Content uploaded by Daniel John Glassbrook
Author content
All content in this area was uploaded by Daniel John Glassbrook on Nov 20, 2017
Content may be subject to copyright.
THE HIGH-BAR AND LOW-BAR BACK-SQUATS:A
BIOMECHANICAL ANALYSIS
DANIEL J. GLASSBROOK,
1
SCOTT R. BROWN,
1
ERIC R. HELMS,
1
SCOTT DUNCAN,
1
AND
ADAM G. STOREY
1,2
1
Sports Performance Research Institute New Zealand (SPRINZ), Auckland University of Technology, Auckland, New
Zealand; and
2
High Performance Sport New Zealand (HPSNZ), Auckland, New Zealand
ABSTRACT
Glassbrook, DJ, Brown, SR, Helms, ER, Duncan, S, and
Storey, AG. The high-bar and low-bar back-squats: a bio-
mechanical analysis. J Strength Cond Res XX(X): 000–000,
2017—No previous study has compared the joint angle and
ground reaction force (vertical force [Fv]) differences
between the high-bar back-squat (HBBS) and low-bar back-
squat (LBBS) above 90% 1 repetition maximum (1RM). Six
male powerlifters (POW) (height: 179.2 67.8 cm; body-
weight: 87.1 68.0 kg; age: 21–33 years) of international
level, 6 male Olympic weightlifters (OLY) (height: 176.7 6
7.7 cm; bodyweight: 83.1 613 kg; age: 22–30 years) of
national level, and 6 recreationally trained male athletes
(height: 181.9 68.7 cm; bodyweight: 87.9 615.3 kg;
age: 23–33 years) performed the LBBS, HBBS, and both
LBBS and HBBS (respectively) up to and including 100%
1RM. Small to moderate (d= 0.2–0.5) effect size differences
were observed between the POW and OLY in joint angles
and Fv, although none were statistically significant. However,
significant joint angle results were observed between the
experienced POW/OLY and the recreationally trained group.
Our findings suggest that practitioners seeking to place
emphasis on the stronger hip musculature should consider
the LBBS. Also, when the goal is to lift the greatest load
possible, the LBBS may be preferable. Conversely, the HBBS
is more suited to replicate movements that exhibit a more
upright torso position, such as the snatch and clean, or to
place more emphasis on the associated musculature of the
knee joint.
KEY WORDS joint angles, ground reaction forces, powerlifting,
Olympic weightlifting
INTRODUCTION
The squat is one of the most common exercises in
strength and conditioning. The movement is
widely accepted as valid and reliable for the assess-
ment and improvement of lower extremity/trunk
strength, function, and resilience to injury (4,9,10) and an
effective exercise in injury rehabilitation (19). These benefits
are possible through the contributions of the quadriceps,
hamstrings, gluteal, triceps surae, and lumbar erector muscle
groups to the completion of the movement (9,25). In fact, it
is predicted that more than 200 muscles are active through-
out the completion of a single repetition (31,36). The squat
itself is in essence a simple movement, despite the great
number of active muscles throughout. In strength and con-
ditioning, load can be applied to the squat movement via
several methods, for example dumbbells, kettlebells, and
a range of other weighted implements. However, perhaps
most commonly, load is applied via a barbell and in one of
2 ways: (a) as a front-squat, where a barbell is placed ante-
riorly on the shoulder and (b) as a back-squat, where the
barbell is placed posteriorly to the shoulder and across the
trapezius musculature (16). The focus of this article will be
the back-squat.
There are 2 different variations of the back-squat, differ-
entiated by the placement of the barbell on the trapezius
musculature. The traditional “high-bar” back-squat (HBBS)
is performed with the barbell placed across the top of the
trapezius just below the process of the C7 vertebra and is
commonly used by Olympic weightlifters (OLY) to simulate
the catch position of the Olympic weightlifting competition
lifts, the snatch and clean and jerk (41). Conversely, the
“low-bar” back-squat (LBBS) places the barbell on the lower
trapezius, just over the posterior deltoid and along the spine
of the scapula (41). The LBBS is commonly used in com-
petitive powerlifting (where the back-squat is 1 of the 3
competition lifts), as it may enable higher loads to be lifted
(32). This could be due to the maximization of posterior
displacement of the hips and increased force through the
hip joints in comparison with the knee joints (37). The differ-
ences in bar position between the HBBS and LBBS result in
an altered center of mass. Therefore, movement strategies
result to maintain the bodies’ center of mass within its base
Address correspondence to Daniel J. Glassbrook, daniel.glassbrook@
aut.ac.nz.
00(00)/1–18
Journal of Strength and Conditioning Research
2017 National Strength and Conditioning Association
VOLUME 00 | NUMBER 00 | MONTH 2017 | 1
Copyright ª2017 National Strength and Conditioning Association
of support (BOS). These movement strategies may manifest
as: changes in (a) joint angles of the lower extremity kinetic
chain and (b) ground reaction forces (vertical force [Fv]).
When comparing the HBBS with LBBS, several differ-
ences present themselves. In powerlifting, there are compe-
tition regulations that each lifter must comply with in order
for each lift to count toward their competition total (21).
One such regulation is for sufficient “depth” to be reached
in the squat. That is, there must be sufficient flexion of the
knees and lowering of the hips toward the ground, so that
“the top surface of the legs at the hip joint is lower than the
top of the knees” (21). In comparison, the HBBS is not
directly included as a competition lift in Olympic weightlift-
ing. Therefore, in training, OLY typically squat to a depth
that replicates the final catch position of the snatch and
clean and jerk. This often manifests as a deeper squat posi-
tion than powerlifting regulation depth, characterized by
greater flexion at the hip, knee, and ankle joints. Previous
research has shown that the angle at peak knee flexion is
generally smaller in the HBBS (e.g., 70–908), in comparison
with the LBBS (e.g., 100–1208) (5,11,13,17,18,20,24,27,37,38).
Interestingly, some studies have reported the reverse
(17,24,37). These conflicting results (although not explicitly
stated by the authors) are likely to be the raw joint angles
and not the actual angle (Figure 1).
Moreover, previous research specifically comparing the
HBBS with the LBBS shows that the LBBS is defined by
a smaller absolute trunk angle, and therefore greater forward
lean to maintain the barbell over the center of mass (2,14,41).
The unique position of the LBBS results in (a) a decreased
trunk lever arm when placing the bar lower on the back, (b)
a greater emphasis on the stronger musculature of the hip
rather than the musculature of the knee joint, and (c) an
increase in stability and a potential decrease in stress placed
on the lumbar region and ankle, when compared with the
HBBS (34,37). These factors may contribute to understand-
ing why the LBBS typically allows for greater loads to be
lifted. However, these kinematic findings are not definitive,
and there are mixed results in the literature for the size of
HBBS and LBBS trunk angles at peak hip flexion
(5,11,13,17,20,24,27,29,37). Similarly, no conclusive differen-
ces between the HBBS and LBBS ankle joint angles can be
drawn, in reference to previous literature (13,17,24,34,37).
As the position of the barbell on the trapezius influences
the joint angles of the back-squat, there is also a resultant
influence on the Fv produced. The position of the upper
body (i.e., hip joint angle) has a large impact on the location
and magnitude of the resultant Fv because of its larger mass.
Because of the LBBS tending to allow for greater loads to be
lifted, it would be expected that the Fv produced would be
greater than that with the HBBS. However, the 2 studies
that have specifically compared the Fv profiles of the HBBS
and LBBS provide contradictory results to this expectation
(15,37). The results of these 2 studies may indicate that,
although the LBBS typically allows for greater load to be
lifted through apparent mechanical advantages such as
Figure 1. Actual and raw joint angles of the hip, knee, and ankle: taken
from (A) the left end of the barbell, (B) the right end of the barbell, (C)
acromion process, (D) greater trochanter, (E) lateral epicondyle of the
femur, (F) lateral malleolus, (G) the top of the heal lift of the lifting shoe,
and (H) the base of the fifth metatarsal.
Figure 2. Representation of the order of familiarization and testing dates for the comparison group.
High-Bar vs. Low-Bar Squats
2
Journal of Strength and Conditioning Research
the
TM
Copyright ª2017 National Strength and Conditioning Association
a decreased trunk lever arm, these mechanical advantages
are not effectively displayed by Fv. Furthermore, the results
of these studies specifically may have arisen because of the
level of expertise of the participant with performing the
LBBS, as the authors chose to target the HBBS in recruit-
ment as the focus for expertise. Therefore, further research is
warranted to understand the Fv differences between the
HBBS and LBBS, in particular with loads greater than
90% 1 repetition maximum (1RM).
The existing literature provides some insight into the
kinematic and kinetic differences between the HBBS and
LBBS. However, there is no consensus as to the differences
between the 2 back-squat barbell positional variations. At
present, no previous study has compared the joint angles and
Fv of the HBBS and LBBS above 90% 1RM, and some
results may have been confounded by inadequate familiar-
ization. Thus, the purpose of this study was to compare and
contrast the differences in joint angles and Fv of the HBBS
and LBBS, up to and including maximal effort, in an effort to
create a full profile of the 2 bar back-squat variations in
groups both well versed and newly introduced to these
movements. The results of this investigation will add to
the current body of knowledge of Olympic weightlifting
and powerlifting practice alike and provide an understanding
of why the LBBS may allow for a greater load to be lifted.
