiological benefits resulting from resistance
training are many, but notable benefits include increases in
phosphagen stores, contractile proteins, anaerobic power
output, muscle architecture, fibre pennation, protein syn-
thesis, tissue remodelling and hypertrophy of fast-twitch
muscle fibres [9–12]. Furthermore, resistance training has
been found to increase maximum strength and therefore
increase the rate of force development [13]. It is widely
accepted that swimming performance is highly dependent
on power and muscular strength [14–18], with the latter
identified as a major determinant of success in competitive
swimming [19]. Specifically, upper body strength is
imperative in swimming as the majority of propulsive
forces [20–22] and swimming velocity [22–25] are gener-
ated by the upper body musculature. Previous research [20]
has also found strong correlations (r=0.93) between
upper body strength and swimming performance.
These points aside, there are key gaps in the scientific
literature on swimming relating to the effectiveness of
resistance-training programmes in improving muscular
strength, and, more significantly, the resulting effects on
key technical parameters of swimming performance. Fur-
thermore, it is not understood whether or how these effects
transfer to improved swimming velocity and performance.
Consequently, the aims of this systematic literature review
are (1) to explore the transfer of resistance-training
modalities to swimming performance in competitive
swimmers, and (2) to examine the effects of resistance
training on key technical parameters of swimming
performance.
2 Methods
2.1 Research Approach
The MEDLINE, PubMed, SPORTDiscus and Web of Sci-
ence online databases were searched using specific search
terms (swimming) AND (strength, resistance, power, cross-
training) AND (humans) NOT (waterpolo, triathlon) for the
period from January 1988 up to and including January 2016.
Search terms were modified accordingly to fit the require-
ments or nuances of the databases used.
2.2 Study Criteria
Studies were included in this systematic review if they met
the following criteria: (1) journal articles with outcome
measures related to swimming performance; (2) competi-
tive-, regional-, national- or elite-level swimmers engaged
in a structured resistance and swimming training pro-
gramme; and (3) resistance-training modalities using tra-
ditional approaches as well as other novel approaches
where resistance is applied to the body. Exclusion criteria
were (1) subjects with a mean age\16 years; (2) untrained,
novice, masters and paraplegic swimmers; (3) triathletes
and waterpolo players; (4) swimmers with injuries or ill-
nesses; and (5) studies focusing on starts and turns.
2.3 Study Selection
Data were extracted using the Preferred Reporting Items
for Systematic Reviews and Meta-Analyses (PRISMA)
guidelines [26]. The search process (illustrated in Fig. 1)
included a hierarchy of assessment whereby studies were
first assessed by journal title (and duplicates removed),
second by abstract, and third by full article review, when
the journal article was either included or excluded based on
the inclusion and exclusion criteria.
2.4 The Physiotherapy Evidence Database (PEDro)
The Physiotherapy Evidence Database (PEDro) scale [27]
was used to rate the quality of the literature. The PEDro
scale consists of 11 items related to scientific rigor,
including eligibility criteria, random allocation strategy,
concealed allocation, follow-up comparison, baseline
E. Crowley et al.
123
Author's personal copy
comparison, blinding of subjects, therapists and assessors,
intention to treat, between-group analysis, and both point
and variability measures. The scale gives 1 point if the
study satisfies the criteria and 0 points if not. Studies
scoring 9–10 on the PEDro scale are considered to be
methodologically excellent, those scoring 6–8 are consid-
ered good, those scoring 4–5 are considered fair, and those
scoring \4 are considered methodologically poor. The
principal investigator scored all studies according to this
scale.
3 Results
From the 14 papers identified (see Fig. 1), 10 examined dry-
land resistance training and four examined swim-specific
resistance-training methods. The dry-land resistance group
was divided into three subgroups: (1) biokinetic swim bench
[28]; (2) traditional weight training [7,16,18,19,29–32];
and (3) core training [33]. The swim-specific resistance
group included resistive-band training [2], arms-only train-
ing [34], drag-suit training [35] and the measurement of
active drag (MAD) device system [36].
Table 1shows the weekly training programme, com-
petition level of swimmers, number of subjects, sex, age
and anthropometric characteristics, PEDro scale scores,
control and resistance-training sample size, and strength
and swimming performance tests. PEDro scores for all
studies reviewed ranged from 4 [16,29,31]to7[2,33,34].
Notably, 50% of the studies did not randomly allocate
subjects to groups [7,16,19,29,31,33,36]. Other studies
[16,29,30] did not assign a control group. Table 2shows
the specific resistance-training programmes, and baseline
and post-intervention results with respect to the relevant
strength tests, stroke length, stroke rate and swimming
performance.
The implementation of various resistance-training
interventions showed improvements in stroke length,
stroke rate and swimming performance. A significant
(p\0.05) increase in stroke length and swimming per-
formance was observed within the traditional resistance-
training groups [31,32], and a significant (p\0.05)
increase in stroke rate and swimming performance was also
observed within the traditional resistance-training groups
[7,31]. Girold et al. [2] found a significant (p\0.05)
increase in stroke rate and 100 m swimming performance,
Fig. 1 Schematic
representation of the data
extraction protocol. The
PRISMA flowchart was used to
illustrate the inclusion and
exclusion criteria used in this
systematic review. PRISMA
Preferred Reporting Items for
Systematic Reviews and Meta-
Analyses
Impact of Resistance Training on Swimming Performance
123
Author's personal copy
Table 1 Summary of studies, subject background, PEDro scale and tests used to examine the effect of resistance training on swimming performance
References Training programme Level N,
sex
Age and anthropometrics
(mean ±SD)
a
PEDro
scale
RT (N) Control
(N)
Strength test
a
Swim test
a
Biokinetic swim-bench training
Roberts
et al. [28]
10 weeks
6/week swim
3/week RT
Collegiate 16, M 19.1 ±2.1 years; 1.83 ±0.76 m;
75.67 ±10.06 kg
5 8 8 Biokinetic swim-
bench test
Isokinetic
dynamometer
91.44 m fc swim
(mock meet)
Traditional resistance training
Girold et al. [32] 4 weeks
10/week swim (2 h)
3/week RT
National 12, M
12, F
21.8 ±3.9 years; 1.74 ±0.08 m;
66 ±9kg
6 8 8 Isokinetic
dynamometer
50 m fc swim
Strass [31] 6 weeks
NR swim/week
4/week RT
Competitive 17, M
2, F
Control group: 17.8 ±3.9 years;
1.79 ±0.79 m; 67.6 ±9.1 kg
RT group: 16.6 ±1.2 years;
1.78 ±0.72 m; 65.2 ±9.4 kg
3 10 9 Isometric barbell arm
extension
25 m fc swim
50 m fc swim
Trappe and
Pearson [19]
11 weeks
5600 m/day swim
2/week RT
Collegiate 10, M 20.1 ±1.2 years; 1.83 ±0.26 m;
76.5 ±2.8 kg
6 5 5 Biokinetic swim
bench
Swim power (tethered
isokinetic device)
22.9 m fc swim
365.8 m fc swim
Aspenes et al. [7] 12 weeks
C6/week swim
2/week RT
Competitive 8, M
12, F
Control group: 15.9 ±1.1 years;
1.73 ±0.06 m; 58.3 ±6.6 kg
RT group: 17.5 ±2.9 years;
1.71 ±0.09 m; 58.9 ±10.2 kg
6 11 9 Tethered swim
Bilateral shoulder
extension
50 m fc swim
100 m fc swim
400 m fc swim
Girold et al. [18] 6 weeks
5/week swim
2/week RT
Regional/national 10, M
11, F
Control group: 16.5 ±1.5 years;
1.71 ±0.11 m; 62 ±4kg
RT group: 16.5 ±2.5 years;
1.71 ±0.09 m; 64 ±8kg
6 7 (7 in assisted
group)
7 Isokinetic
dynamometer
50 m fc swim
Tanaka et al. [16] 8 weeks
6/week swim
3/week RT
Collegiate 24, M Control group: 19.5 ±0.26 years;
182.5 ±2.12 m; 76.39 ±2.16 kg
RT group: 19.17 ±0.32 years;
1.81 ±1.47 m; 77.05 ±1.93 kg
5 12 12 Biokinetic swim
bench
22.9 m tethered fc
swim
365.8 m fc swim
Manning
et al. [30]
9 weeks
13–16.5 km
/day swim
3/week RT
National 7, M Control group: NR
RT group: 16.49 ±0.81 years;
1.79 ±0.51 m; 70.5 ±4.12 kg
47 No
control
Isokinetic
dynamometer
1 min ergometer
Standing long jump
22.9 m fc swim
91.44 m fc swim
182.88 m fc swim
E. Crowley et al.
123
Author's personal copy
Table 1 continued
References Training programme Level N,
sex
Age and anthropometrics
(mean ±SD)
a
PEDro
scale
RT (N) Control
(N)
Strength test
a
Swim test
a
Song et al. [29] 24 weeks
NR swim/week
2–4/week RT
National 10, M 16–23 years; 1.72–1.89 m; 67.5–87.5
kg
410 No
control
Maximum isometric
(lower back, grip
and arm pulling
strength)
Maximum isotonic
(power clean, bench
press, dead lift and
squat strength)
50 m run
Basketball throw
Jump (sargent and
standing long jump)
Swimming
competitions
Core training
Weston et al. [33] 12 weeks
8/week swim
3/week RT
National 10, M
10, F
Control group: 16.7 ±0.2 years;
1.7 ±0.03 m; 63 ±3kg
RT group: 15.7 ±1.2 years;
1.72 ±0.06 m; 63 ±5kg
7 10 10 Straight arm pull
down
Timed prone bridge
50 m fc swim
Swim-specific resistance training
Girold et al. [2] 3 weeks
10/week swim
2/week RT
Regional/ national 16, M
21, F
16.5 ±3 years;
1.71 ±0.13 m;
67 ±21 kg
715(11in
assisted
group)
11 Isokinetic
dynamometer
100 m fc swim
Konstantaki
et al. [34]
6 weeks
C5/week swim
(1.5 h)
3/week RT
County 15, M Control group:
16 ±3 years;
1.75 ±0.05 m;
72 ±8kg
RT group:
16 ±3 years;
1.75 ±0.05 m;
72 ±8kg
7 8 7 186 m arms-only fc
swim
372 m fc swim
Dragunas
et al. [35]
5 weeks
9/week swim (18 h)
3/week RT
Collegiate 10, M
8, F
Control group:
19.0 ±1.8 years
RT group:
19.3 ±0.87 years
6 9 9 50 m drag-suit swims 50 m fc swims
Impact of Resistance Training on Swimming Performance
123
Author's personal copy
Weston et al. [33] found large improvements in 50 m
swimming performance in the core training group, and
Dragunas et al. [35] found a small effect (p\0.05) on
stroke length and a trivial effect on 50 m swimming per-
formance in a drag-suit training group. There was a trivial
effect (p\0.001) on stroke length, and small effects
(p\0.05) on both stroke rate and 50 m swimming per-
formance in the control group.
