Upper and lower body power tests predict serve performance in national and international level male tennis players

Abstract

Successful outcome of the tennis serve is primarily dependent on speed (1), which is a result of effective full kinetic chain function (2-5). Specific mechanical determinants, kinetic and kinematic variables, that contribute to serve speed have been identified (2, 4, 6-9). However, current tests of physical fitness in tennis do not target all mechanical determinants (10-20). Implementation of tests that target known mechanical determinants in full kinetic chain movements have been called for (3, 21). The purpose of this study was to determine the relationship between full kinetic chain tests of dynamic postural control, force and power that target known mechanical determinants of tennis serve performance.

Full text

Upper and lower body power tests predict serve performance
Upper and lower body power tests predict serve
performance in national and international level
male tennis players
O. Eriksrud 1,A.Ghelem2, F. Henrikson 3, J. Englund 3,N.Brodin3
1Department of Physical Performance, Norwegian School of Sports of Science, P. O Box 4014 Ullevaal Stadion, 0806 Oslo, Norway,2Ali Ghelem AB, Stockholm, Sweden, and
3Karolinska Institutet, Department of Neurobiology, Care Sciences and Society, Division of Physiotherapy, 141 83 Huddinge, Sweden
Testing |Serve |Power |Tennis |International Level
Headline
Successful outcome of the tennis serve is primarily depen-
dent on speed (1), which is a result of e↵ective full kinetic
chain function (2-5). Specific mechanical determinants, kinetic
and kinematic variables, that contribute to serve speed have
been identified (2, 4, 6-9). However, current tests of physi-
cal fitness in tennis do not target all mechanical determinants
(10-20). Implementation of tests that target known mechan-
ical determinants in full kinetic chain movements have been
called for (3, 21).
Aim.Determine the relationship between full kinetic chain
tests of dynamic postural control, force and power that target
known mechanical determinants of tennis serve performance.
Methods
Athletes.Twelve recreational to semi-professional (interna-
tional 5; national 4; regional 3) Swedish tennis players (age
28.3±10.3 years; body mass 80.0±6.0 kg; height 184.0±4.9 cm;
experience 17.5±5.5 years; mean±D) volunteered and com-
pleted the study. Nine players were right handed (determined
by serve hand) and the same side foot was considered as domi-
nant, also. Exclusion criterium was the inability to participate
in tennis and injuries in the past two months that limited
training for more than two weeks. All sub jects were given
written and verbal information of the study prior to obtaining
written informed consent. The study was conducted accord-
ing to the Declaration of Helsinki and the ethical standards of
Karolinska Institute (Stockholm, Sweden).
Design. The present study used a cross-sectional design were
force, power and dynamic postural control tests were corre-
lated to serve speed.
Methodology
Dynamic postural control. The hand reach star excursion
balance test (HSEBT) is a reliable and valid test of dynamic
postural control (22), which consist of a total of eight hori-
zontal hand reaches: anterior (A0), left anterolateral (L45),
left lateral (L90), left posterolateral (L135), posterior (P180),
right posterolateral (R135), right lateral (R90) and right an-
terolateral (R45) measured in cm, and left (L) and (R) ro-
tational reaches (LROT and RROT) measured degrees. The
HSEBT reaches included in this study were selected based
on their ability to capture three-dimensional joint movement
interactions of the shoulder, trunk and lower extremities sim-
ilar to those utilized during the di↵erent phases of the tennis
serve (5, 23-26). Specifically, the R90, R135, P180, L135,
RROT reaches were performed on the dominant foot, and the
R90, P180, RROT reaches on the non-dominant foot (Figure
2). A detailed description of testing procedures is described
elsewhere (22).
Te nn i s s er v e p er f o rm a n ce . Each subject performed their
own serve specific warm-up with their own equipment and pre-
ferred position behind the baseline. The two serve targets used
were defined by dividing the service box into two equal halves,
then the rectangle closest to the service line was divided into
three equal rectangles with the rectangle closest to the singles
sideline and service center line used as targets (Figure 1). The
subjects performed two serves towards both boxes before a 2-
minute rest period. The athletes continued serving until line
judges of the Royal Tennis Club of Stockholm approved five
serves for each box. Serve speed was measured using a stan-
dard camera set up of Playsight Smartcourt (Kochav Yair,
Israel). Average serve speed (vserve) was calculated for all
valid serves.
Forc e an d p ower . The force and power tests that tar-
geted serve specific mechanical determinants (in parenthesis)
were: 1) countermovement jump (CMJ) (leg drive) (2, 9), 2)
isokinetic single leg squat (SLS) (leg drive) (2, 9), 3) dominant
hand vertical press (vertical shoulder and leg drive) (2, 7-9),
4) bilateral arm anterior overhead push (hip and trunk flex-
ion) (7, 8) and 5) dominant hand anterior push (rotation) (27)
(Figure 3). All tests were carried out using a robotic resistance
device (1080 Quantum, 1080 Motion Nordic AB, Stockholm,
Sweden).
The testing procedures were standardized as follows: 1)
starting position of subject on a custom-made floor (foot posi-
tion in X and Y coordinates (cm) and orientation in 45-degree
increments based on HSEBT definitions with A0 defined as
the subject facing the machine) and 2) position of adjustable
arm: low (floor height), high (2,4 m above floor), knee height
(NK) or shoulder height (NS) (Figure 3). Resistance for power
tests was set to 10% body mass except for CMJ and SLS (37
kg). Concentric and eccentric speed was set to 8 m s1and
6ms1respectively for the power test, whereas 0.5 m s1
Fig. 1. Overview of tennis court and how the service box was divided into two
targets.
sportperfsci.com 1 SPSR - 2018 |Dec |42 |v1