METHODS
Experimental Approach to the Problem
To determine why the LBBS may allow for greater loads
to be lifted than the HBBS, both squat styles were
performed by experienced and inexperienced lifters. The
HBBS was performed by experienced OLY, and the LBBS
by experienced powerlifters (POW), up to and including
100% of 1RM. Recreationally trained athletes served as
a comparison group and performed both the HBBS and
LBBS. It is assumed that the experienced OLY and POW
have a better technique than the recreationally trained
athlete; however, it is important to acknowledge this may
not be strictly true in practice. A profile of each squat was
created through analysis of kinematic joint angles and
kinetic Fv differences.
Subjects
Six male POW (height: 179.2 67.8 cm; bodyweight: 87.1 6
8.0 kg; age: 21–33 years) of international (i.e., Oceania cham-
pionships) level volunteered to participate in the LBBS
group. In addition, 6 male OLY (height: 176.7 67.7 cm;
bodyweight: 83.1 613 kg; age: 22–30 years) who had pre-
viously qualified for national championship–level competi-
tion volunteered to participate in the HBBS group. All POW
routinely performed the LBBS in training and competition,
and all OLY routinely performed the HBBS in training.
Finally, 6 recreationally trained male athletes (height:
181.9 68.7 cm; bodyweight: 87.9 615.3 kg; age: 23–33
years) volunteered as a comparison group, and each
TABLE 1. Mean loads lifted across all %1RM ranges.*†
% Range Variable OLY POW HBCOM LBCOM
OLY vs. POW
Difference; 690% CI
HBCOM vs. LBCOM
Difference; 690% CI
BW (kg) 83.2 613.0 87.1 68.0 87.9 615.3 87.9 615.3
74–83 Load (kg) 136.6 623.5 140.9 620.1 99.9 613.2 103.0 616.2 12.5 623.8 4.0 67.6
BW 1.6 60.2 1.6 60.3 1.2 60.2 1.2 60.2 0.1 60.3 0.1 60.1
84–93 Load (kg) 152.5 623.1 159.2 621.8 116.4 612.9 121.7 618.8 9.4 626.6 6.0 69.5
BW 1.8 60.2 1.9 60.4 1.3 60.2 1.4 60.2 0.0 60.3 0.1 60.1
94–99 Load (kg) 164.0 624.7 174.6 620.1 128.7 612.4 136.5 621.6 7.2 624.2 7.9 68.0
BW 2.0 60.2 2.0 60.4 1.5 60.2 1.6 60.2 0.0 60.3 0.1 60.1
100 Load (kg) 169.5 626.5 181.2 621.8 135.2 611.1 143.4 620.7 11.8 625.4 8.2 611.1
BW 1.9 60.3 2.1 60.4 1.6 60.2 1.6 60.2 0.1 60.3 0.1 60.1
*1RM = 1 repetition maximum; BW = times body weight; CI = confidence interval; HBCOM = comparison high-bar back-squat; LBCOM = comparison low-bar back-squat; OLY =
Olympic weightlifters; POW = powerlifters.
†All data are presented as mean 6SD.
Journal of Strength and Conditioning Research
the
TM
|
www.nsca.com
VOLUME 00 | NUMBER 00 | MONTH 2017 | 3
Copyright ª2017 National Strength and Conditioning Association
participant was required to perform both the LBBS and
HBBS in a randomized order, after 2 familiarization sessions
with both types of squat. All participants were free of injury
and had $1 year’s strength training experience (POW: 5.05
64.56 years; OLY: 3.75 62.72 years; recreational: 8.67 6
3.5 years) consisting of $3 training sessions per week for the
POW and OLY. The comparison group volunteers were
required to train the back-squat in $1 training sessions per
week. Because of small participant numbers (n= 6 for each
group), the results of this study may not provide a full rep-
resentation of the differences between each squat type. Some
differences may be due to sampling error.
Before testing, written informed consent was received
from each participant, and all testing conditions were
examined and approved by the Auckland University of
Technology Ethics Committee (14/398).
Procedures
Powerlifters and Olympic Weightlifters. The POW and OLY
were required to attend only one session of approximately
3 hours in duration. A full
“level 2” anthropometric
assessment was performed on
all athletes by an experienced
International Society for the
Advancement of Kinanthrop-
ometry anthropometrist, fol-
lowed by an LBBS 1RM test
for the POW, and a HBBS
1RM test for the OLY.
Comparison Group. The recrea-
tionally trained athletes were
required to attend 4 separate
sessions over the course of 1
week: 2 guided 1-hour famil-
iarization sessions, 1 personal
familiarization session, and 1
3-hour–long testing session
(Figure 2). The first familiariza-
tion session comprised of the 1RM testing protocol for HBBS
and LBBS with loads up to 60% of self-reported or predicted
1RM. Self-reported 1RM values (performed within the last 6
months) for either back-squat variation were used to estimate
load progressions. Pilot testing determined that the load of
the unknown back-squat variation would be around 90% of
the known back-squat 1RM regardless of which squat style
was routinely performed. Thus, the loads for the familiariza-
tion session were estimated from 1 known 1RM for 1 back-
squat variation and a predicted 1RM at 90% of the known
1RM. The second familiarization session was performed 2
days later and comprised the same HBBS and LBBS protocol
in the same order as the first familiarization session, up to 80%
1RM of the self-reported and predicted 1RM for either back-
squat variation.
In both the first and second familiarization sessions for
each participant, the resistance exercise–specific rating of
perceived exertion (RPE) scale (43) was used to ensure
that intensity and predicted attempt weight values were
correct. In the first familiarization session, an RPE value of
3 or less (i.e., “light to little
effort”) was expected to be
reported in line with the
percentages of the 1RM (50
and 60%). If this was not
achieved, the predicted
weight values were changed
for the second familiarization
session. In the second famil-
iarization session, the same
RPE values of 3 or less were
employed for the 50 and 60%
of predicted 1RM sets. After
that, a self-reported RPE of 5
or less (i.e., “light effort with
TABLE 2. Mean loads lifted effect sizes and percentage differences.*
% Range Variable
OLY vs. POW HBCOM vs. LBCOM
Effect size % Difference Effect size % Difference
74–83 Load (kg) 0.3†3.2 0.3†3.0
BW 0.1 0.2 0.3†2.5
84–93 Load (kg) 0.2†4.4 0.3†4.4
BW 0.0 0.8 0.4†3.9
94–99 Load (kg) 0.2†6.5 0.5z5.7
BW 0.0 2.5 0.6z5.1
100 Load (kg) 0.2†6.9 0.4†5.7
BW 0.1 3.0 0.5z5.2
*BW = times body weight; HBCOM = comparison high-bar back-squat; LBCOM = com-
parison low-bar back-squat; OLY = Olympic weightlifters; POW = powerlifters.
†Small effect d$0.2.
zModerate effect d$0.5.
TABLE 3. Distance of COP to bar results.*†
% Range OLY (mm) POW (mm) HBCOM (mm) LBCOM (mm)
74–83 219 642 244 631 260 645 257 618
84–93 220 640 258 639 251 642 272 625
94–99 223 629 246 631 258 635 259 638
100 224 640 274 652 239 649 251 618
*COP = center of pressure; HBCOM = comparison high-bar back-squat; LBCOM =
comparison low-bar back-squat; OLY = Olympic weightlifters; POW = powerlifters.
†Negative numbers represent the bar a distance behind the COP. All COP data are
presented as mean 6SD.
High-Bar vs. Low-Bar Squats
4
Journal of Strength and Conditioning Research
the
TM
Copyright ª2017 National Strength and Conditioning Association
TABLE 4. Kinematic results.*†
%1RM
Range Joint Variable
HBBS LBBS
OLY
angle (8)
HBCOM
angle (8)
OLY vs. HBCOM
Difference;
690% CI
OLY vs. POW
Difference;
690% CI
POW
angle (8)
LBCOM
angle (8)
POW vs. LBCOM
Difference;
690% CI
HBCOM vs.
LBCOM
Difference;
690% CI
74–83 Hip Peak flexion 69 6764656678610 59 696164368363
ROM 100 68 105 695610 6 611 109 611 101 699612 4 64
Knee Peak flexion 54 6759683699611 62 611 63 681611 4 64
ROM 116 67 110 611z3611 5 612 114 612 104 610 8 613 5 65
Ankle Peak
dorsiflexion
90 65886646626590659068067264
ROM 33 64326306416633663064266263
84–93 Hip Peak flexion 69 696466z6686611 59 686163z367363
ROM 100 69 105 610 6 611 8 611 111 611 99 6913611 5 65
Knee Peak flexion 56 6761684687611 63 612 67 654610 6 66
ROM 114 67 107 611 5 610 1 612 113 613 101 6612612 6 66
Ankle Peak
dorsiflexion
91 64906626526590659167167163
ROM 33 64306526426634673064466062
94–99 Hip Peak flexion 71 610 69 66z46912612 59 696165z269868
ROM 98 610 100 610 4 611 11 613 110 614 100 610 9 614 0 63
Knee Peak flexion 56 6765688694610 62 612 68 655610 3 64
ROM 113 68 103 612 8 612 2 612 114 613 101 6711612 6 66
Ankle Peak
dorsiflexion
90 65916616606690659267267162
ROM 33 64286446516633672963466162
100 Hip Peak flexion 71 696866z36812612 59 610 63 66z468565
ROM 97 610 101 610 4 610 11 612 109 613 96 611 13 613 5 65
Knee Peak flexion 56 67§ 65 66z§9697611 63 612 73 66z10 610 7 67
ROM 113 69 103 69z10 610 0 612 113 614k95 68zk 18 618 8 68
Ankle Peak
dorsiflexion
90 65926616606691669366367262
ROM 32 63§ 27 64§ 5 6516633682764666064
*1RM = 1 repetition maximum; CI = confidence interval; HBBS = high-bar back-squat; HBCOM = comparison high-bar back-squat; LBBS = low-bar back-squat; LBCOM =
comparison low-bar back-squat; OLY = Olympic weightlifters; POW = powerlifters; ROM = range-of-motion.