Several method-related issues were apparent in the
review, including the absence of control groups [29,30], a
lack of randomised controlled trials [7,16,19,29–31,33,
36], and sex bias [16,19,28–30,34], with more specific
method-related issues requiring greater attention. Inter-
vention durations ranged from 3 [2] to 24 weeks [29]. Low
participant numbers were also a feature of the included
studies, ranging from 7 [30]to37[2]. A range of statistical
analysis methods were used, including significance levels,
effect sizes and confidence intervals. Different timing
systems were used across the reviewed studies, with some
using stopwatches [19,33,35], others using electronic
timing systems [28–30,36], and several studies not spec-
ifying how the timing data were collected [2,7,16,
18,32,34]. Different pool lengths were used in the studies
reviewed, including 25 m pools [2,7,18,32], 22.9 m
pools [16,19,28,30,34] and 50 m pools [29,33,36]; in
some cases the pool size was not stated [31,35].
A range of strength assessment tools were used to
monitor the changes in strength across the reviewed stud-
ies. Many studies [2,18,28,30,32] used an isokinetic
dynamometer to measure arm extensor and flexor strength
at different angular velocities, including concentric,
eccentric and isometric contractions. Other methods used
to quantify changes in strength included bilateral shoulder
extension [7], a straight arm pull down test [33] and a
shoulder extension test [30]. Swimming-specific strength
assessment tools, such as the biokinetic swim bench
[16,28], MAD device [36] and tethered swimming [7],
were also used to assess changes in strength.
4 Discussion
4.1 Method-Related Considerations
There were important method-related inconsistencies that
needed consideration when critically evaluating the find-
ings of this systematic review. Common method-related
considerations focus on the lack of control groups [29,30],
lack of randomised controlled trials [7,16,19,29–31,33,
36] and sex bias [16,19,28–30,34], with more specific
method-related issues requiring greater attention. To begin
with, interventions included in this review varied in dura-
tion from 3 [2] to 24 weeks [29], with most studies lasting
Table 1 continued
References Training programme Level N,
sex
Age and anthropometrics
(mean ±SD)
a
PEDro
scale
RT (N) Control
(N)
Strength test
a
Swim test
a
Toussaint and
Vervoorn [36]
10 weeks
8/week swim (1.5 h and
4.5 km/swim session)
3/week RT
Competitive
swimmers
(competed at
Dutch nationals)
16, M
6, F
Control group: 18.5 ±3.3 years;
1.78 ±0.08 m;
69.2 ±7.8 kg
RT group:
18.4 ±2.1 years;
1.79 ±0.07 m;
72.3 ±8kg
6 11 11 MAD-device 25 m fc swim
50 m fc swim
Mmale, Ffemale, RT resistance training, fc front crawl, PEDro Physiotherapy Evidence Database scale, MAD measurement of active drag, NR not reported, SD standard deviation
a
In the interest of brevity, baseline and post-test distances were all converted to metres
E. Crowley et al.
123
Author's personal copy
Table 2 Summary of resistance-training programme, relevant strength tests, technical parameters (stroke length, stroke rate) and swimming performance results
References Resistance-training programme Strength tests
a
Stroke length Stroke rate Swimming performance (m.s
-1
)
b
Biokinetic swim-bench training
Roberts
et al. [28]
Biokinetic swim bench; (4 910 s
max) 94; rest 30 s
Control group:
no significant difference
RT group:
No significant difference
Control group: (m)
1.77 ±0.14 to 1.79 ±0.12
RT group: (m)
1.87 ±0.37 to 1.9 ±0.31
Control group: (5 strokes.min
-1
)
62.6 ±4.6 to 61.0 ±3.9
RT group: (5 strokes.min
-1
)
60.3 ±5.2 to 58.3 ±8.2
Control group:
91.44 m: 1.83 ±0.04 to
1.81 ±0.04
RT group:
91.44 m: 1.76 ±0.08 to
1.81 ±0.06
Traditional resistance training
Girold
et al. [32]
Strength: 3 sets, 6 reps; pull ups,
latissimus pull downs, swim-bench
latissimus pull downs; rest 2 min;
80–90% 1RM
Control group:
No significant difference
RT group:
Extensors—concentric
(60°s
-1
): 11.2 ±13.6*
Extensors—concentric
(180°s
-1
): 16.9 ±11.7*
Control group: (m)
2.08 ±0.03 to 2.10 ±0.02
RT group: (m)
2.05 ±0.01 to 2.11 ±0.01*
Control group: (cycles.min
-1
)
53.5 ±2.4 to 53.9 ±2.3
RT group: (cycles.min
-1
)
54.7 ±4.1 to 55.9 ±2.7
Control group:
NR
RT group:
50 m: 2 ±1.3%*
Strass [31] Training the arm extensor muscles
using a barbell (80–90%):
3 reps 93 at 90%; 2 reps 93at
95%; 1 rep 91 at 100%
Assisted exercises for the arm flexors
and trunk also conducted but NR
Control group:
NR
RT group:
Max arm extension force
(N):
354 ±62 to 396 ±61**
Rate of force development
(N s
-1
):
4.01 ±0.8 to 4.96 ±1.2**
Control group: (m)
NR
RT group: (m)
25 m:
2.01 ±0.24 to
2.16 ±0.26**
50 m:
1.88 ±0.1 to
2.01 ±0.24**
Control group: (cycles.min
-1
)
NR
RT group: (cycle.min
-1
)
25 m
55 ±4 to 53.5 ±3.4*
50 m:
56.7 ±3.2 to 54.7 ±3.6*
Control group:
NR
RT group:
25 m: 1.83 ±0.1 to 1.91 ±0.1**
50 m: 1.77 ±0.08 to 1.81 ±0.08**
Aspenes
et al. [7]
5 reps 93; latissimus pull downs
Swimming: 4 94 min fc high
intensity;
rest 3 min active; 90–95% and
60–75%, respectively
Control group:
Bilateral shoulder extension:
277.9 ±44.2 to
310.7 ±56.2*
Tethered swim:
114.4 ±17.3 to
118.1 ±18.3
RT group:
Bilateral shoulder extension:
318.8 ±89.8 to
383.5 ±89.3**
Tethered swim:
124.9 ±23.2 to
133.5 ±21.9**
Control group: (m)
1.74 ±0.3 to 1.80 ±0.15
RT group: (m)
1.68 ±0.17 to 1.73 ±0.16
Control group: (hertz)
0.885 ±0.078 to 0.872 ±0.078
RT group: (hertz)
0.953 ±0.090 to 0.930 ±0.074
Control group:
50 m: 1.69 ±0.1 to 1.71 ±0.1
c
100 m: 1.54 ±0.1 to 1.56 ±0.12
c
400 m: 1.38 ±0.08 to 1.38 ±0.08
c
RT group:
50 m: 1.73 ±0.12 to 1.75 ±0.11
c
100 m: 1.59 ±0.1 to 1.61 ±0.1
c
400 m: 1.38 ±0.16 to 1.4 ±0.08
c
*
Impact of Resistance Training on Swimming Performance
123
Author's personal copy
Table 2 continued
References Resistance-training programme Strength tests
a
Stroke length Stroke rate Swimming performance (m.s
-1
)
b
Girold
et al. [18]
Dry-land: 6 reps 93; bench press,
pull up, barbell draws, squats,
plyometrics;
80–90%; rest 2 min
Control: 19(run/cycle) [60–70%]
Control group:
Extensors (180
8
s
-1
):
15.7 ±10.6*
RT group:
Extensors—isometric:
45.5 ±38.7*
Extensors—concentric
(60
8
s
-1
): 33.7 ±27.6*
Extensors (180
8
.s
-1
):
35.2 ±31.9*
Flexors—isometric:
39.5 ±32.4*
Control group: (m)
1.56 ±0.09 to 1.56 ±0.08
RT group: (m)
1.61 ±0.11 to 1.59 ±0.09
Control group: (cycle.min
-1
)
47.8 ±3.7 to 48.7 ±3.7*
RT group: (cycle.min
-1
)
48.9 ±4.98 to 50.7 ±3.71
Control group:
50 m: 0.9 ±1.2%
RT group:
50 m: 2.8 ±2.5%*
Trappe and
Pearson
[19]
Weight-assisted group: dips and pull
ups, until failure. Set 1: no weight;
Set 2: 13.6 kg; Set 3: 22.7 kg
Traditional weight-lifting group:
latissimus pull downs, elbow
extension and flexion, bent-arm fly,
quadriceps extensions, hamstring
flexion. 8–12 reps; 3 sets/exercise
Weighted assisted group:
Pull ups:
0 kg: 9.6 ±2.6 to 16 ±2.6
(66.7%)
13.6 kg: 9.2 ±0.7 to
26.3 ±5.6 (185.8%)
22.7 kg: 12 ±1.1 to
62.5 ±6 (420.8%)
Tricep dips:
0 kg: 20.2 ±3.8 to
41 ±6.9 (102.9%)
13.6 kg: 12.6 ±1.5 to
40 ±3.1 (217.5%)
22.7 kg: 14.4 ±2to
80.5 ±4.2 (459%)
No statistical significance
reported
Biokinetic swim bench NR*
Traditional weight-training
group:
NR
Weeks 4–12
Weighted assisted group: (m)
2.09 ±0.05 to 2.17 ±0.07
Traditional weight-training
group: (m)
2.24 ±0.17 to 2.24 ±0.12
Weeks 4–12
Weighted assisted group: (time to
complete 4 stroke cycles)
5.83 ±0.14 to 5.84 ±0.25
Traditional weight-training group: (time
to complete 4 stroke cycles)
6.29 ±0.67 to 6.01 ±0.36
(time to complete 4 stroke cycles)
Weighted assisted group:
22.9 m: 2.04 ±0.01 to 2.1 ±0.05*
365.8 m: 1.43 ±0.02 to
1.49 ±0.01*
c
Traditional weight-training group:
22.9 m: NR
365.8 m: 1.43 ±0.3 to
1.49 ±0.02*
c
E. Crowley et al.
123
Author's personal copy
Table 2 continued
References Resistance-training programme Strength tests
a
Stroke length Stroke rate Swimming performance (m.s
-1
)
b
Tanaka
et al. [16]
8–12 reps; 3 sets; tricep dips, pull
ups, latissimus pull downs, tricep
extensions, bent over fly
Control group:
Biokinetic swim-bench test:
NR*
22.9 m tethered fc swim:
NR*
RT group:
Biokinetic swim-bench test:
NR*
22.9 m tethered fc swim:
NR*
NR (m) NR Control group:
22.9 m fc: no values reported*
RT group:
22.9 m fc: no values reported*
Manning
et al. [30]
Circuit; 1 min on and rest 30 s
Weeks 1–3: 30%; weeks 4–6: 40%;
Weeks 7–9: 50% 1RM
Exercises: leg extension, bench
press, sit-ups with weight, shoulder
press, skipping, latissimus pull
down, leg flexion, arm curl, cycle
ergometer, shoulder shrug, side
bend
Control group:
NR
RT group:
Plantar and dorsiflexion:
R: 17.21 ±78.03 to
16.99 ±120.15*
L: 21.98 ±82.71 to
16.87 ±125.65*
Knee extension and flexion:
R: 116.57 ±610.77 to
111.8 ±736.48*
L: 93.97 ±586.93 to
85.91 ±701.62*
Shoulder extension and
flexion:
R: 68.08 ±334.99 to
65.18 ±427.49*
L: 70.55 ±332.55 to
55.83 ±431.52*
Ergometer (rev.min
-1
):
146.83 ±10.4 to
167.33 ±16.95*
Not recorded Not recorded Control group:
NR
RT group:
22.9 m: 0.95 ±0.03 to
0.99 ±0.04
c
91.44 m: 1.81 ±0.35 to
1.82 ±0.2
c
182.88 m: 1.64 ±0.04 to
1.66 ±0.06
c
Impact of Resistance Training on Swimming Performance
123
Author's personal copy
Table 2 continued
References Resistance-training programme Strength tests
a
Stroke length Stroke rate Swimming performance (m.s
-1
)
b
Song et al.