†All angle data are presented as mean 6SD.
zp#0.05 HBCOM vs. LBCOM.
§p#0.05 OLY vs. HBCOM.
kp#0.05 POW vs. LBCOM.
Journal of Strength and Conditioning Research
the
TM
|
www.nsca.com
VOLUME 00 | NUMBER 00 | MONTH 2017 | 5
Copyright ª2017 National Strength and Conditioning Association
at-least 6 more repetitions possible”) was expected for the
70 and 80% of 1RM sets. If these RPE values were not
achieved, the predicted 1RMs for both back-squat varia-
tions were changed for the final testing session. In the
period between the second familiarization session and
the final testing session, a self-directed familiarization ses-
sion was included for each participant to reinforce the
skills learned in the previous familiarization sessions and
to provide a chance to practice each bar position before
the testing. Each participant was asked not to exceed an
RPE of 5 in this session and to do no more than 3 sets. The
final testing session was performed 3 days later and com-
prised a full anthropometric assessment, followed by
a 1RM test of both the HBBS and LBBS in random order,
so that half of the comparison group performed the HBBS
first and the other half performed the LBBS first. This
randomized order was employed to minimize any fatigue
affect from performing 2 maximal squat tests in one testing
session.
Back-Squat 1 Repetition Maximum Testing Protocol
All squats were completed in line with the International
Powerlifting Federation’s competition rules (21). Both the
HBBS and LBBS were deemed to be successful lifts if the
athlete was able to safely lower the bar to a minimum
accepted depth (the top surface of the legs at the hip joint
is lower than the top of the knees) or lower, through
a bending of the knees, and then recover at will to a stance
with knees locked, without the aid of any spotters. The
OLY participants were instructed to squat to the usual
depth they perform in training. Specific focus was placed
on ensuring correct depth was obtained, the legs were
TABLE 5. Kinematic effect sizes and percentage differences.*
%
Range Joint Variable
OLY vs. HBCOM POW vs. LBCOM OLY vs. POW
HBCOM vs.
LBCOM
Effect
size
%
Difference
Effect
size
%
Difference
Effect
size
%
Difference
Effect
size
%
Difference
74–83 Hip Peak flexion 0.5z4.3 0.2†2.7 0.4†16.4 0.7z5.1
ROM 0.3†7.4 0.4†8.3 0.3†8.0 0.6z3.9
Knee Peak flexion 0.2†5.1 0.0 1.8 0.4†12.0 0.6z7.1
ROM 0.2†2.5 0.3†8.8 0.2†1.9 0.7z5.2
Ankle Peak
dorsiflexion
0.3†1.9 0.0 0.6 0.2†0.7 0.3†2.3
ROM 0.1 15.9 0.2†10.1 0.1 1.7 0.3z5.9
84–93 Hip Peak flexion 0.4†7.9 0.2†3.1 0.3†16.2 0.8§ 4.2
ROM 0.3†5.3 0.1 11.9 0.4†10.2 0.6z5.7
Knee Peak flexion 0.3†8.7 0.2†6.4 0.3†11.0 0.6z9.5
ROM 0.2†6.2 0.6z12.4 0.1 0.8 0.6z6.2
Ankle Peak
dorsiflexion
0.2†0.8 0.0 0.8 0.2†0.4 0.2†1.2
ROM 0.2†10.2 0.3†14.6 0.2†2.4 0.0 1.5
94–99 Hip Peak flexion 0.2†2.4 0.1 2.9 0.6z19.5 2.3§ 11.8
ROM 0.2†2.3 0.4†9.4 0.4†10.9 0.1 0.2
Knee Peak flexion 0.4†13.7 0.2†7.9 0.2†9.9 0.3†4.0
ROM 0.4†10.0 0.5z12.4 0.1 0.6 0.6z1.6
Ankle Peak
dorsiflexion
0.1 1.0 0.1 1.9 0.0 0.1 0.2†0.9
ROM 0.4†15.4 0.3†13.9 0.1 1.9 0.2†3.2
100 Hip Peak flexion 0.2†3.8 0.3†6.6 0.7z20.7 1.3§ 7.9
ROM 0.2†3.6 0.5z13.2 0.5z10.5 0.6z4.9
Knee Peak flexion 0.7z14.3 0.5z13.7 0.3†10.4 0.9§ 10.9
ROM 0.6z9.9 0.8§ 18.9 0.0 0.1 0.8§ 7.7
Ankle Peak
dorsiflexion
0.1 1.3 0.2†2.9 0.0 0.2 0.6z1.9
ROM 0.7z18.3 0.5z22.0 0.1 1.9 0.0 1.1
*HBCOM = comparison high-bar back-squat; LBCOM = comparison low-bar back-squat; OLY = Olympic weightlifters; POW =
powerlifters; ROM = range-of-motion.
†Small effect d$0.2.
zModerate effect d$0.5.
§Large effect d$0.8.
High-Bar vs. Low-Bar Squats
6
Journal of Strength and Conditioning Research
the
TM
Copyright ª2017 National Strength and Conditioning Association
TABLE 6. Kinetic results 74–83% 1RM.*†
Phase Variable
HBBS LBBS
OLY HBCOM
OLY vs.
HBCOM
Difference;
690% CI
OLY vs.
POW
Difference;
690% CI POW LBCOM
POW vs.
LBCOM
Difference;
690% CI
HBCOM vs.
LBCOM
Difference;
690% CI
Eccentric Mean bar v(m$s
21
) 0.51 60.13 0.44 60.11 0.09 60.19 0.05 60.10 0.54 60.12z0.38 60.09z0.20 60.20 0.06 60.07
Peak Fv (N$kg
21
)3863§ 26 64§ 10 610 1 633762.69z26 63z969162
RFD (0–50 ms)
(N$s
21
)
2,746 61,080§ 845 6318§ 2,190 62,190 862 6948 2,294 6824 1,102 6339 1,213 61,213 231 6319
RFD (0–100 ms)
(N$s
21
)
3,657 61,788k1,570 6539 2,396 62,396 1,319 61,553 3,058 61,376k1,877 6415 1,377 61,641 337 6436
Concentric Mean bar v(m$s
21
) 0.51 60.05 0.49 60.11 0.07 60.08 0.03 60.06 0.57 60.08z0.55 60.11z0.09 60.10 0.04 60.05
Peak Fv (N$kg
21
)3863§ 31 627§ 10 610 1 633763z27 64z868161
RFD (0–50 ms)
(N$s
21
)
2,013 6737 816 6416 1,131 61,131 311 61,046 2,002 61,089 707 6166 1,319 61,317 85 6283
RFD (0–100 ms)
(N$s
21
)
3110 61502 1391 6608 1695 61922 344 61634 3287 61474 1258 6328 2084 62084 98 6359
*1RM = 1 repetition maximum; CI = confidence interval; Fv = vertical force; HBBS = high-bar back-squat; HBCOM = comparison high-bar back-squat; LBBS = low-bar back-
squat; LBCOM = comparison low-bar back-squat; OLY = Olympic weightlifters; POW = powerlifters; RFD = rate of force development.
†All kinetic data are presented as mean 6SD.
zp#0.05 POW vs. LBCOM.
§p#0.05 OLY vs. HBCOM.
kp#0.05 OLY vs. POW.
Journal of Strength and Conditioning Research
the
TM
|
www.nsca.com
VOLUME 00 | NUMBER 00 | MONTH 2017 | 7
Copyright ª2017 National Strength and Conditioning Association
completely locked out at the conclusion of each repeti-
tion, and no downward movement was observed on the
ascent.
Before testing, each participant’s beltless 1RM was esti-
mated. If in normal training, the participant did not use
a weight belt, the athlete’s predicted beltless 1RM was
used. If the participant used a weight belt in normal train-
ing, and had a known belted 1RM, this belted 1RM was
used to predict the athlete’s beltless 1RM. Pilot testing
determined that the beltless 1RM is approximately 90%
of a belted 1RM. Weightlifting shoes (comprised of a hard
sole and a slightly raised heel) were required to be worn
by all participants, and the heel height was required to be
within the range of 1.5–2.0 cm. All participants were
accustomed to wearing weightlifting shoes. No other sup-
portive aids beyond the use of wrist wraps were allowed
to be worn during the test. Before all testing procedures,
each participant completed a standardized dynamic
warm-up.
The 1RM testing protocol was adapted from Matuszak
et al. (26) and consisted of the participants performing 8
repetitions at 50% of the predicted 1RM, 3 repetitions at
60%, 2 repetitions at 70%, and 1 repetition at 80 and 90%.
Additional warm-up sets, before the initial 8 repetition set
with 50% 1RM, were permitted with ,50% 1RM load if the
participant desired to do so as to better replicate their nor-
mal warm-up procedures. After the 90% of predicted 1RM
lift, the participant was consulted as to what weight they
would like to attempt for a maximal 1RM lift. An experi-
enced strength coach along with the use of a Gymaware
Powertool (Kinetic Performance Technology, Canberra,
Australia) to measure the mean concentric velocity of the
movement assisted athletes in attempt selection to get as
close to a true beltless 1RM as possible. Previous research
has shown that maximal squat attempts performed by
experienced lifters are typically performed at approximately
0.2 m$s
21
(0.24 60.04 m$s
21
) (43). Commonly, a lift at
95% 1RM was performed before attempting the predicted
maximal 1RM. After each successful attempt, small weight
increments (1–5 kg) were made to obtain a true maximum.