[29]
Adaptation phase:
11 exercises; 10–12 reps 93,
40–70%, 4 weeks, 4/week; max
strength phase: 9 exercises; 3–7
reps 94–5,
85–100%, 4/week; power endurance
phase: 6 exercises 7–10 93,
70–80%, 2/week
RT group:
Lower back strength:
1.4–17.3%*
Grip strength:
Dominant arm:
7.3–8.3%**
Nondominant arm:
5.9–6.7%*
Ball throw:
5.7–9.7%**
Sargent jump:
8.6–11.6%**
Standing long jump:
2.2–5%*
Not recorded Not recorded Mean improvement:
(?) 1.3%
Range:
(-) 6.9% to (?)4.2%
[p-value NR]
Core training
Weston
et al. [33]
Prone bridge and side bridge:
30 s (weeks 1–2), 60 s (weeks 3–4),
90 s (weeks 5–8), 120 s (weeks
9–12)
Straight leg raises and bird dog: 10
reps (weeks 1–2), 15 (weeks 3–4),
20 (weeks 5–6), 25 (weeks 7–10),
30 (weeks 11–12)
Overhead squat and sit and twist (?5
reps weeks 1–4): 10 (3 kg; weeks
1–2), 10 (4 kg; weeks 3–4), 15
(5 kg; weeks 5–6), 20 (6 kg; weeks
7–8), 20 (7 kg; weeks 9–10), 25
(7 kg)
Shoulder press: 10 reps (weeks 1–4),
15 (weeks 5–6), 20 (weeks 7–10),
25 (weeks 11–12)
Control group:
Straight arm pull down:
(?)2.5; -2.9 to 8.1
Prone bridge:
(?)4.7; 0.2–9.3
RT group:
Straight arm pull down:
(?)26.2; 19.6–33.1
Prone bridge:
(?)14.1; 9.2–19.2
Not recorded Not recorded 50 m fc swim:
90% confidence interval
Control group:
50 m: 0.7%; 1.6 to 0.2
Intervention group:
50 m: 2.7%; 4.2 to 1.1 (large)
E. Crowley et al.
123
Author's personal copy
Table 2 continued
References Resistance-training programme Strength tests
a
Stroke length Stroke rate Swimming performance (m.s
-1
)
b
Swim-specific resistance training
Girold
et al. [2]
Resisted swim:
6930 s max fc, rest 30 s
Control group:
Flexors—isometric:
15.5 ±16.8*
RT group:
Flexors—concentric
(60
8
s
-1
): 8.6 ±9.1*
Flexors—concentric
(180
8
s
-1
): 14 ±15.3*
Extensors—isometric:
31.5 ±24.9*
Control group: (m.min
-1
)
1.95 ±0.2 to 2.01 ±0.19
RT group: (m.min
-1
)
2.03 ±0.18 to 2.01 ±0.18
Control group: (cycle.min
-1
)
43.5 ±4.92 to 42.2 ±3.56
RT group: (cycle.min
-1
)
42.32 ±4.98 to 43.01 ±3.91*
Control group:
100 m: 1.47 ±0.13 to 1.46 ±0.13
c
RT group:
100 m: 1.48 ±0.1 to 1.51 ±0.09*
c
Konstantaki
et al. [34]
Arms only:
Elastic band around ankles and pull
buoy in between upper leg
1333 m arms only
Control group:
1.6 ±0.5%
RT group:
14.02 ±3.6%*
Not recorded Not recorded Control group:
372 m: 1.15 ±0.03 to 1.16 ±0.03
c
RT group:
372 m: 1.17 ±0.04 to 1.18 ±0.06
c
Dragunas
et al. [35]
Drag suit:
Set 1: 3 922.9 m sprint ?182.8 m
submax swim
Set 2: 4 922.9 m sprints ?30 s
interval ?182.8 m submax swim
Set 3: 16 925 m sprints ?1 min
interval
Drag-suit group:
NR
Control group:
NR
Control group: (m)
SE =0.015
Effect size =0.133
(trivial)***
RT group: (m)
SE =0.017
Effect size =0.333 (small)*
Control group: (strokes.min
-1
)
SE =0.652
Effect size =0.226 (small)*
RT group: (strokes.min
-1
)
SE =0.652
Effect size =0.195 (trivial)
Control group:
50 m: Effect size =0.265(small)*
RT group:
50 m: Effect size =0.006(trivial)*
Impact of Resistance Training on Swimming Performance
123
Author's personal copy
between 3 and 12 weeks. The greater the duration of the
intervention, the greater the probability of a significant
adaptation occurring. Participant numbers also ranged
from 7 [30]to37[2]. Specifically, low participant num-
bers reduced the statistical power of the studies and
therefore the merits of such studies needed to be evalu-
ated in this context. In terms of statistical reporting, the
use of significance levels, effect sizes and confidence
levels were inconsistent, with more recent research (e.g.
Weston et al. [33]) employing effect sizes and confidence
intervals. It is important to note that the range of
improvement of athletes is related to the athlete’s initial
training state, and more elite performers have a reduced
range of improvement. Therefore, effect sizes are
becoming a more popular and relevant statistic for anal-
ysis of athletic populations. Different timing systems
were used across the reviewed studies, with some using
stopwatches [19,33,35], others using electronic timing
systems [28–30,36] and several studies not specifying
how the timing data were collected [2,7,16,18,32,34].
The stopwatch has been shown to produce faster times
(0.04–0.24) than electronic timing systems [37,38]and
would be deemed an unreliable method for collecting data
in a number of sports/contexts. Finally, the different types
of pools that were used in the studies reviewed make it
hard to determine whether increasesinswimmingper-
formance were due to an increase in swimming velocity
or an increase in ground reaction forces while turning and
diving. Within this review, some studies used 25 m pools
[2,7,18,32], others used 22.9 m pools [16,19,28,
30,34] and some used 50 m pools [29,33,36]. The pool
size was not stated in some of the reviewed studies
[31,35]. These method-related considerations in study
designs are important when interpreting the findings of
these studies.
4.2 Strength Diagnostics
A range of resistance-training interventions were consid-
ered in this review. To assess the effectiveness of these,
valid and reliable strength tests are required to monitor
changes in performance following the interventions.
Pichon et al. [39] stated that strength tests need to be
specific to the movement pattern of swimming, while
Newton et al. [10] emphasised the importance of assessing
multi-joint exercises in strength testing rather than isolated
exercises, since multi-joint exercises represent the coor-
dinated effort, whereas isolated exercises may lack validity
as they do not examine the overall force generation
capabilities.
For strength assessment, many studies have used the
isokinetic dynamometer [2,18,28,30,32] and tested arm
extensor and flexor strength at different angular velocities.
Table 2 continued
References Resistance-training programme Strength tests
a
Stroke length Stroke rate Swimming performance (m.s
-1
)
b
Toussaint
and
Vervoorn
[36]
Sample programme:
20 923 m lengths; every 35 s: 1
length sprinting on the POP, the
other swimming back slowly
10 923 m lengths kicking
6923 m lengths; every 110 s: 39
(1 length sprint on POP, 1 length
butterfly)
Control group:
Power (W):
143.4 ±39.4 to
142.5 ±39
Force (N):
82.8 ±18.7 to 80.9 ±17.5
RT group:
Power (W):
160.4 ±43.0 to
171 ±51.3*
Force (N):
91 ±19.7 to 94 ±22.7***
Not recorded Not recorded Control group:
25 m: 1.88 ±0.11 to 1.89 ±0.11
c
50 m: 1.74 ±0.1 to 1.74 ±0.09
c
RT group:
25 m: 1.9 ±0.13 to 1.92 ±1.11
c
50 m: 1.75 ±0.1 to 1.77 ±0.1
c
m.s
-1
metres per second, 8.s
-1
degrees per second, rev.min
-1
revolutions per minute, NNewtons, N.s
-1
Newtons per second, Wwatts, SE standard error, -increase in time, ?decrease in time, RT
resistance training, RM repetition maximum, reps repetitions, max maximum, submax submaximum, fc front crawl, POP push-off pads, Rright, Lleft, NR not reported, * indicates p\0.05, ** indicates
p\0.01, *** indicates p\0.001
a
In the interest of brevity, only significant p-values and effect sizes (where reported) are shown
b
In the interest of brevity, swimming performance is reported in metres per second
c
For consistency, seconds were converted to metres per second
E. Crowley et al.
123
Author's personal copy
Although arm extension is very applicable to front crawl
(freestyle) swimming, this test lacks specificity of the
multi-joint movements and changes in velocities through-
out the propulsive phase that occur in front-crawl swim-
ming. Isometric testing used the isokinetic dynamometer
conducted with the arm at a 90°position, which represents
the specific mid-stroke position [40]. However, while it
does not mimic the dynamic nature of swimming, isometric
testing has been shown to have a strong relationship
(r=0.61–0.72) with dynamic strength performance [41].