Between 3 and 5 minutes of rest was allowed between sets
before the next weight was attempted.
Biomechanical Instrumentation
Two embedded force platforms (Model AM6501; Bertec
Corp., Columbus, OH, USA), were used to collect all
kineticsquatdataatasamplingrateof1,000Hz.The
kinetic variables of interest included mean bar velocity
(m$s
21
); peak Fv (N$kg
21
); rate of force development
(RFD)(0–50ms)(N$s
21
); and RFD (0–100 ms) (N$s
21
)
for both the eccentric and concentricphases.TheRFD
variableswerechoseninlinewithprevioussquatresearch
(8). Mean bar velocity was chosen over peak bar velocity
for a better representation of each athlete’s ability to move
load throughout the whole lifting phase (concentric/
eccentric) (22). Rate of force development is the change
inforceoveragiventime(33),andtheeccentricphaseof
each movement is where the body lowers and slows to
TABLE 7. Kinetic effect sizes and percentage differences 74–83% 1RM.*
Phase Variable
OLY vs. HBCOM POW vs. LBCOM OLY vs. POW
HBCOM vs.
LBCOM
Effect
size
%
Difference
Effect
size
%
Difference
Effect
size
%
Difference
Effect
size
%
Difference
Eccentric Mean bar v(m$s
21
) 0.3†15.9 1.0§ 40.2 0.3†5.3 0.7z12.7
Peak Fv (N$kg
21
) 1.5§ 43.7 1.9§ 40.9 0.2†1.3 0.4†0.7
RFD (0–50 ms)
(N$s
21
)
1.2§ 224.8 0.9§ 108.1 0.6z19.7 0.7z23.3
RFD (0–100 ms)
(N$s
21
)
0.8§ 132.9 0.6z62.9 0.6z19.6 0.7z16.4
Concentric Mean bar v(m$s
21
) 0.6z2.4 0.6z4.0 0.4†11.9 0.8§ 11.7
Peak Fv (N$kg
21
) 1.6§ 42.0 1.6§ 37.6 0.2†0.7 1.1§ 2.5
RFD (0–50 ms)
(N$s
21
)
0.8§ 146.7 0.7z183.3 0.2†0.6 0.3†15.5
RFD (0–100 ms)
(N$s
21
)
0.6z123.6 0.8§ 161.2 0.1 5.4 0.2†10.5
*1RM = 1 repetition maximum; Fv = vertical force; HBCOM = comparison high-bar back-squat; LBCOM = comparison low-bar
back-squat; OLY = Olympic weightlifters; POW = powerlifters; RFD = rate of force development.
†Small effect d$0.2.
zModerate effect d$0.5.
§Large effect d$0.8.
High-Bar vs. Low-Bar Squats
8
Journal of Strength and Conditioning Research
the
TM
Copyright ª2017 National Strength and Conditioning Association
TABLE 8. Kinetic results 84–93% 1RM.*†
Phase Variable
HBBS LBBSt
OLY HBCOM
OLY vs.
HBCOM
Difference;
690% CI
OLY vs. POW
Difference;
690% CI POW LBCOM
POW vs.
LBCOM
Difference;
690% CI
HBCOM vs.
LBCOM
Difference;
690% CI
Eccentric Mean bar v
(m$s
21
)
0.48 60.09 0.39 60.08 0.09 60.19 0.00 60.10 0.51 60.10z0.35 60.10z0.16 60.16 0.04 60.04
Peak Fv
(N$kg
21
)
40 63§ 28 65§ 10610 2 633863z27 63z10 610 0 63
RFD (0–50
ms)
(N$s
21
)
2,258 6943 857 6737 1,088 61,188 517 6957 1,857 6648z493 6112z1,425 61,425 362 6745
RFD (0–
100 ms)
(N$s
21
)
3,413 61,587 1,552 61,233 1,727 62,147 715 61,648 2,896 61,226z950 674z1,987 61,987 602 61,247
Concentric Mean bar v
(m$s
21
)
0.41 60.06 0.40 60.05 0.06 60.06 0.00 60.07 0.44 60.09z0.42 60.09z0.06 60.08 0.03 60.05
Peak Fv
(N$kg
21
)
39 64§ 28 65§ 10 610 2 643863z29 64z868163
RFD (0–50
ms)
(N$s
21
)
2,278 6921 889 6324 1,282 61,282 865 61,023 1,617 6838 705 6243 1,036 61,036 183 6333
RFD (0–
100 ms)
(N$s
21
)
3,303 61,632 1,325 6674 1,930 61,994 1,024 61,727 2,686 61,448z964 6223z1,871 61,871 357 6719
*1RM = 1 repetition maximum; CI = confidence interval; Fv = vertical force; HBBS = high-bar back-squat; HBCOM = comparison high-bar back-squat; LBBS = low-bar back-
squat; LBCOM = comparison low-bar back-squat; OLY = Olympic weightlifters; POW = powerlifters; RFD = rate of force development.
†All kinetic are data presented as mean 6SD.
zp#0.05 POW vs. LBCOM.
§p#0.05 OLY vs. HBCOM.
Journal of Strength and Conditioning Research
the
TM
|
www.nsca.com
VOLUME 00 | NUMBER 00 | MONTH 2017 | 9
Copyright ª2017 National Strength and Conditioning Association
a point of zero velocity, immediately before the start of the
concentric ascent. The eccentric RFD is measured in
the time before this change from the eccentric phase to
the concentric phase. The 2 force platforms were arranged
next to each other in the middle of the collection space to
increase the chances of obtaining complete foot contact
from each foot during the required movements. Kinemat-
ics were collected by 9 infrared cameras (T10S; Vicon
Motion System Ltd., Oxford, United Kingdom) strategi-
cally placed around the force platforms in the collection
space. The cameras were arranged so that each marker
was always visible to a minimum of 3 cameras to allow
for reconstruction of three-dimensional (3D) trajectories.
The collection space was calibrated with an error of no
greater than 0.2 (route mean squared in camera pixels; the
difference between the two-dimensional image of each
marker on the camera sensor and the 3D reconstructions
of those markers projected back to the cameras sensor) for
each camera before each data collection session, and
a point of origin was positioned at the corner of one of
the force platforms to establish a local relationship
between the camera positions and the laboratory origin.
Data from 8 reflective markers (10-mm diameter) placed
in specific locations were used to analyze bar path and
joint angles throughout the squat movement using Vicon
Nexus software (Version 1.8.5; Vicon Motion System Ltd.,
Oxford, United Kingdom). The joint angles were calcu-
lated as the angle between a parent segment (i.e., thigh
or femur) and a child segment (i.e., shank or tibia).
Markers were placed in the center of both ends of the
barbell and on the right side of the athletes’ bodies in
specific anatomical locations after previous research (28)
(Figure 1). The markers were placed on the following lo-
cations: acromion process, greater trochanter, lateral epi-
condyle of the femur, lateral malleolus, top of the heal lift
of the lifting shoe and in line with the lateral malleolus,
and base of the fifth metatarsal to create 5 rigid segments.
Data Reduction
Subsequent to the testing sessions, the 2 force platforms
were combined and all data were filtered with a low-pass
fourth-order zero-lag Butterworth filter using a cut-off
frequencyof16Hzinacustom-madeLabVIEWprogram
(Version 14.0; National Instruments Corp., Austin, TX,
USA) based on residual analysis and visual inspection of
the kinematic and kinetic data. Kinematic variables of
interest were gathered through an individual analysis
within the start and finish of the squat to calculate the
range-of-motion (ROM) (peak flexion—initial or finishing
flexion) and peak flexion angles for the hip, knee, and ankle
joints. Peak joint flexion was recorded as the angle at the
lowest point of the lift, and peak extension at the highest
point of the lift. The hip ROM in the sagittal plane was
derived from the anterior angle between the thorax (trunk)
and the thigh, the knee ROM was derived from the posterior
angle between the thigh and the shank, and the ankle ROM
TABLE 9. Kinetic effect sizes and percentage differences 84–93% 1RM.*
Phase Variable
OLY vs. HBCOM POW vs. LBCOM OLY vs. POW
HBCOM vs.
LBCOM
Effect
size
%
Difference
Effect
size
%
Difference
Effect
size
%
Difference
Effect
size
%
Difference
Eccentric Mean bar v(m$s
21
) 0.3†22.7 0.9§ 46.0 0.0 6.1 0.9§ 10.5
Peak Fv (N$kg
21
) 1.5§ 42.5 2.1§ 39.6 0.4†4.3 0.1 2.1
RFD (0–50 ms)
(N$s
21
)
0.7z163.5 1.5§ 276.5 0.4†21.6 0.4†73.7
RFD (0–100 ms)
(N$s
21
)
0.6z119.9 1.1§ 204.8 0.3†17.8 0.4†63.3
Concentric Mean bar v(m$s
21
) 0.7z5.5 0.5z5.1 0.0 6.8 0.4†7.7
Peak Fv (N$kg
21
) 1.3§ 39.7 1.6§ 32.0 0.4†3.6 0.3†2.1
RFD (0–50 ms)
(N$s
21
)
0.8§ 156.3 0.9§ 129.2 0.6z40.9 0.5z26.0
RFD (0–100 ms)
(N$s
21
)
0.7z149.3 0.9§ 178.6 0.4†23.0 0.4†37.4
*1RM = 1 repetition maximum; Fv = vertical force; HBCOM = comparison high-bar back-squat; LBCOM = comparison low-bar
back-squat; OLY = Olympic weightlifters; POW = powerlifters; RFD = rate of force development.
†Small effect d$0.2.
zModerate effect d$0.5.
§Large effect d$0.8.
High-Bar vs. Low-Bar Squats
10
Journal of Strength and Conditioning Research
the
TM
Copyright ª2017 National Strength and Conditioning Association
TABLE 10. Kinetic results 94–99% 1RM.*†
Phase Variable
HBBS LBBS
OLY HBCOM
OLY vs.
HBCOM
Difference;
690% CI
OLY vs. POW
Difference;
690% CI POW LBCOM
POW vs.
LBCOM
Difference;
690% CI
HBCOM vs.
LBCOM
Difference;
690% CI
Eccentric Mean bar v
(m$s
21
)
0.47 60.09 0.36 60.10 0.12 60.15 0.04 60.13 0.45 60.12 0.34 60.07 0.10 60.14 0.03 60.07
Peak Fv
(N$kg
21
)
41 64z29 64z§11611 2 643963k28 63§k11 611 1 61
RFD (0–50
ms)
(N$s
21
)
2,018 61,110 811 6500 1,272 61,479 383 61,275 1,618 61,107 687 6140 893 61,207 123 6504
RFD (0–
100 ms)
(N$s
21
)
2,953 61,658 1,413 6957 1,344 62,274 477 61,665 2,371 61,266 1,071 6402 1,302 61,418 300 6848
Concentric Mean bar v
(m$s
21
)
0.32 60.03 0.31 60.05 0.06 60.06 0.02 60.05 0.31 60.05 0.31 60.04 0.01 60.06 0.02 60.03
Peak Fv
(N$kg
21
)
41 65z29 64z11 611 2 643963k30 64k969161
RFD (0–50
ms)
(N$s
21
)
2,083 6906z706 6525z992 61,154 327 6935 1,595 6818 575 6342 1,016 61,016 141 6294
RFD (0–
100 ms)
(N$s
21
)
3,425 61,412 1,062 6815 1,880 61,880 498 61,481 2,761 61,258k870 6461k1,866 61,866 224 6542
*1RM = 1 repetition maximum; CI = confidence interval; Fv = vertical force; HBBS = high-bar back-squat; HBCOM = comparison high-bar back-squat; LBBS = low-bar back-
squat; LBCOM = comparison low-bar back-squat; OLY = Olympic weightlifters; POW = powerlifters; RFD = rate of force development.
†All kinetic data are presented as mean 6SD.
zp#0.05 OLY vs. HBCOM.
§p#0.05 HBCOM vs. LBCOM.
kp#0.05 POW vs. LBCOM.
Journal of Strength and Conditioning Research
the
TM
|
www.nsca.com
VOLUME 00 | NUMBER 00 | MONTH 2017 | 11
Copyright ª2017 National Strength and Conditioning Association
was derived from the angle between the shank and the foot. In
all cases, the actual angle is presented as opposed to the raw
angle (Figure 1). To obtain kinetic variables of interest, all
repetitions were individually analyzed during the eccentric
phase (from the initiation of a negative [downward] velocity
of the right-side bar marker to the instant the marker reached
zero velocity [full depth]) and concentric phase (from the
initiation of a positive [upward] velocity of the right-side
bar marker to the instant the marker reached zero velocity
a second time [the top]).
To obtain kinematic variables of interest, all repetitions
were individually analyzed within the start and finish of the
squat movement to calculate the ROM (peak flexion—initial
flexion) and peak flexion angles for the hip, knee, and ankle
joints. From the sagittal plane, the hip ROM was derived
from the anterior angle between the thorax (trunk) and the
thigh, the knee ROM was derived from the posterior angle
between the thigh and the shank and the ankle ROM was
derived from the angle between the shank and the foot. In all
cases, the actual angle is presented as opposed to the raw
(Figure 1).
Statistical Analyses
Before analyses, data were split into 4 categories according
to trials where a single squat was completed and the %1RM
load achieved in testing: (a) 74–83%, (b) 84–93%, (c) 94–
99%, and (d) 100%. This was necessary because of the var-
iation in the number of single repetition trials completed
before a true 1RM was achieved between participants. If
multiple trials were completed within a 1RM range for a par-
ticipant, the results were averaged so each participant effec-
tively had 1 trail per category. Generalized linear mixed
models using a normal distribution with an identity link
and unstructured covariance structure were used to estimate
the difference in outcome variables between bar height and
subject group across all 4 load groups, while adjusting for the
random effect of subject. In an unstructured covariance
matrix, each variance and each covariance value is estimated
uniquely from the data, resulting in the best possible model
fit (39). Robust standard errors, constructed using the “sand-
wich estimator” of the covariance structure, were used to
control for possible misspecifications of the correlation struc-
ture. An alpha of 0.05 was used to determine significant
associations. Multiple pairwise comparisons were corrected
for inflation of type 1 error using the Bonferroni method
(e.g., for all pairwise comparisons in a fixed factor with 3
groups, significance level was divided by 3). For all variables,
Cohen’s dstatistic was calculated as the estimated marginal
means divided by the square root of N multiplied by the SE
(i.e., the SD) to provide additional information on the mag-
nitude of the associations, with 0.2, 0.5, and 0.8 representing
small, moderate, and large effects, respectively (3). The anal-
ysis used IBM SPSS Statistics v. 23.0.0.0 (IBM, Armonk, NY,
USA) software.
RESULTS
Initially, a comparison of the HBBS performed by the OLY
and comparison groups (comparison high-bar back-squat
TABLE 11. Kinetic effect sizes and percentage differences 94–99% 1RM.*
Phase Variable
OLY vs. HBCOM POW vs. LBCOM OLY vs. POW
HBCOM vs.
LBCOM
Effect
size
%
Difference
Effect
size
%
Difference
Effect
size
%
Difference
Effect
size
%
Difference
Eccentric Mean bar v(m$s
21
) 0.6z27.9 0.5z33.5 0.2†2.4 0.4†6.4
Peak Fv (N$kg
21
) 1.2§ 43.9 2.0§ 41.5 0.3†4.2 1.3§ 2.4
RFD (0–50 ms)
(N$s
21
)
0.6z148.9 0.5z135.6 0.2†24.7 0.2†18.0
RFD (0–100 ms)
(N$s
21
)
0.4†109.0 0.7z121.4 0.2†24.5 0.3†31.9
Concentric Mean bar v(m$s
21
) 1.2§ 5.5 0.1 1.1 0.3†3.9 0.5z2.7
Peak Fv (N$kg
21
) 1.2§ 42.4 1.6§ 32.1 0.3†4.1 0.7z3.5
RFD (0–50 ms)
(N$s
21
)
0.6z195.0 0.8§ 177.6 0.6z30.6 0.4†22.9
RFD (0–100 ms)
(N$s
21
)
0.7z222.3 0.9§ 217.2 0.2†24.0 0.4†22.1
*1RM = 1 repetition maximum; Fv = vertical force; HBCOM = comparison high-bar back-squat; LBCOM = comparison low-bar
back-squat; OLY = Olympic weightlifters; POW = powerlifters; RFD = rate of force development.
†Small effect d$0.2.
zModerate effect d$0.5.
§Large effect d$0.8.
High-Bar vs. Low-Bar Squats
12
Journal of Strength and Conditioning Research
the
TM
Copyright ª2017 National Strength and Conditioning Association
TABLE 12. Kinetic results 100% 1RM.*†
Phase Variable
HBBS LBBS
OLY HBCOM
OLY vs.
HBCOM
Difference;
690% CI
OLY vs. POW
Difference;
690% CI POW LBCOM
POW vs.
LBCOM
Difference;
690% CI
HBCOM vs.
LBCOM
Difference;
690% CI
Eccentric Mean bar v
(m$s
21
)
0.48 60.09z0.34 60.09z0.14 60.14 0.03 60.11 0.44 60.14§ 0.31 60.06§ 0.14 60.14 0.04 60.07
Peak Fv
(N$kg
21
)
42 64z29 64z13 613 2 634062§ 29 63§ 11 611 0 62
RFD (0–50
ms)
(N$s
21
)
2,240 6852z634 6372z1,606 61,606 490 6905 1,750 6878§ 375 6337§ 1,375 61,375 258 6413
RFD (0–
100 ms)
(N$s
21
)
3,062 61,681 1,052 6650 2,010 62,010 406 61,660 2,656 61,485§ 676 6581§ 1,980 61,980 376 6769
Concentric Mean bar v
(m$s
21
)
0.22 60.03 0.20 60.03 0.02 60.04 0.01 60.05 0.21 60.06 0.23 60.05 0.03 60.07 0.04 60.06
Peak Fv
(N$kg
21
)
41 64z30 63z12 612 2 644062§ 31 63§ 9.00 69.00 1 62
RFD (0–50
ms)
(N$s
21
)
1,734 6916z629 6248z1,105 61,105 86 61,197 1,820 61,332 507 6222 1,313 61,313 122 6179
RFD (0–
100 ms)
(N$s
21
)
3,218 61,572z1,049 6480z2,169 62,169 202 61,972 3,016 62,153 676 6254 2,341 62,341 374 6528
*1RM = 1 repetition maximum; CI = confidence interval; Fv = vertical force; HBBS = high-bar back-squat; HBCOM = comparison high-bar back-squat; LBBS = low-bar back-
squat; LBCOM = comparison low-bar back-squat; OLY = Olympic weightlifters; POW = powerlifters; RFD = rate of force development.