Nonetheless, it is clear that isometric testing has the ability
to examine maximum force at specific stroke phases and
may therefore be a useful strength assessment tool for
swimming coaches and practitioners. Other studies have
used more specific dynamic methods to measure maximal
strength, including shoulder extension as this is important
in the propulsive phase of front-crawl swimming. Of par-
ticular note, Aspenes et al. [7] assessed bilateral shoulder
extension strength by means of a cable crossover apparatus
from a starting position of 170 ±10°of shoulder flexion,
Weston et al. [33] conducted a straight arm pull down test,
and Manning et al. [30] conducted a shoulder extension
test. Biokinetic swim bench [16,28], MAD device [36] and
tethered swimming [7] are three other methods that are
deemed specific to swimming performance. The biokinetic
swim bench and the MAD device are discussed further in
Sects. 4.3.1 and 4.4.4, respectively. Tethered swimming
has been shown to be similar to front-crawl swimming in
terms of muscle activation [42] and propulsive forces
(r=0.92, p\0.01) [43]. This may be a useful tool to
monitor the transference of resistance-training exercises to
actual swimming performance.
4.3 Dry-Land Resistance-Training Modalities
4.3.1 Biokinetic Swim-Bench Training
The biokinetic swim bench is posited to mimic swimming
performance without the use of the lower extremities [14].
The swimmer lies prone on a sliding bench with a slight
incline, arms outstretched over his/her head and hands
secured in hand paddles. The swimmer is then able to pull
along the sliding bench and therefore mimic the kinematics
of front-crawl swimming. Insufficient elbow control during
this movement can generate inaccurate findings [31] and
this is an important limitation that needs attention. The
maximal power produced on the swim bench has been
shown to have a strong relationship (r=0.92) with
swimming velocity in semi-tethered conditions [44];
however, the biokinetic swim bench still lacks specificity
as the pulling pathway of the hand is longer and the dis-
tribution of pulling forces at various joint angles is not
similar to swimming in water [45]. Tanaka et al. [16] used
the biokinetic swim bench to monitor strength improve-
ments from a traditional resistance-training programme.
Despite finding no significant improvement in maximal
power produced on the biokinetic swim bench, significant
(p\0.05) improvements in swimming performance were
observed. This highlights the necessity for specificity when
choosing resistance-training exercises.
Roberts et al. [28] designed an intervention with high-
velocity swim training and biokinetic swim-bench training
sessions performed three times weekly in a 22.9 m pool. It
is important to note that pool length and the inclusion of all
male participants affects the practical applications of this
study, as mentioned in Sect. 4.1. The results showed that
swimming performance in both the control group and the
biokinetic swim-bench group did not improve. This may be
due to the intervention group being more fatigued because
of the biokinetic swim-bench intervention period, com-
pared with the control group. Because of the limited
research and conflicting results, it is hard to determine the
transfer of biokinetic swim-bench training to swimming
performance. Lack of improvement may be due to the
increased stability provided by lying on a stable surface
compared with the relative instability of the water, and the
inability to replicate the drag propulsion relationship on
land. The biokinetic swim bench lacks specificity and does
not transfer to significantly improve swimming
performance.
4.3.2 Traditional Resistance Training
Traditional resistance training is widely used in many
sports and involves conventional gym-based resistance-
training exercises such as bench press, latissimus pull
downs, tricep extensions, tricep dips, bent-arm flies, pull
ups and squats. Low-volume, high-velocity/force resis-
tance-training programmes resulted in significant
improvements in swimming performance. Trappe and
Pearson [19] found that a resistance-training programme
for 6 weeks resulted in a significant improvement in
22.9 m front crawl sprint performance. Girold et al. [32]
found a 2% increase in 50 m performance in a resistance-
training group, which is similar to the 2.1% increase
reported by Strass [31] and the 2.8% increase reported by
Girold et al. [18]. Strass [31] also found improvements in
maximal explosive force production compared with maxi-
mal force production, and the authors proposed that this
could be due to various neuromuscular adaptations. Neu-
romuscular adaptations refer to improved motor unit
recruitment, synchronisation, co-contraction, rate coding,
intra- and inter-neuromuscular coordination and neural
inhibition. This gain in explosive force production trans-
ferred to swimming performance with a significant increase
in velocity (p\0.01). Furthermore, Aspenes et al. [7]
Impact of Resistance Training on Swimming Performance
123
Author's personal copy
employed a similar training intervention method and found
a 20.3% improvement in dry-land strength. The control
group increased by 11.8%, and it may be possible that the
control group increased dry-land strength due to famil-
iarisation with the testing procedures, or the improvement
resulted from the prescribed swimming programme.
Despite this, improvements did not transfer to significant
improvements in 50 and 100 m swimming performance;
however, 400 m performance did improve significantly,
with a mean decrease in swimming time of 3.74 s.
By contrast, the applications of high-volume resistance-
training programmes appear to have no significant effect on
swimming performance. Song et al. [29] used a periodised
gym-based programme (24 weeks) with four phases: (1)
adaptation; (2) maximum strength; (3) power endurance;
and (4) maximum strength and power/endurance. In the
case of this study, strength improvements were observed,
ranging from a 21% increase in sit-ups to a 20–25%
increase in power clean. It is likely that the initial increase
in strength can be associated firstly with neuromuscular
adaptations affecting fast-twitch muscle fibres, and, sec-
ondly, with improvement of the exercise technique (i.e.
power clean). Manning et al. [30] designed a traditional
dry-land resistance circuit, consisting of various upper
body, lower body and core exercises, each performed for
1 min to mimic the swimmer’s event duration. The training
programme was designed to target specific muscle groups
involved in front-crawl swimming, and exercises were
completed with maximum velocity to maximally recruit the
appropriate motor units. Despite increased muscle power
most likely resulting from neurological adaptation, Man-
ning et al. [30] did not find a significant increase in
swimming performance, which may be due to the hyper-
trophic nature of the training programme prescribed, and
residual fatigue may have been a key factor. Tanaka et al.
[16] found that swimmers in a resistance-training inter-
vention group increased strength by 27–35%, but this
improvement did not transfer to swimming performance.
The heavy demands of both swimming and resistance
training may have caused local muscular fatigue and
inhibited the development of maximal swimming power,
although serum cortisol, an indicator for overtraining, did
not change over the course of the season. This reinforces
that training modalities need to be swim-specific to trans-
late to improved swimming performance.
The results of this systematic review are consistent with
the view that traditional resistance training improves
swimming performance. Girold et al. [32], Girold et al.
[18], Strass [31] and Aspenes et al. [7] all found that tra-
ditional resistance-training methods increased dry-land
strength and found significant improvements in swimming
performance. Each of these studies used high velocities
during the concentric phase and this has been linked to
greater neuromuscular adaptations and recruitment of type
II muscle fibres [46–48]. Girold et al. [32], Girold et al.
[18] and Aspenes et al. [7] all employed low repetition
ranges (1–6 repetitions) with a low number of sets (B3
sets), whereas Strass [31] prescribed a range of sets across
a low number of repetitions for one exercise (i.e. three sets
of three repetitions, three sets of two repetitions and one set
of one repetition). This method of programming will result
in less neuromuscular fatigue. These adaptations, along
with greater specificity (resistance exercises specific to the
event, both in terms of muscle groups and velocity, to
produce optimal performance enhancements), which has
been mentioned in several of the studies, may lead to the
significant transfer to swimming performance
[7,17–19,30]. This indicates that lower volume (low
number of sets and repetitions), high-intensity (high
velocity/force) resistance-training programmes induce less
neuromuscular fatigue and greater strength and neuro-
muscular improvements, which may positively influence
swimming performance [11]. This is in agreement with
Ramos Veliz et al. [49], who prescribed a low-volume,
high-velocity/force resistance-training programme to elite
waterpolo players. However, no control group [29], sex
bias [7,18,31,32], lack of randomised controlled design
[7,29,31] and intervention durations lasting less than
8 weeks [18,31,32] are a feature of the previously men-
tioned studies. The specific timing system used was only
documented in one study [29]. Furthermore, three of the
studies used a 25 m pool [7,18,31] and one study used a
50 m pool [29], with Strass [31] not reporting pool length.
These shortcomings can alter the practical application of
these studies, with intervention length, lack of control
groups and randomised controlled design affecting the
quality and interpretation of the studies. Conversely,
Tanaka et al. [16], Trappe and Pearson [19] and Manning
et al. [30] found an increase in dry-land strength, with no
increase in swimming performance. This may have been
due to the hypertrophic nature of these programmes
(Table 2)[50], which can impair performance by increas-
ing resistive forces and decreasing the force-to-mass ratio.
The increase in volume of these programmes would have
caused an increase in neuromuscular fatigue and would
therefore have inhibited the anabolic environment neces-
sary for recovery and strength development.
4.3.3 Core Training
Core training is commonly practiced in elite sport. In
swimming, core stability is essential due to the unsta-
ble nature of water. During each stroke cycle, propulsive
forces are produced through the hand, which create a
dynamic reaction about the rotational axes of the vertebrae
causing an increase in lateral movement, increase in
E. Crowley et al.
123
Author's personal copy
excessive kicking movements and therefore a decrease in
propulsive efficiency. Due to this instability, muscular
strength and endurance in core muscles is required to
absorb these forces. Weston et al. [33] demonstrated sig-
nificant improvements in swimmers’ core function and
swimming performance due to the inclusion of an isolated
core programme, while Dingley et al. [51] found similar
results with paraplegic athletes, where the increase in
swimming performance was largely attributed to the
increase in core stability of the athletes. Dynamic exercises
such as ‘bird dog’ [33] enhanced stability and control of the
gluteus maximus, external oblique abdominis, gluteus
medius and other specific muscle groups. Weston et al. [33]
identified difficulties in designing a core programme that
matched the biomechanical complexity of the core muscle
functions in swimming. Furthermore, the use of a manual
stopwatch to determine times would have affected the
quality of data obtained in this study. Despite this, the
study found moderate to large improvements in peak
electromyography (EMG) activity of the latissimus dorsi,
external oblique abdominis and lumbar extensor muscles
during isolated core tests. Therefore, the introduction of
core exercises may have increased stability in the lumbar
and thoracic regions through a variety of these exercises;
however, it is important to highlight that no randomised
controlled trials have been conducted specifically exploring
this.
4.4 Swim-Specific Resistance Training
4.4.1 Resisted-Band Swim Training
Resistive bands have been used as a resistance-training
exercise modality and have been shown to have significant
effects on swimming performance. In this training modal-
ity, an elastic band is securely attached to the diving block
and a belt with the elastic band is attached around the
swimmer’s waist. The athlete swims out against the resis-
tance of the elastic band. Girold et al. [2] found a 1.9%
improvement in performance over 100 m following resis-
tive-band training, however inconsistencies in the appli-
cation of the methods in this study rendered the findings
unreliable. This included a 3-week intervention period, no
reporting of the specific timing system used and the use of
a 25 m pool, all of which could have affected the quality of
the study. In a subsequent study, Girold et al. [18] found
significant improvements (p\0.05) in 50 m swimming
performance, with a 32% increase in strength, but unfor-
tunately the use of combined resisted and assisted training
excluded this study from the current systematic review.
Girold et al. [2] showed that the use of in-water resistance
training can increase swimming performance, and empha-
sised the need for swim-specific resistance training
[52,53], which is in agreement with Delecluse et al. [54],
who suggested that high resistances enhance power due to
adaptive changes to the neuromuscular system. The
greatest improvements in strength were observed in the
isometric condition which was conducted at a 908angle at
the arm; this is the specific mid-stroke swimming position
between the pulling and pushing phase [55]. Therefore, a
gain in muscle strength in this position may have been the
significant indicator of improvement in swimming perfor-
mance. This emphasises the need for specificity of training
modalities. Santos-Garcia et al. [56] showed that an
increase in swimming performance can be observed after
four rounds of a swim-specific resistance-training exercise
(i.e. resistive elastic bands), followed by a maximum
sprint. This suggests the requirement for extensive warm-
ups to ensure maximum swimming performance. It is
plausible that the use of resistive elastic bands as a resis-
tance-training tool promotes an increase in swimming
performance [2,56] and provides a dynamic resistance-
training method that develops muscle strength specific to
swimming. This is consistent with De Villarreal et al. [57],
who found that specific in-water resistance exercises
improved waterpolo-specific performance skills, re-em-
phasising the necessity for specificity.
4.4.2 Arms-Only Training
Arms-only training is commonly used in elite swimming
and is believed to increase upper body resistance in a
swimming-specific manner by overloading the specific
upper body muscles; therefore, arms-only training can be
seen as a resistance exercise rather than an endurance
exercise. When compared with whole-body swimming,
arms-only swimming (using a pull buoy to increase the
buoyancy of the swimmer) reduces maximal oxygen
uptake [58,59] at the same relative intensity. This training
modality can be used to reduce total training load while
maintaining swimming fitness and volume. Konstantaki
et al. [34] replaced regular swimming training with arms-
only training three times a week, consisting of swimming
exercises and drills (i.e. breathing drills, one arm only,
hand paddles, pull buoy, etc.). Significant improvements
were noted in arms-only peak exercise intensity, ventila-
tory threshold and movement economy, but with no sig-
nificant transfer to swimming performance. This lack of
transfer may be due to the loss of coordination between the
arms and legs, as swimming speed, type of kick used (2, 4
or 6 beat) and swimming proficiency are factors that rely
on coordination. Isolation of the arms may add additional
external torque due to the changes in body roll and stability
in the water. It may be suggested that a longer period of
time is needed for the body to adapt to this increase in
external torque, with sex bias, unknown timing system and
Impact of Resistance Training on Swimming Performance
123
Author's personal copy
a 22.9 m pool all affecting the quality and practical
application of this study. The study also used a pull buoy,
which provides extra buoyancy and support, and which, in
turn, reduces the activation and stabilisation of the core
muscles. This reinforces the vital role of the core muscles
in maintaining an optimal body position, as described by
Morris et al. [60], for elite swimming performance. Despite
its limitations, arms-only training still has a very important
role in swimming performance. It may be necessary to
isolate the legs, using an elastic band around the ankles
rather than a pull buoy, to create greater core activation and
maintain an optimal body position [60]. While the research
is not conclusive, it is plausible that an integration of arms-
only swimming in the training programme may improve
the transfer to swimming performance.
4.4.3 Drag-Suit Training
Training specificity is a key element in the enhancement of
swimming performance. Drag suits have been worn by
swimmers worldwide and are commercially advertised as a
training tool. Drag suits are traditionally considered an
endurance training modality but due to the increase in
propulsion and consequent increase in work output, their
use can be considered a resistance exercise. The concept
behind drag suits is that the mesh clothing retains water and
therefore increases the resistive drag forces. Dragunas et al.
[35] showed no significant improvement in swimming
performance while wearing a drag suit for 5 weeks com-
pared with the control group. This may be due to the
adaptation from drag-suit training requiring longer than
5 weeks, with sex bias, use of a stopwatch, and not
reporting pool length all affecting the quality and results of
this study. Training under drag-suit training conditions on a
daily basis may change the body position and ‘feel for the
water’ [61], which may result in slower swimming times
without the drag suit. Previous research has shown that
there are no acute alterations in kinematic characteristics of
front-crawl swimming while using parachutes [62], a
similar concept to drag-suit training. To date, no research
has examined the chronic effect of parachute or drag-suit
training on kinematic characteristics, and it is unknown
whether chronic training under drag-suit conditions alters
body position and affects swimming performance.
4.4.4 Measurement of Active Drag (MAD) Device
Toussaint and Vervoorn [36] used the MAD device, a
specific in-water system to measure mechanical power
output as well as active drag during front-crawl swimming,
in their study [22]. The MAD device consists of 16 push-
off pads, 1.35 m apart on a 23 m horizontal rod 0.8 m
below the water surface. The system requires the swimmer
to push off the resistance pads submerged under the water
at fixed points. The force exerted on these fixed points is
then recorded through a force transducer. This device is
highly specific to swimming but requires several famil-
iarisation trials to produce a reliable and valid result. A
training device known as the POP (push-off pad) device
was used to mimic the MAD device during the 10-week
training intervention period designed by Toussaint and
Vervoorn [36]. Toussaint and Vervoorn [36] found no
significant improvements in 25 and 50 m swimming per-
formance; however, competition results showed significant
improvements in 50 and 200 m swimming performance
compared with the control group, but 100 m swimming
performance did not significantly improve compared with
the control group. This increase in power, velocity and
force may be due to the added resistance encountered by
propelling off the fixed pads rather than water. Clarys et al.
[63] found similar EMG patterns when comparing the
MAD device and free swimming, which would suggest that
the MAD device is task-specific. The high correlation
between competition swimming performance and force,
velocity and power, from this study, may suggest that the
MAD device is a sensitive device to monitor training
improvements and the transfer of resistance-training
modalities. The authors of this study also suggest that the
MAD device can force the swimmer to swim more sym-
metrically due to the fixed depth and distance of the pads,
which is an important practical feature of the MAD device.
Although the findings of this study are positive, the com-
pliance of the participants was poor and therefore the
findings need to be interpreted with this in mind. Further-
more, no randomised controlled trials have been conducted
in this area and the available data should therefore be
interpreted with caution.
5 Biomechanical Parameters
Stroke length and stroke rate are the two biomechanical
parameters that are investigated in the majority of studies,
and have been linked to improvements in swimming per-
formance. Many studies have highlighted the importance of
stroke length in improving swimming performance
[2,20,64–68] and therefore maintaining mechanical
propulsive efficiency throughout the race. Stroke rate has
been determined as a key factor in 50 m performance
compared with other swimming distances [18], and has
been known to influence swimming velocity [3,64,67].
Swimming velocity is the product of stroke length (metres
per stroke cycle) and stroke rate (strokes per minute) [69],
therefore these technical parameters are important deter-
minants of swimming velocity. Researchers have attempted
to find the optimal relationship of stroke length to stroke
E. Crowley et al.
123
Author's personal copy
rate for swimming performance [70] and have found that
training, intensity, physiological capabilities, race dis-
tances, sex and swimming technique all influence the
stroke length/stroke rate relationship [71]. Expert swim-
mers have been shown to maintain their stroke length and
stroke rate longer than non-expert swimmers, with coaches
employing various training strategies to maintain these
technical parameters and achieve higher swimming
velocities [72]. Five different combinations of these
parameters exist: (1) increase in stroke length and stroke
rate; (2) increase in stroke length and decrease in stroke
rate; (3) increase in stroke rate and decrease in stroke
length; (4) increase in stroke length and stabilisation of
stroke rate; and (5) increase in stroke rate and stabilisation
of stroke length [73]. It is therefore imperative that the
calculations of stroke length and stroke rate are accurate
and it is essential that this is considered before discussing
the effect of resistance training on these technical
parameters.
5.1 Calculation of Stroke Length and Stroke Rate
The calculations of stroke length and stroke rate included in
this review are shown in Table 3. Many methods are used to
determine stroke length and stroke rate, with different units
being implemented, as observed in Table 3.Various
method-related issues affect stroke rate and length deter-
mination, with the major factor being the method used to
identify the start and finish of each stroke cycle. Due to
measurement inconsistencies, a standardised model to
determine stroke length and stroke rate is needed to ensure
accurate reporting. The implementation of landmark regis-
tration could make the measurement of stroke length and
stroke rate more valid and reliable. Landmark registration is
an intermediate step in functional data analysis, a branch of
statistical analysis that presents curves or families of curves
[74]. Landmark registration identifies visible features or
landmarks and realigns curves to ensure the landmarks of
multiple trials (curves) are consistent. This technique allows
a more effective comparison between participants without
resorting to distortions of the time-base via conventional
temporal normalisation techniques. In swimming, the tran-
sition of the arms from the propulsive phase (under water) to
the recovery phase (above the water) will cause a clear
difference in velocity, force and displacement of the arm-
stroke cycle, and this provides an easily identifiable land-
mark in the curve. The advantage of landmark registration is
the identification of movement patterns where the sequence
of events is clearly identifiable. The disadvantage with this
technique is that landmark registration will introduce a
change to the raw data and therefore realigned data needs to
be interpreted with some caution. A clear description of this
analysis is provided by Deluzio et al. [74].
5.2 Stroke Length
Hay et al. [68] demonstrated that the improvement in
velocity over a season was due to improvements in stroke
length. Improvements in performance were observed with
an increase in stroke length, and therefore an increase in
velocity [68]. A decrease in velocity has been shown to be
directly associated with a decrease in stroke length
[67,75]. Stroke length can therefore be shown to be an
important factor in freestyle swimming [65,66]. Girold
et al. [32] and Strass [31] found an increase in stroke
length, and this transferred to an increase in swimming
performance. These studies used traditional resistance
training with low repetitions (Girold et al. [32], six repe-
titions; Strass [31], three repetitions). Aspenes et al. [7] and
Girold et al. [18] found no such increases in stroke length
but significant increases in 400 m and 50 m swimming
performance, respectively. Aspenes et al. [7] and Girold
et al. [18] also used traditional resistance training with low
repetitions (five and six repetitions, respectively). The
results here are conflicting, therefore further clarity/addi-
tional research is needed. Toussaint and Vervoorn [36]
conducted 25 and 50 m sprint times outside of the com-
petition, used to monitor pre- and post-training effects of
the POP training device on swimming performance. During
these tests, significant reductions in the number of strokes
taken were recorded in the resistance-training group: 25 m,
19 ±2.3 to 18.6 ±2; and 50 m, 46.4 ±3.7 to 44.9 ±3.9.