†All kinetic data are presented as mean 6SD.
zp#0.05 OLY vs. HBCOM.
§p#0.05 POW vs. LBCOM.
Journal of Strength and Conditioning Research
the
TM
|
www.nsca.com
VOLUME 00 | NUMBER 00 | MONTH 2017 | 13
Copyright ª2017 National Strength and Conditioning Association
[HBCOM]) and the LBBS performed by the POW and
comparison groups (comparison low-bar back-squat
[LBCOM]) was completed to determine whether the
comparison group data could be combined with the OLY
or the POW for the high- and low-bar positions, respec-
tively. Significant joint angle differences were observed in
knee flexion (p= 0.04) and ankle ROM (p= 0.04) at 100% of
1RM for HBBS (OLY vs. HBCOM) and in knee ROM (p=
0.02) at 100% 1RM for the LBBS (POW vs. LBCOM). Sig-
nificant differences for several kinetic variables across all 4
percentage ranges of 1RM for both HBBS (OLY vs.
HBCOM) and LBBS (POW vs. LBCOM) were also
observed. Therefore, in the following sections, the data have
been analyzed with all 4 groups displayed independently.
Load
The mean loads are presented in Tables 1 and 2. No signif-
icant differences were observed between OLY and POW and
between HBCOM and LBCOM. However, on an average,
the POW group lifted greater loads compared with the OLY
group across all ranges of load (d= 0.3, 0.2, 0.2, and 0.2 for
ranges of 74–83%, 84–93%, 94–99%, and 100% 1RM, respec-
tively). Small effect sizes indicated that greater loads and
loads relative to body weight were lifted by the LBCOM
group than the HBCOM group for the 74–83% (d= 0.3
and 0.3, respectively), and 84–93% (d= 0.3 and 0.4, respec-
tively) 1RM ranges, but only for load at 100% 1RM (d= 0.4).
Moderate effect sizes indicated that greater loads were lifted
by the LBCOM in comparison with the HBCOM group at
94–99% 1RM in both load and load relative to body weight
(d= 0.5 and 0.6, respectively) and at 100% 1RM in load
relative to body weight (d= 0.5).
Center of Pressure
The mean distances of the bar from the center of pressure
(COP) are presented in Table 3. In the experienced OLY and
POW groups, there is a distinct difference between the 2 bar
positions. The LB BS performed by the POW shows a greater
average distance from the bar to the COP. In the less expe-
rienced COM group, the same difference is generally
observed between the HBBS and LBBS, but is much less
pronounced.
Kinematics
Differences in the estimated marginal means for the
kinematic variables are presented in Tables 4 and 5. No
significant differences were observed between the OLY
and POW groups in any condition. A significantly larger
knee flexion angle was observed in the HBCOM when
compared with the OLY group (p=0.04;d=0.7;%Dif-
ference = 14.3) at 100% 1RM. Conversely, the OLY group
displayed a significantly larger ankle ROM than the
HBCOM group at 100% (p=0.04;d= 0.07; % Difference =
18.3). The only significant difference between the POW
and LBCOM groups was observed at 100% 1RM, with the
POW group demonstrating a significantly larger knee ROM
(p= 0.02; d= 0.8; % Difference = 18.9). The majority of
significant results were observed between the HBCOM and
TABLE 13. Kinetic effect sizes and percentage differences 100% 1RM.*
Phase Variable
OLY vs. HBCOM POW vs. LBCOM OLY vs. POW
HBCOM vs.
LBCOM
Effect
size
%
Difference
Effect
size
%
Difference
Effect
size
%
Difference
Effect
size
%
Difference
Eccentric Mean bar v(m$s
21
) 0.9§ 40.0 0.9§ 46.4 0.2†7.7 0.5z11.2
Peak Fv (N$kg
21
) 1.9§ 44.8 2.9§ 39.7 0.3†3.5 0.2†0.2
RFD (0–50 ms)
(N$s
21
)
1.3§ 253.5 1.1§ 366.4 0.4†28.0 0.5z68.9
RFD (0–100 ms)
(N$s
21
)
0.8§ 191.1 0.9§ 292.9 0.2†15.3 0.4†55.6
Concentric Mean bar v(m$s
21
) 0.4†9.9 0.3†10.6 0.1 4.5 0.5z17.7
Peak Fv (N$kg
21
) 1.7§ 39.3 2.1§ 29.2 0.3†3.2 0.7z4.5
RFD (0–50 ms)
(N$s
21
)
0.9§ 175.6 0.7z259.1 0.0 4.7 0.6z24.1
RFD (0–100 ms)
(N$s
21
)
1.0§ 206.7 0.8§ 346.5 0.1 6.7 0.6z55.3
*1RM = 1 repetition maximum; Fv = vertical force; HBCOM = comparison high-bar back-squat; LBCOM = comparison low-bar
back-squat; OLY = Olympic weightlifters; POW = powerlifters; RFD = rate of force development.
†Small effect d$0.2.
zModerate effect d$0.5.
§Large effect d$0.8.
High-Bar vs. Low-Bar Squats
14
Journal of Strength and Conditioning Research
the
TM
Copyright ª2017 National Strength and Conditioning Association
LBCOM. Significant differences were observed in knee
ROM at 74–83% 1RM (p= 0.04), peak hip flexion at 84–
93% 1RM (p= 0.02), peak hip flexion at 94–99% 1RM (p=
0.00) and peak hip flexion (p= 0.00), peak knee flexion (p=
0.01), and knee ROM at 100% 1RM (p= 0.02). In all cases,
the HBCOM group displayed larger angles, except for peak
knee flexion at 100% 1RM where the LBCOM was greater.
No significant interactions between load and group were
detected.
Kinetics
Kinetic differences in estimated marginal means are
presented in Tables 6–13. The only significant difference
observed between the OLY and POW groups across all
percentage ranges of 1RM was in the eccentric phase
RFD (0–50 ms) at 74–83% 1RM (p= 0.03). Small effects
were observed for a variety of variables across all 4 ranges
of load (%1RM). Moderate kinetic effects showing
a greater OLY RFD were also observed in the eccentric
phase of the squat at 74–83% 1RM 0–50 milliseconds
(d= 0.6) and 0–100 milliseconds (d= 0.6). Moderately
larger effects were also observed in the concentric phase
in the OLY at 84–93% 1RM at 0–50 milliseconds (d=0.6)
and at 94–99% 1RM (0–50 ms) (d= 0.6). Only 1 significant
difference between the HBCOM and LBCOM was
observed. The HBCOM group produced a significantly
greater peak Fv in the eccentric phase at 94–99% 1RM
(p=0.05;d= 0.9; % Difference = 2.4) (Tables 10 and 11).
A large number of significant differences (p#0.05)
were observed across all load ranges, in both the
eccentric and concentric phases for OLY vs. HBCOM
and for POW vs. LBCOM (Tables 6, 8, 10, 12). In all cases
of significant difference, the more experienced OLY and
POW groups produced larger forces than those produced
by the less experienced HBCOM and LBCOM groups,
respectively.
DISCUSSION
The purpose of this study was to compare and contrast the
differences in kinematics and kinetics between the HBBS
and LBBS to understand why the LBBS might typically
allow for greater loads to be lifted (32). Originally, the
HBBS and LBBS were compared by combining experi-
enced populations (OLY and POW) with the same bar
position in resistance-trained individuals (HBCOM and
LBCOM). However, initial analyses revealed differences
between groups using the same bar position (i.e., between
HBCOM and OLY and between LBCOM and POW,
respectively). Therefore, each group was compared inde-
pendently to examine the kinematic and kinetic differences
that arise as a function of bar position (i.e., high-bar and
low-bar position) and experience level (i.e., OLY high-bar
vs. POW low-bar).
To the best of our knowledge, this is the first study to
compare the kinematic and kinetic differences of the HBBS
and LBBS using loads $90% 1RM. The main findings of this
investigation were: (a) statistically significant results were
observed in both joint angles and kinetics between the
OLY and HBCOM groups and between POW and LBCOM
groups; (b) although not significant, a small effect size indi-
cated that greater loads were lifted for each of the percent-
age 1R M ranges for the LBB S when comparing the POW vs.
OLY (d= 0.2–0.3). In addition, small (d$0.2) and moderate
(d$0.5) effect sizes indicated that the LBCOM group lifted
greater loads and loads relative to body weight across all
ranges of %1RM; (c) no significant differences were observed
in kinematics between the OLY and POW groups, in any
conditions, and only one significant difference was observed
between the OLY and POW groups in kinetics. However,
small (d$0.2), moderate (d$0.5), and large (d$0.8)
effects were observed across all ranges of load between
OLY and POW; (d) significantly larger joint angles were
observed on the HBCOM in comparison with the LBCOM
in knee ROM at 74–83% and 100% 1RM and peak flexion at
84–93%, 94–99%, and 100% 1RM. The LBCOM, however,
did produce a larger knee flexion angle at 100% 1RM than
the HBCOM; and (e) only 1 significant difference was
observed between the HBCOM and LBCOM groups in
kinetics. The HBCOM group produced a significantly larger
peak Fv at 94–99% 1RM in the eccentric phase.