By contrast, Tanaka et al. [16] found that an increase in
resistance-training performance did not result in an
increase in stroke length and found no significant
improvement in swimming performance. Tanaka et al. [16]
found that improvements through resistance training did
not transfer to propulsive forces needed for an increase in
stroke length. The resistance-training group increased
strength by 25–35%, but no transfer to swimming perfor-
mance was found. This may have been due to the hyper-
trophic nature of this intervention (Table 2). Dragunas
et al. [35] found a significant increase in stroke length but
no increase in swimming performance, similar to the
above. While these results are inconclusive, they suggest
that low repetitions with high velocity/force are needed for
improving stroke length when prescribing resistance-
training exercises. This may be because increasing stroke
length demands higher strength levels, as evidenced by
Toussaint and Vervoorn [36].
5.3 Stroke Rate
Using stepwise regression analysis, Girold et al. [2] found
stroke rate to be an important factor in increasing swim-
ming performance over 100 m. This study showed that
resisted swims caused the swimmer to increase his/her
Impact of Resistance Training on Swimming Performance
123
Author's personal copy
stroke rate, providing sufficient propulsive forces to move
forward rather than being pulled back by the resisted band.
The fact that an increase in stroke rate was only observed in
the second half of the 100 m test trial suggests an increase
in muscular strength endurance due to the resisted swims.
Theoretical analyses show that stroke rate is the primary
Table 3 Summary of units and methods used to calculate stroke length and stroke rate
References Stroke length Stroke rate
Units Method Units Method
Biokinetic swim-bench training
Roberts
et al. [28]
NR Method NR Cycle.min
-1
Time taken to perform five stroke cycles per
minute
Traditional resistance training
Girold et al.
[32]
Metres All 50 m trials were recorded using a video
recorder. Stroke length was analysed using
the video software, in 10 m segments. The
mean of the total number of stroke cycles per
50 m was used to calculate stroke length
Cycle.min
-1
All 50 m trials were recorded using a video
recorder. Three complete stroke cycles per
each 25 m, from a 10 m segment. The mean
of the total number of stroke cycles per 50 m
was used
Strass [31] Metres Stroke length was calculated from the
following equation: velocity =stroke
length 9stroke frequency/60
Cycle.min
-1
Stroke frequency was calculated using a
stopwatch that was calibrated over four stroke
cycles
Aspenes
et al. [7]
Metres Calculated by using a speedometer by attaching
a line to the swimmer and analysing the data.
Stroke length was calculated when the right
hand entered the water
Hertz Calculated using a speedometer by attaching a
line to the swimmer and analysing the data.
Stroke rate was calculated when the right
hand entered the water
Girold et al.
[18]
Metres All 50 m trials were video recorded and
measured with picture digitiser software.
Stroke length over 10 m, every 25 m between
7.5 and 17.5 m
Cycle.min
-1
All 50 m trials video recorded and measured
with picture digitiser software. Three
complete stroke cycles, over 10 m, every
25 m between 7.5 and 17.5 m
Trappe and
Pearson
[19]
Metres The following equation was used:
velocity =stroke length 9stroke frequency/
60
Time to
complete
four stroke
cycles
Recorded during the last 83.6 m of the 91.4 m
race. Calculated by the time it took to
complete four stroke cycles and divided by 4
Tanaka et al.
[16]
Metres The following equation was used:
velocity =stroke length 9stroke frequency/
60
NR During the 365.8 m swim, stroke rate was
calculated by the time taken for four complete
stroke cycles at three time points
Manning
et al. [30]
Not
recorded
Not recorded Not recorded Not recorded
Song et al.
[29]
Not
recorded
Not recorded Not recorded Not recorded
Core training
Weston
et al. [33]
Not
recorded
Not recorded Not recorded Not recorded
Swim-specific resistance training
Girold et al.
[2]
Metres The following equation was used:
velocity =stroke length 9stroke frequency/
60
Cycle.min
-1
Stroke rate was calculated for three stroke
cycles, and was measured every 25 m
Konstantaki
et al. [34]
Not
recorded
Not recorded Not recorded Not recorded
Dragunas
et al. [35]
Metres An average distance per stroke was taken from
20 to 30 and 35 to 45 each 50 m, using a
video performance device, and distance per
stroke was then calculated
Strokes.min
-1
An average stroke rate was taken from 20 to
30 m and 35 to 45 m each 50 m, using a
video performance device, and stroke rate
was then calculated
Toussaint
and
Vervoorn
[36]
Not
recorded
Not recorded Not recorded Not recorded
NR not reported
E. Crowley et al.
123
Author's personal copy
factor determining swimming performance at higher
swimming velocities [68]. By contrast, Roberts et al. [28]
and Aspenes et al. [7] found a decrease in stroke rate, along
with a non-significant [7,28] increase in stroke length. This
suggests that by increasing stroke length there may be a
decrease in stroke rate, and by increasing stroke rate there
may be a decrease in stroke length. This is not uncommon
as Girold et al. [2] found a significant decrease in stroke
length with a significant increase in stroke rate. At maxi-
mum velocity, stroke rate is the most significant factor
[3,18,67]; however, as race distance increases to C400 m,
stroke rate becomes less of a determining factor [75–77],
but will increase due to a decrease in stroke length at
volitional exhaustion [1,77–80], with the fastest swimmers
maintaining stroke length [20,64,67,76]. Therefore, an
individualised combination of stroke rate and stroke length
needs to be established to ensure optimal performance, but
for now there is no evidence to support this.
6 Concurrent Training
The complex nature of high-performance swimming
training requires a large volume of swimming training with
the complementary aid of resistance training. This simul-
taneous integration of resistance and swimming training
within a periodised training programme is known as con-
current training [81,82]. Swimming training includes both
a low-intensity endurance training (LIT) component (in-
creases mitochondrial density, oxidative capacity of muscle
fibres, alterations in substrate metabolism and therefore the
athlete’s aerobic capacity [83]) and a high-intensity
endurance training (HIT) component (improves delivery of
oxygen to working muscles through increased stroke vol-
ume, improved heat tolerance, improved ability to produce
and utilise adenosine triphosphate (ATP), improved gly-
cogenolytic capacity and increased buffering capacity
[84]). This provides the essential physiological adaptations
needed to improve swimming performance. Regardless of
the positive benefits of LIT, HIT and resistance training, it
has been well-documented that concurrent training results
in conflicting adaptation responses [81,82,85].
Molecular physiologists refer to myofibrillar hypertro-
phy as the sole adaptation to resistance training within the
current literature, and do not acknowledge other neural
adaptations that contribute to the increased rate of force
production (i.e. musculotendinous stiffness, motor unit
recruitment, and intermuscular and intramuscular coordi-
nation) [86]. Conversely, in this review, neural adaptations
have been found to be primarily responsible for an increase
in swimming performance. Mammalian target of rapamy-
cin complex 1 (mTORC1) has been identified as a key
molecular pathway for resistance training that increases
mechanical stimuli, growth factors and nutrients to pro-
mote protein synthesis. Adenosine monophosphate-acti-
vated protein kinase (AMPK) and calmodulin-dependent
protein kinase II (CaMKII) are activated by endurance
training and promote intracellular concentrations of Ca
2?
,
oxygen, lactate and other physiological responses to
coordinate mitochondrial biogenesis, improved substrate
utilisation and capillary density to enhance oxidative
capacities [85]. It is believed that endurance training
inhibits mTORC1 and other regulators of muscle hyper-
trophy, with HIT interfering with the key molecular path-
ways associated with resistance training more than LIT.
HIT may compromise force and power production during
resistance training and theoretically limit the activation of
type II muscle fibres and units [87], thereby causing
increased residual fatigue. Endurance training inhibits
protein synthesis and stimulates protein breakdown, which
limits muscle hypertrophy by affecting the positive net
balance of protein synthesis needed for hypertrophy to take
place [88]. AMPK has a direct inhibitory effect on
mTORC1 and therefore negatively regulates protein syn-
thesis and upregulates protein degradation [89], with sirtuin
1 (SIRT1) emerging as another potential mechanism by
which mTORC1 and protein synthesis may be suppressed
[90]. Conversely, some studies have shown that both pro-
tein synthesis [91] and mTORC1 signalling [92] are not
affected by resistance training, with some studies showing
that concurrent training improves adaptations [93]. Wilson
et al. [82] hypothetically stated that greater hypertrophy
can be observed in concurrent training studies when
resistance and endurance training were performed on sep-
arate days. Lundberg et al. [94] found that significant time
between sessions resulted in elevated metabolic signalling.
Botonis et al. [95] reported similar results in waterpolo
players, finding that resistance training and endurance
training conducted on separate days resulted in improve-
ments in both strength and swimming performance. These
studies show that LIT has little, or any, effect on resistance-
training adaptations. Other factors that modulate the
molecular responses to resistance training include training
status, age, genetic factors, muscle phenotype [96,97],
training modalities and nutrient availability [85].
The current research shows conflicting results as
molecular signals produced from endurance training have
been shown to interfere with hypertrophic-related resis-
tance-training adaptations; however, significant time peri-
ods and specific training modalities show favourable
results. Baar [98] provides a series of scientific-based
recommendations to optimise concurrent training, as fol-
lows: (1) endurance training should be performed in the
morning, with at least 3 h between endurance and resis-
tance training to allow AMPK and SIRT1 levels to return
to baseline; (2) resistance training should be followed by a
Impact of Resistance Training on Swimming Performance
123
Author's personal copy
leucine-rich protein source to promote protein synthesis;
(3) athletes should fully refuel after endurance training as
AMPK can be activated by low glycogen levels and SIRT1
is activated by caloric restriction, which will affect adap-
tation responses to resistance training; and (4) it is optimal
to perform resistance training after LIT rather than HIT.
Future research should focus on the interference of
endurance training on resistance training in relation to
neural adaptations, rather than just myofibrillar hyper-
trophic adaptations.
7 Conclusions
The current literature shows that resistance training can
improve swimming performance in response to various
types of resistance-training programmes. The dynamic
correspondence theory [12] provides a practical set of
guidelines for designing a resistance-training programme
for swimming. This theory outlines that (1) joint angular
ranges should be similar; (2) weaknesses within the system
should be highlighted and improved; (3) overload is
essential (i.e. move moderate loads at high velocities or
high loads at similar velocities specific to swimming); (4)
resistance training should be specific to the time frame of
swimming events; (5) resistance training should be specific
to the contractions performed (i.e. concentric, eccentric,
isometric, etc.); and (6) exposure to complex interactions
such as multi-joint movement patterns should be similar to
movement patterns performed in swimming. Further to the
points above, Aspenes et al. [7] stated that technical
swimming post-resistance training will help the likelihood
of the transfer of resistance training to swimming
performance.