Surprisingly, no significant differences were observed
between the experienced OLY and POW groups for any
joint angles. It was expected that the OLY would display
a greater angle at peak hip flexion because of the more
upright torso position and a smaller knee flexion angle. In
this study, small to moderate magnitudes of effect (d$0.2–
0.5) were observed at all 4 percentages of 1RM, indicating
that the OLY group demonstrated a larger hip angle dis-
played at peak flexion by the OLY group at all percentages
of 1RM tested. Previous research by Fry et al. (14) and
Wretenberg et al. (41) demonstrated a larger hip angle in
the HBBS and a greater forward lean in the LBBS. However,
the squats were performed only at 50% and 65% 1RM,
respectively, in these aforementioned studies, and the results
also failed to reach statistical significance. Therefore, it is
possible to surmise that OLY consistently demonstrate
a larger hip angle and therefore a more upright torso posi-
tion when performing the HBBS when compared with the
LBBS performed by POW. The knee joint findings of this
study were similar to those reported in other studies
(5,11,13,17,18,20,24,27,37,38), and it appears that the OLY
displays a smaller peak knee flexion angle (i.e., greater depth)
than what is seen during the POW. However, the difference
was not pronounced, as there were no significant differences
observed; but, there were small to moderate magnitudes of
change (d$0.2–0.5).
Interestingly, however, significant differences were
observed in the hip and knee joints, between the HBBS
performed by the HBCOM group, and the LBBS performed
by the LBCOM. The significant differences between these
Journal of Strength and Conditioning Research
the
TM
|
www.nsca.com
VOLUME 00 | NUMBER 00 | MONTH 2017 | 15
Copyright ª2017 National Strength and Conditioning Association
2 groups in joint angles are in line with the previous
literature, and this indicates that there may have been an
influence of experience on the significant results in this study
and in the findings of previous research. The smaller hip
angle, and greater knee angle shown by the POW group in
this study, indicates a greater posterior displacement of the
hip, a more vertical shank, and therefore a greater ankle
angle. However, this study showed no significant differences
in ankle joint angles between the OLY and POW groups.
Instead, only 1 significant difference was presented in the
ankle ROM between OLY and HBCOM at 100% 1RM (p=
0.04; d= 0.7; % Difference = 18.3). Previous investigations
have shown no definitive differences between the ankle joint
angles of the HBBS and LBBS (13,17,24,34,37). The ankle
joint angle results of this study further support these previous
findings between experienced populations (i.e., OLY and
POW), but may indicate differences in experienced vs. less
experienced groups of HBBS practitioners (i.e., OLY and
HBCOM) at maximal effort.
The upper body has a larger mass than the lower body,
and therefore humans are inherently unstable and require
effective control mechanisms to constantly resist perturba-
tion (40). This inherent instability is expressed in 3 planes of
motion when load is added to the upper body via a barbell,
as in the case of the HBBS and LBBS (35). The COP is the
point on the ground at which the Fv vector originates and is
a representation of the center of mass (COM) that accounts
for the whole body’s weight (including the external bar load)
(1). It can be argued that the COM/COP will be in the same
position with both the HBBS and LBBS; however, the var-
iation in position of the bar forces the segments of the body
to adapt differently to maintain the COM within the ath-
lete’s BOS, and therefore combat a loss of balance. A change
in 1 body segment will typically result in a change in the
other segments (12). The distance of the bar from the COP
can help indicate the level of change in these segments,
particularly when paired with kinematic joint angle data.
The results of this study indicate that the mechanisms the
body employs to maintain the balance of its system are con-
centrated at the hip and not at the knee or ankle joint. At the
deepest part of each squat, we found the distance of the bar
behind the COP was larger in the LBBS (55 639 mm) than
in the HBBS (21 636 mm) (Table 3). Anthropometric dif-
ferences (e.g., lower limb length) between participants here
would create variability if such a measure was to come from
a joint center. Instead, the distance from the COP accounts
better for the combined mass of the participant and external
bar load. These findings exemplify the effects of the low-bar
position being further down the back on the lower trapezius
musculature and also indicate a more vertical torso in the
HBBS. To maintain the position of the barbell on the
shoulders and to keep the body’s COM within the BOS,
the lifter must adopt a smaller torso angle when performing
the LBBS. In addition, a wider stance is also often employed
when performing the LBBS (10), and anecdotally it is per-
formed to suit the hip structure of the lifter to allow them to
obtain the required depth. An increased stance width also
acts to effectively increase the BOS, and therefore allows for
the bar to be a further distance from the COP, without ex-
iting the BOS. Thus, the smaller hip angle demonstrated in
this study may allow greater loads to be lifted with the
LBBS, because of the decreased moment arm, greater
emphasis on the strong hip musculature, and the aforemen-
tioned increased stability (34,37).
The only significant difference observed between the OLY
and POW groups across all percentage ranges of 1RM was in
the eccentric phase RFD (0–50 ms) at 74–83% 1RM (p=
0.03). However, small (d$0.2) and moderate (d$0.5)
magnitudes of change were observed for several variables
(Tables 7, 9, 11, 13). The OLY and POW who took part in
this study were all of a high level, and consequently they
lifted loads that were similar to each other when presented
relative to body weight, but not in terms of actual load
(Tables 1 and 2). Although not statistically significant, the
POW on an average lifted greater loads for each percentage
of 1RM. Previous research has shown that as load is
increased, there is a resulting increase in the Fv produced
that is proportionate to the increase in load (6,7,13,23,42).
With this in mind, it was expected that the results of this
study would show that the POW had the ability to generate
greater Fv levels during the LBBS, because of the larger
loads typically lifted. However, this did not occur. Instead,
no significant differences were observed between the POW
and OLY groups, and only small effects (d$0.2) were
observed for Fv. These effects are also in direct contrast to
Goodin (15), who showed the HBBS to produce larger Fv,
when compared with the LBBS, with loads of 20–80% 1RM,
in HBBS-dominant athletes. In this investigation, the Fv
levels were shown to be significantly greater in the LBBS
than in the HBBS only between the less experienced
HBCOM and LBCOM groups in the eccentric phase at
94–99% 1RM (p= 0.05; d= 1.3; % Difference = 2.4). This
indicates that the LBBS may in fact be a more efficient
technique of squatting large loads in proportion to the lifter’s
bodyweight. Even though greater loads were lifted by the
POW, when compared with the OLY for each set, the Fv
produced was relatively the same; thus, the mechanical
advantage can be attributed to kinematic joint angle differ-
ences. An analysis of the lower limb and trunk muscle activ-
ity throughout the squat for both the HBBS and LBBS is
necessary to supplement these conclusions. Such an analysis
will create a greater understanding as to the level of muscle
mass that is deemed to be active throughout each squat style.
These findings may provide an insight into the reasons for
differing kinetic results, through muscle activity results.
The resistance-trained men in this study were recruited as
a comparison group, and they did not have any specific
expertise in either the HBBS or LBBS. As a result, the
techniques displayed by the comparison group had many
significant kinetic differences when compared with the
High-Bar vs. Low-Bar Squats
16
Journal of Strength and Conditioning Research
the
TM
Copyright ª2017 National Strength and Conditioning Association
well-trained OLY and POW athletes (Tables 6–13). In addi-
tion, significant differences were also observed in several
joint angles between the OLY and POW groups versus the
HBCOM and LBCOM groups (Tables 4 and 5). Therefore, it
can be concluded that resistance training experience and tech-
nical proficiency have a strong influence on the associated
joint angle kinematics and kinetics. Thus, the level of experi-
ence of an individual may be a useful predictor of squatting
technical performance. This notion and the results of this
study are supported by the work of Miletello et al. (30),
who reported differences in kinetic and kinematic variables
measured at the knee when 3 different POW groups, of vary-
ing experience, performed the LBBS. In order of highest skill
to least skilled, the P OW groups were: competitive collegiate;
competitive high school; and novice. Future studies should
look to specifically include only well-trained athletes when
comparing the HBBS with LBBS, to minimize the dilution
of results from less experienced populations.
The significant differences observed between the experi-
enced (i.e., OLY and POW) groups and the less experienced (i.
e., HBCOM and LBCOM) groups indicate that the time
spent familiarizing each comparison participant with both
squat styles was insufficient to create expertise in both styles
before testing. The differences in joint angles between the 2
bar positions in the comparison group can also be attributed
to a lack of expertise in both squat styles. Another limitation
to this study was the low number of participants representing
each group, as this reduced the statistical power of the model.
Athletes competing at a high level were targeted to make up
the experienced OLY and POW groups (i.e., international and
national level, respectively). Therefore, the pool of potential
participants was automatically reduced. Moreover, athletes
were also recruited from different gyms in different stages of
competition preparation at the time of testing. As a result of
the reduced sample size, the effect size data should be
carefully considered rather than interpreting the findings
based on statistical significance alone. Future studies should
look to compare larger cohorts of experienced HBBS and
LBBS participants up to and including 100% of 1RM, with the
further addition of muscle activity analysis, to complete a full
profile of each squat style and improve statistical power.
PRACTICAL APPLICATIONS
This study provided evidence to suggest that the LBBS is
a more efficient way of squatting large loads, as demon-
strated by comparable kinetic results to the HBBS, despite
greater absolute loads being lifted. This study also indicates
that resistance-trained individuals should not be compared/
combined with well-trained athletes when comparing such
a technical movement as the HBBS or LBBS, as there is an
apparent influence of expertise on the performance of these
techniques. With regard to training adaptations, practi-
tioners seeking to place emphasis on the stronger hip
musculature should consider the LBBS, as the greater
forward lean of the movement ensures the hip muscles are
engaged more so than the HBBS. It is also recommended
that when the goal is to lift the greatest load possible, the
LBBS may be preferable. Conversely, the HBBS is more
suited to replicate movements that exhibit a more upright
torso position, such as the snatch and clean, or to place more
emphasis on the associated musculature of the knee joint.