The findings of this review are as follows: for optimal
transfer, low-volume, high-velocity/force, swim-specific
resistance-training programmes showed a positive transfer
to swimming performance, with the lack of transfer from
arms-only swimming and the biokinetic swim bench due to
the isolation of the arms, suggesting that core training may
have a significant role to play in swimming performance
due to weaknesses within the system. In relation to the
effect of resistance-training modalities to technical
parameters, the literature is sparse and inconclusive but
trends show that for improving stroke length, low-volume,
high-velocity/force training is required due to increasing
stroke length demanding a high level of strength. Resisted
swims could present a viable training modality for
increasing stroke rate, as shown by Girold et al. [2]. We
have also suggested that landmark registration is a reliable
and valid method to avoid stroke length and stroke rate
inconsistencies within the data. Identification of suit-
able strength tests for swimmers requires further
investigation as a gold-standard test has yet to be docu-
mented in the current literature. Use of the isokinetic
machine and tethered swims seems to correlate well, both
with an increase in strength of the arm extensors in the
concentric condition and with swimming performance,
respectively. It is therefore clear that the development of a
wide range of functional resistance modalities will facili-
tate the transfer of resistance training to swimming
performance.
The lack of high-quality methodological studies using
elite swimmers makes the literature hard to interpret. It is
important to note that a thorough documentation of methods
and results is crucial for accurate interpretations. Specific
attention should be given to intervention duration, use of
effect sizes, timing systems and pool size used as these can
have an effect on outcome measures. Some studies outside
the scope of this review documented their training inter-
ventions in a clear and concise manner [48,49,57] and this
is needed for studies of this nature. Future research needs to
focus on long-term intervention periods, with full docu-
mentation of swimming training programmes and periodis-
ation plans, and use of the dynamic correspondence theory
[12], to provide a practical set of guidelines for designing the
resistance-training intervention.
Compliance with Ethical Standards
Funding No sources of funding were used to assist in the preparation
of this article.
Conflicts of interest Emmet Crowley, Andrew J. Harrison and Mark
Lyons declare that they have no conflicts of interest relevant to the
content of this review.
References
1. Wakayoshi K, D’Acquisto J, Cappaert J, et al. Relationship
between metabolic parameters and stroking technique character-
istics in front crawl. In: Troup JP, Hollander AP, Strasse D,
Trappe SW, Cappaert JM, Trappe TA, editors. Biomechanics and
medicine in swimming VII. London: Taylor & Francis; 1996.
p. 152–8.
2. Girold S, Calmels P, Maurin D, et al. Assisted and resisted sprint
training in swimming. J Strength Cond Res. 2006;20(3):547–54.
3. Wakayoshi K, D’Acquisto L, Cappaert J, et al. Relationship
between oxygen uptake, stroke rate and swimming velocity in
competitive swimming. Int J Sports Med. 1995;16(01):19–23.
4. Gonza
´lez-Boto R, Salguero A, Tuero C, et al. Monitoring the
effects of training load changes on stress and recovery in swim-
mers. J Physiol Biochem. 2008;64(1):19–26.
5. Aspenes ST, Karlsen T. Exercise-training intervention studies in
competitive swimming. Sports Med. 2012;42(6):527–43.
6. Morouc¸o PG, Marinho DA, Amaro NM, et al. Effects of dry-land
strength training on swimming performance: a brief review.
J Hum Sport Exerc. 2012;7(2):553–9.
7. Aspenes S, Kjendlie P-L, Hoff J, et al. Combined strength and
endurance training in competitive swimmers. J Sport Sci Med.
2009;8(3):357–65.
E. Crowley et al.
123
Author's personal copy
8. Garrido N, Marinho DA, Reis VM, et al. Does combined dry land
strength and aerobic training inhibit performance of young
competitive swimmers. J Sport Sci Med. 2010;9(2):300–10.
9. Saltin B, Gollnick P. Skeletal muscle adaptability: significance
for metabolism and performance. In: Peachey LD, Adrain RH,
Geiger SR, editors. Handbook of Physiology. Baltimore: Wil-
liams and Wilkins; 1983. p. 555–631.
10. Newton RU, Cormie P, Cardinale M. Principles of athletic
training. In: Cardinale M, Newton R, Nosaka K, editors. Strength
and conditioning: biological principles and practical applications.
Oxford: Wiley-Blackwell; 2011. p. 255–70.
11. Haff GG, Nimphius S. Training principles for power. J Strength
Cond Res. 2012;34(6):2–12.
12. Goodwin JE, Cleather DJ. The biomechanical principles under-
pinning strength and conditioning. In: Jefferys I, Moody J, edi-
tors. Strength and conditioning for sports performance. Oxon:
Routledge; 2016. p. 62–5.
13. Suchomel TJ, Nimphius S, Stone MH. The importance of mus-
cular strength in athletic performance. Sports Med.
2016;46(10):1–31.
14. Sharp RL, Troup JP, Costill DL. Relationship between power and
sprint freestyle swimming. Med Sci Sports Exerc.
1981;14(1):53–6.
15. Costill D, Rayfield F, Kirwan J, et al. A computer based system
for the measurement of force and power during front crawl
swimming. J Swim Res. 1986;2(1):16–9.
16. Tanaka H, Costill DL, Thomas R, et al. Dry-land resistance
training for competitive swimming. Med Sci Sports Exerc.
1993;25(8):952–9.
17. Tanaka H, Swensen T. Impact of resistance training on endurance
performance. Sports Med. 1998;25(3):191–200.
18. Girold S, Maurin D, Dugue
´B, et al. Effects of dry-land vs.
resisted-and assisted-sprint exercises on swimming sprint per-
formances. J Strength Cond Res. 2007;21(2):599–605.
19. Trappe SW, Pearson DR. Effects of weight assisted dry-land
strength training on swimming performance. J Strength Cond
Res. 1994;8(4):209–13.
20. Smith DJ, Norris SR, Hogg JM. Performance evaluation of
swimmers. Sports Med. 2002;32(9):539–54.
21. Toussaint HM, Beek PJ. Biomechanics of competitive front crawl
swimming. Sports Med. 1992;13(1):8–24.
22. Hollander AP, de Groot G, van Ingen Schenau GJ, et al. Con-
tribution of the legs to propulsion in front crawl swimming. In:
Ungerechts BE, Wilke K, Reischle K, editors. Swimming science
V. International Series of Sport Sciences, vol. 18. Champaign:
Human Kinetics; 1988. p. 39–44.
23. Bucher W. The influence of the leg kick and the arm stroke on the
total speed during the crawl stroke. In: Lewillie L, Clarys JP,
editors. Swimming II. Balitimore: University Park Press; 1975.
p. 180–7.
24. Deschodt V, Arsac L, Rouard A. Relative contribution of arms
and legs in humans to propulsion in 25-m sprint front-crawl
swimming. Eur J Appl Physiol Occup Physiol. 1999;80(3):192–9.
25. Zamparo P, Pendergast D, Mollendorf J, et al. An energy balance
of front crawl. Eur J Appl Physiol. 2005;94(1–2):134–44.
26. Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting items
for systematic reviews and meta-analyses: the PRISMA state-
ment. Ann Intern Med. 2009;151(4):264–9.
27. Maher CG, Sherrington C, Herbert RD, et al. Reliability of the
PEDro scale for rating quality of randomized controlled trials.
Phys Ther. 2003;83(8):713–21.
28. Roberts AJ, Termin B, Reilly M, et al. Effectiveness of biokinetic
training on swimming performance in collegiate swimmers.
J Swim Res. 1991;7(3):5–11.
29. Song H-S, Park D-H, Jung D-S. The effect of periodized strength
training application on the Korea national team. Int J Appl Sports
Sci. 2009;21(2):122–45.
30. Manning J, Dooly-Manning C, Terrell D, et al. Effects of a power
circuit weight training programme on power production and
performance. J Swim Res. 1986;2(1):24–9.
31. Strass D. Effects of maximal strength training on sprint perfor-
mance of competitive swimmers. In: Ungerechts B, Wilke K,
Reischle K, editors. Swimming science. V International Series on
Sport Sciences, vol. 18. Champaign: Human Kinetics; 1988.
p. 149–56.
32. Girold S, Jalab C, Bernard O, et al. Dry-land strength training vs.
electrical stimulation in sprint swimming performance. J Strength
Cond Res. 2012;26(2):497–505.
33. Weston M, Hibbs AE, Thompson KG. Isolated core training
improves sprint performance in national-level junior swimmers.
Int J Sports Physiol Perform. 2015;10(2):204–10.
34. Konstantaki M, Winter E, Swaine I. Effects of arms-only
swimming training on performance, movement economy, and
aerobic power. Int J Sports Physiol Perform. 2008;3(3):294–304.
35. Dragunas AJ, Dickey JP, Nolte VW. The effect of drag suit
training on 50-m freestyle performance. J Strength Cond Res.
2012;26(4):989–94.
36. Toussaint HM, Vervoorn K. Effects of specific high resistance
training in the water on competitive swimmers. Int J Sports Med.
1990;11(03):228–33.
37. Fry AC, Kraemer WJ. Physical performance characteristics of
American collegiate football players. J Strength Cond Res.
1991;5(3):126–38.
38. Hetzler RK, Stickley CD, Lundquist KM, et al. Reliability and
accuracy of handheld stopwatches compared with electronic
timing in measuring sprint performance. J Strength Cond Res.
2008;22(6):1969–76.
39. Pichon F, Chatard J-C, Martin A, et al. Electrical stimulation and
swimming performance. Med Sci Sports Exerc.
1995;27(12):1671–6.
40. Maglischo E, Maglischo C, Zier D, et al. The effect of sprint-
assisted and sprint-resisted swimming on stroke mechanics.
J Swim Res. 1985;1(2):27–33.
41. McGuigan MR, Winchester JB. The relationship between iso-
metric and dynamic strength in college football players. J Sports
Sci Med. 2008;7(1):101–5.
42. Bollens E, Annemans L, Vaes W, et al. Peripheral EMG com-
parison between fully tethered and free front crawl swimming. In:
Ungerechts B, Wilke K, Reischle K, editors. Swimming science
V. International Series of Sport Sciences, vol. 18. Champaign:
Human Kinetics; 1988. p. 173–81.
43. Morouc¸ o P, Keskinen KL, Vilas-Boas JP, et al. Relationship
between tethered forces and the four swimming techniques per-
formance. J Appl Biomech. 2011;27(2):161–9.
44. Shimonagata S, Taguchi M, Miura M. Effect of swimming
power, swimming power endurance and dry-land power on 100 m
freestyle performance. In: Chatard JC, editor. Biomechanics and
medicine in swimming IX. Saint Etienne: University of Saint-
Etienne; 2003. pp. 391–6.
45. Clarys J. Hydrodynamics and electromyography: ergonomics
aspects in aquatics. Appl Ergon. 1985;16(1):11–24.
46. Coyle EF, Feiring DC, Rotkis TC, et al. Specificity of power
improvements through slow and fast isokinetic training. Appl
Physiol. 1981;51:1437–42.
47. Kanehisa H, Miyashita M. Specificty of velocity in strength
training. Euro J Appl Physiol. 1983;52:104–6.
48. Sadowski J, Mastalerz A, Gromisz W, et al. Effectiveness of the
power dry-land training programmes in youth swimmers. J Hum
Kinet. 2012;32:77–86.
Impact of Resistance Training on Swimming Performance
123
Author's personal copy
49. Ramos Veliz R, Requena B, Suarez-Arrones L, et al. Effects of
18-week in-season heavy-resistance and power training on
throwing velocity, strength, jumping, and maximal sprint swim
performance of elite male water polo players. J Strength Cond
Res. 2014;28(4):1007–14.
50. Gollnick PD. Relationship of strength and endurance with
skeletal muscle structure and metabolic potential. Int J Sports
Med. 1982;3:26–32.
51. Dingley AA, Pyne DB, Youngson J, et al. Effectiveness of a dry-
land resistance training programme on strength, power, and
swimming performance in paralympic swimmers. J Strength
Cond Res. 2015;29(3):619–26.
52. Stewart AM, Hopkins WG. Seasonal training and performance of
competitive swimmers. J Sports Sci. 2000;18(11):873–84.
53. Costill D. Training adaptations for optimal performance. In:
Keskinen KL, Komi PV, Hollander AP, editors. Biomechanics
and medicine in swimming VIII. Jyva
¨skyla
¨: University of
Jyva
¨skyla
¨; 1999. p. 381–90.
54. Delecluse C, Van Coppenolle H, Willems E, et al. Influence of
high resistance and high velocity training on sprint performance.
Med Sci Sports Exerc. 1995;27:1203–9.
55. Maglischo EW, Maglischo CW, Zier DJ, et al. The effect of
sprint-assisted and sprint-ressisted swimming on stroke mechan-
ics. J Swimming Res. 1985;1:27–33.
56. Jua
´rez Santos-Garcı
´a D, Gonza
´lez-Rave
´JM, Legaz Arrese A,
et al. Acute effects of two resisted exercises on 25 m swimming
performance. Isokinet Exerc Sci. 2013;21(1):29–35.
57. De Villarreal ES, Suarez-Arrones L, Requena B, et al. Enhancing
performance in professional water polo players: dryland training,
in-water training, and combined training. J Strength Cond Res.
2015;29(4):1089–97.
58. Ribeiro J, Figueiredo P, Sousa A, et al.
_
VO
2
kinetics and meta-
bolic contributions during full and upper body extreme swimming
intensity. Eur J Appl Physiol. 2015;115(5):1117–24.
59. Rodrı
´guez F, La
¨tt E, Ju
¨rima
¨e J, et al. VO
2
kinetics in all-out arm
stroke, leg kick and whole stroke front crawl 100-m swimming.
Int J Sports Med. 2016;37(03):191–6.
60. Morris KS, Osborne MA, Shephard ME, et al. Velocity, aerobic
power and metabolic cost of whole body and arms only front
crawl swimming at various stroke rates. Eur J Appl Physiol.
2016;116(5):1075–85.
61. McGowan CJ, Pyne DB, Raglin JS, et al. Current warm-up
practices and contemporary issues faced by elite swimming
coaches. J Strength Cond Res. 2016;30(12):3471–80.
62. Telles T, Barbosa AC, Campos MH, et al. Effect of hand paddles
and parachute on the index of coordination of competitive crawl-
strokers. J Sports Sci. 2011;29(4):431–8.
63. Clarys JP, Toussaint HM, Bollens E, et al. Muscular specificity
and intensity in swimming against a mechanical resistance: sur-
face EMG in MAD-and free swimming. In: Ungerechts BE,
Reischle K, Wilke K, editors. Swimming V. Champaign: Human
Kinetics; 1988. p. 191–9.
64. Figueiredo P, Zamparo P, Sousa A, et al. An energy balance of
the 200 m front crawl race. Eur J Appl Physiol.
2011;111(5):767–77.
65. Chatard JC, Mujika I. Training load and performance in swim-
ming. In: Keskinen KL, Komi PV, Hollander AP, editors.
Biomechanics and medicine in swimming VIII. Jyva
¨skyla
¨:
University of Jyva
¨skyla
¨; 1999. p. 429–34
66. Costill D, Sharp R, Troup J. Muscle strength: contributions to
sprint swimming. Swim World. 1980;21:29–34.
67. Craig AB, Skehan PL, Pawelczyk JA, et al. Velocity, stroke rate,
and distance per stroke during elite swimming competition. Med
Sci Sports Exerc. 1985;17(6):625–34.
68. Hay J, Guimaraes A, Grimston S. A quantitative look at swim-
ming biomechanics. Swimming Technique. 1983;20(2):11–7.
69. Craig AB, Pendergast DR. Relationships of stroke rate, distance
per stroke, and velocity in competitive swimming. Med Sci
Sports Exerc. 1979;11:278–83.
70. Schnitzler C, Seifert L, Chollet D. Arm coordination and per-
formance level in the 400-m front crawl. Res Q Exerc Sport.
2011;82(1):1–8.
71. Laffite LP, Vilas-Boas JP, Demarle A, et al. Changes in physi-
ological and stroke parameters during a maximal 400-m free
swimming test in elite swimmers. Can J Appl Physiol.
2004;29(1):17–31.
72. Alberty M, Potdevin F, Dekerle J, et al. Changes in swimming
technique during time to exhaustion at freely chosen and con-
trolled stroke rates. J Sports Sci. 2008;26(11):1191–200.
73. Huot-Marchand F, Nesi X, Sidney M, et al. Is improvement in
performance linked to higher stroke length values in top-level
100-m front crawl swimmers? J Hum Mov Stud. 2005;6(1):12–8.
74. Deluzio KJ, Harrison AJ, Coffey N, et al. Analysis of biomech-
nical waveform data. In: Robertson DGE, Caldwell GE, Hamill J,
Kamen G, Whittlesey SN, editors. Research methods in biome-
chanics. 2nd ed. Champaign: Human Kinetics; 2014. p. 317–37.
75. Fernandes RJ, Marinho DA, Barbosa TM, et al. Is time limit at
the minimum swimming velocity of VO
2
max influenced by
stroking parameters? Percept Mot Skills. 2006;103(1):67–75.
76. Laffite LP, Vilas-Boas JP, Demarle A, et al. Changes in physi-
ological and stroke parameters during a maximal 400-m free
swimming test in elite swimmers. Can J Appl Physiol.
2004;29(S1):S17–31.
77. Alberty M, Sidney M, Pelayo P, et al. Stroking characteristics
during time to exhaustion tests. Med Sci Sports Exerc.
2009;41(3):637.
78. Marinho DA, Boas JPV, Keskinen KL, et al. Behaviour of the
kinematic parameters during a time to exhaustion test at VO
2max
in elite swimmers. J Hum Movement Stud. 2006;51:1–10.
79. Keskinen KL, Komi PV. Stroking characteristics of front crawl
swimming during exercise. J Appl Biomech. 1993;9(3):219–26.
80. Fernandes R, Vilas-Boas J. Time to exhaustion at the VO
2max
velocity in swimming: a review. J Hum Kinet. 2012;32:121–34.
81. Leveritt M, Abernethy PJ, Barry BK, et al. Concurrent strength
and endurance training. A review. Sports Med.
1999;28(6):413–27.
82. Wilson JM, Marin PJ, Rhea MR, et al. Concurrent training: a
meta-analysis examining interference of aerobic and resistance
exercises. J Strength Cond Res. 2012;26(8):2293–307.
83. Hawley JA. Adaptations of skeletal muscle to prolonged, intense
endurance training. Clin Exp Pharmacol Physiol.
2002;29(3):218–22.
84. Laursen PB, Jenkins DG. The scientific basis for high-intensity
interval training. Sports Med. 2002;32(1):53–73.
85. Fyfe JJ, Bishop DJ, Stepto NK. Interference between concurrent
resistance and endurance exercise: molecular bases and the role
of individual training variables. Sports Med. 2014;44(6):743–62.
86. Cormie P, McGuigan MR, Newton RU. Developing maximal
neuromuscular power. Sports Med. 2011;41(1):17–38.
87. Sale DG. Influence of exercise and training on motor unit acti-
vation. Exerc Sport Sci Rev. 1987;15:95–151.
88. Hawley JA. Molecular responses to strength and endurance
training: are they incompatible? Appl Physiol Nutr Metab.
2009;34(3):355–61.
89. Coffey VG, Hawley JA. The molecular bases of training adap-
tation. Sports Med. 2007;37(9):737–63.
90. Ghosh HS, McBurney M, Robbins PD. SIRT1 negatively regu-
lates the mammalian target of rapamycin. PLoS One.
2010;5(2):e9199.
91. Donges CE, Burd NA, Duffield R, et al. Concurrent resistance
and aerobic exercise stimulates both myofibrillar and
E. Crowley et al.
123
Author's personal copy
mitochondrial protein synthesis in sedentary middle-aged men.
J Appl Physiol. 2012;112(12):1992–2001.
92. Apro W, Wang L, Ponten M, et al. Resistance exercise induced
mTORC1 signalling is not impaired by subsequent endurance
exercise in human skeletal muscle. Am J Physiol Endocrinol
Metab. 2013;305:22–32.
93. Wang L, Mascher H, Psilander N, et al. Resistance exercise
enhances the molecular signalling of mitochondrial biogenesis
induced by endurance exercise in human skeletal muscle. J Appl
Physiol. 2011;111(5):1335–44.
94. Lundberg TR, Fernandez-Gonzalo R, Gustafsson T, et al. Aerobic
exercise alters skeletal muscle molecular responses to resistance
exercise. Med Sci Sports Exerc. 2012;44(9):1680–8.
95. Botonis PG, Toubekis AG, Platanou TI. Concurrent strength and
interval endurance training in elite water polo players. J Strength
Cond Res. 2016;30(1):126–33.
96. Coffey VG, Shield A, Canny BJ, et al. Interaction of contractile
activity and training history on mRNA abundance in skeletal
muscle from trained athletes. Am J Physiol Endocrinol Metab.
2006;290(5):E849–55.
97. Coffey VG, Zhong Z, Shield A, et al. Early signalling responses
to divergent exercise stimuli in skeletal muscle from well-trained
humans. FASEB J. 2006;20(1):190–2.
98. Baar K. Using molecular biology to maximize concurrent train-
ing. Sports Med. 2014;44(2):117–25.
Impact of Resistance Training on Swimming Performance
123
Author's personal copy
A preview of this full-text is provided by Springer Nature.
Content available from Sports Medicine
This content is subject to copyright. Terms and conditions apply.