Future research should look to analyze the muscle activity
differences between the HBBS and LBBS, up to and
including 100% 1RM. The addition of this knowledge to
the results presented in this study will provide a complete
profile of the differences between the HBBS and LBBS.
REFERENCES
1. Benda, BJ, Riley, PO, and Krebs, DE. Biomechanical relationship
between center of gravity and center of pressure during standing.
Rehabil Eng 2: 3–10, 1994.
2. Benz, RC. A Kinematic Analysis of the High and Low Bar Squat
Techniques by Experienced Low Bar Weight Lifters. Masters thesis, West
Chester, PA: West Chester University, 1989.
3. Cohen, J. Statistical power analysis. Curr Dir Psychol Sci 1: 98–101,
1992.
4. Cormie, P, McGuigan, MR, and Newton, RU. Adaptations in
athletic performance after ballistic power versus strength training.
Med Sci Sports Exerc 42: 1582–1598, 2010.
5. Donnelly, DV, Berg, WP, and Fiske, DM. The effect of the direction
of gaze on the kinematics of the squat exercise. J Strength Cond Res
20: 145–150, 2006.
6. Ebben, WE and Jensen, RL. Electromyographic and kinetic analysis
of traditional, chain, and elastic band squats. J Strength Cond Res 16:
547–550, 2002.
7. Ebben, WP, Garceau, LR, Wurm, BJ, Suchomel, TJ, Duran, K, and
Petushek, EJ. The optimal back squat load for potential osteogenesis.
J Strength Cond Res 26: 1232–1237, 2012.
8. Ebben, WP, Kaufmann, CE, Fauth, ML, and Petushek, EJ. Kinetic
analysis of concurrent activation potentiation during back squats
and jump squats. J Strength Cond Res 24: 1515–1519, 2010.
9. Escamilla, RF. Knee biomechanics of the dynamic squat exercise.
Med Sci Sports Exerc 33: 127–141, 2001.
10. Escamilla, RF, Fleisig, GS, Lowry, TM, Barrentine, SW, and
Andrews, JR. A three-dimensional biomechanical analysis of the
squat during varying stance widths. Med Sci Sports Exerc 33: 984–
998, 2001.
11. Escamilla, RF, Fleisig, GS, Zheng, N, Lander, JE, Barrentine, SW,
Andrews, JR, Bergemann, BW, and Moorman, CT. Effects of
technique variations on knee biomechanics during the squat and leg
press. Med Sci Sports Exerc 33: 1552–1566, 2001.
12. Evangelista, P and Alberti, G. The physics of the squat. European
Weightlifing Federation Scientific Magazine 5: 26–39, 2016.
13. Flanagan, SP and Salem, GJ. Bilateral differences in the net joint
torques during the squat exercise. J Strength Cond Res 21: 1220–1226,
2007.
14. Fry, A, Aro, T, Bauer, J, and Kraemer, W. A comparison of methods
for determining kinematic properties of three barbell squat exercises.
J Hum Movement Stud 24: 83, 1993.
15. Goodin, J. Comparison of External Kinetic and Kinematic Variables
Between High Barbell Back Squats and Low Barbell Back Squats Across
a Range of Loads. Masters thesis, Johnson City, TN: East Tennessee
State University, 2015.
16. Gullett, JC, Tillman, MD, Gutierrez, GM, and Chow, JW. A
biomechanical comparison of back and front squats in healthy
trained individuals. J Strength Cond Res 23: 284–292, 2009.
Journal of Strength and Conditioning Research
the
TM
|
www.nsca.com
VOLUME 00 | NUMBER 00 | MONTH 2017 | 17
Copyright ª2017 National Strength and Conditioning Association
17. Hales, ME, Johnson, BF, and Johnson, JT. Kinematic analysis of the
powerlifting style squat and the conventional deadlift during
competition: Is there a cross-over effect between lifts? J Strength
Cond Res 23: 2574–2580, 2009.
18. Han, S, Ge, S, Liu, H, and Liu, R. Alterations in three-dimensional
knee kinematics and kinetics during neutral, squeeze and outward
squat. J Hum Kinetics 39: 59–66, 2013.
19. Heijne, A, Fleming, BC, Renstrom, PA, Peura, GD, Beynnon, BD,
and Werner, S. Strain on the anterior cruciate ligament during
closed kinetic chain exercises. Med Sci Sports Exerc 36: 935–941,
2004.
20. Hooper, DR, Szivak, TK, Comstock, BA, Dunn-Lewis, C,
Apicella, JM, Kelly, NA, Creighton, BC, Flanagan, SD, Looney,
DP, and Volek, JS. Effects of fatigue from resistance training on
barbell back squat biomechanics. JStrengthCondRes28: 1127–
1134, 2014.
21. International Powerlifting Federation. Technical Rules Book.
Differdange, LU: International Powerlifting Federation, 2015.
22. Jidovtseff, B, Harris, NK, Crielaard, JM, and Cronin, JB. Using the
load-velocity relationship for 1RM prediction. J Strength Cond Res
25: 267–270, 2011.
23. Kellis, E, Arambatzi, F, and Papadopoulos, C. Effects of load on
ground reaction force and lower limb kinematics during concentric
squats. J Sports Sci 23: 1045–1055, 2005.
24. Kobayashi, Y, Kubo, J, Matsuo, A, Matsubayashi, T, Kobayashi, K,
and Ishii, N. Bilateral asymmetry in joint torque during squat
exercise performed by long jumpers. J Strength Cond Res 24: 2826–
2830, 2010.
25. Maddigan, ME, Button, DC, and Behm, DG. Lower-limb and trunk
muscle activation with back squats and weighted sled apparatus. J
Strength Cond Res 28: 3346–3353, 2014.
26. Matuszak, ME, Fry, AC, Weiss, LW, Ireland, TR, and McKnight,
MM. Effect of rest interval length on repeated 1 repetition
maximum back squats. J Strength Cond Res 17: 634–637, 2003.
27. McKean, MR, Dunn, PK, and Burkett, BJ. Quantifying the
movement and the influence of load in the back squat exercise. J
Strength Cond Res 24: 1671–1679, 2010.
28. McKenzie, C, Brughelli, M, Whatman, C, and Brown, S. The
influence of optimal handheld load on the technical ability to apply
ground reaction forces during horizontal jumping in female netball
players. Int J Sports Med 37: 318–323, 2015.
29. McLaughlin, TM, Dillman, CJ, and Lardner, TJ. Kinematic model of
performance in the parallel squat by champion powerlifters. Med Sci
Sports 9: 128–133, 1977.
30. Miletello, WM, Beam, JR, and Cooper, ZC. A biomechanical
analysis of the squat between competitive collegiate, competitive
high school, and novice powerlifters. J Strength Cond Res 23: 1611–
1617, 2009.
31. Nisell, R and Ekholm, J. Joint load during the parallel squat in
powerlifting and force analysis of in vivo bilateral quadriceps tendon
rupture. Scand J Sports Sci 8: 63–70, 1986.
32. O’Shea, P. Sports Performance Series: The parallel squat. Natl
Strength Cond Assoc J 7: 4, 1985.
33. Sands, WA, McNeal, JR, and Shultz, BB. Kinetic and temporal
patterns of three types of vertical jump among elite international
divers. Res Sports Med An Int J 9: 107–127, 1999.
34. Sato, K, Fortenbaugh, D, and Hydock, DS. Kinematic changes using
weightlifting shoes on barbell back squat. J Strength Cond Res 26: 28–
33, 2012.
35. Schick, EE, Coburn, JW, Brown, LE, Judelson, DA, Khamoui, AV,
Tran, TT, and Uribe, BP. A comparison of muscle activation
between a smith machine and free weight bench press. J Strength
Cond Res 24: 779–784, 2010.
36. Stoppani, J. Encyclopedia of Muscle & Strength. Champaign, IL:
Human Kinetics, 2006.
37. Swinton, PA, Lloyd, R, Keogh, JW, Agouris, I, and Stewart, AD. A
biomechanical comparison of the traditional squat, powerlifting
squat, and box squat. J Strength Cond Res 26: 1805–1816, 2012.
38. van den Tillaar, R, Andersen, V, and Saeterbakken, AH. The
existence of a sticking region in free weight squats. J Hum Kinetics
42: 63–71, 2014.
39. West, BT, Welch, KB, and Galecki, AT. Linear Mixed Models: A
Practical Guide Using Statistical Software. Boca Raton, FL: CRC
Press, 2014.
40. Winter, DA. Human balance and posture control during standing
and walking. Gait Posture 3: 193–214, 1995.
41. Wretenberg, P, Feng, Y, and Arborelius, UP. High- and low-bar
squatting techniques during weight-training. Med Sci Sports Exerc 28:
218–224, 1996.
42. Zink, AJ, Perry, AC, Robertson, BL, Roach, KE, and Signorile, JF.
Peak power, ground reaction forces, and velocity during the squat
exercise performed at different loads. J Strength Cond Res 20: 658–
664, 2006.
43. Zourdos, MC, Klemp, A, Dolan, C, Quiles, JM, Schau, KA, Jo, E,
Helms, E, Esgro, B, Duncan, S, Merino, SG, and Blanco, R. Novel
resistance training-specific RPE scale measuring repetitions in
reserve. J Strength Cond Res 30: 267–275, 2015.
High-Bar vs. Low-Bar Squats
18
Journal of Strength and Conditioning Research
the
TM
Copyright ª2017 National Strength and Conditioning Association