Upper and lower body power tests predict serve performance
Fig. 2. Horizontal and rotational reaches dominant (left) and non-dominant (right). Visual representation of all reaches (photographs). Horizontal reaches (center graphs)
with average (cm, black line) and standard deviation (±SD, grey shaded area) with correlations (⇢)tov
serve . Rotational reaches ( , circular graphs) with average (±SD)
with correlations (⇢)tov
serve . Value in grey font for non-dominant leg (R135) calculated as average of P180 and R90 for visual presentation.
and 2 m s1were selected for the isokinetic tests. A general
warm-up (10 minutes) preceded familiarization. Then, five
repetitions of maximum e↵ort with a 5-minute rest between
tests were recorded. In order to get the best representation
of strength and power the repetitions with lowest and highest
peak power (Pmax) (power tests) and average force (Favg)
(isokinetic tests) were removed and the average of the three
remaining repetitions used for analysis. The results from the
SLS were also normalized to BW.
Statistical analysis. Serve speed was correlated to force, power
and dynamic postural control tests using two-tailed Spearman
rank-order correlation (⇢) since v
serve was not normally dis-
tributed (Shapiro Wilk s test <.05) (IBM SPSS version 21.0).
Outliers were removed from the analysis using the outlier la-
belling rule (28).
Results
Serve speed (164.1±22.5 km h1) was significantly correlated
(moderate to good) (29) with all power tests (Pmax), ex-
cept for the dominant hand anterior push (⇢=.413, p=.183).
Non-significant correlations were observed between vserv e and
isokinetic strength tests, with fair to moderate (29) correla-
tions for dominant hand vertical press and SLS (Table 1).
All correlations between vserve and dynamic postural control
tests were non-significant, except dominant foot L135 reach
(⇢=.607, p=.036) (Table 1).
Discussion
Serve speed were significantly correlated with power tests that
targeted leg drive (CMJ, dominant hand vertical press), ver-
tical shoulder drive (dominant hand vertical press) and trunk
and hip flexion (bilateral hand anterior press). This is a valu-
able addition to the body of knowledge considering the lim-
ited information available on the influence of musculoskeletal
power on tennis serve performance. Previously, serve speed
has been found to have no correlations with lower extremity
power, squat jump (SJ) and CMJ (13, 19, 20). Our data
contradict these findings with a good correlation (⇢=.715)
between vserve and CMJ Pmax. The greater normalized
Pmax values observed by Girard and co-workers (59.5 to 63.8
Wkg1) can be contributed to the calculation of Pmax in
1080 Quantum which is based on pulling force and speed only.
Furthermore, Ulbricht and co-workers (18) found that power
tests (overhead, forehand and backhand medicine ball throws)
and serve speed to be the best predictors of tennis performance
(national ranking), which indirectly supports our findings. In
addition, Behringer and co-workers found that plyometric con-
ditioning focusing on lower and upper body speed had a bet-
ter e↵ect on serve speed than resistance training at moder-
ate speeds (30), which is supported by our findings with all
power tests having better correlations with vserve than the
force tests. However, one test, dominant leg SLS, approached
significance (Table 1), which agrees with multi- and single joint
lower extremity strength tests having variable relationships to
serve speed (10, 13-15). The absence of significance of the
dominant hand anterior push power test supports the idea
that transverse plane power has a lower significance in deter-
mining serve speed (7, 27).
Even if HSEBT reaches were selected based on eliciting
combinations of joint movements similar to those used in the
preparation and acceleration phase of the serve (5, 24-26)
vserve was only significantly correlated with one reach, the
dominant foot L135 reach. This particular hand reach elicit
sportperfsci.com 2 SPSR - 2018 |Dec |42 |v1

Upper and lower body power tests predict serve performance
Table 1. Power (W), force (N and N kg1) and Spearman correlation coefficients (⇢) to serve ball speed
Test Po we r ⇢pPower⇢p
Dominant hand anterior push 1243 185 .413 [-.181, .838] .183 254 30 -.431 [-.871, .311] .162
Dominant hand vertical press 874 178 .650 [.007, .977] .022* 240 32 .459 [-.178, .759] .134
Bilateral hand overhead anterior push 1046 208 .643 [.419, 739] .024* 218 32 .119 [-.463, .694] .713
CMJ 1316 237 .715 [.356, .899] .009* NA NA NA
Dominant leg single leg squat NA NA NA 884 161 .524 [-.068, .853] .080
Normalized dominant leg single leg squat NA NA NA 11.1 2.0 .566 [-.092, .902] .055
Non-dominant leg single leg squat NA NA NA 871 110 .218 [-.309, .697] .519
Normalized non-dominant leg single leg
squat NA NA NA 11.0 1.5 .391 [-.253, .791] .235
Note: * p<0.05; NA: Not Assessed
Fig. 3. Description of power and isokinetic strength tests.
CMJ: Countermovement jump; NS: Normalized to shoulder height; NK: Normalized to knee height; A0=Anterior (0 ) reach; R90=Right lateral (90 ) reach; P180=Posterior
(180 ) reach; L90=Left lateral (L90) reach
dominant arm shoulder flexion, trunk extension, minimal ip-
silateral rotation and contralateral lateral flexion (23). The
non-significant reaches were either bilateral or non-dominant
hand reaches and might therefore lack specificity to the serve.
Based on these findings it appears that the combination of
dominant shoulder flexion, previously found to be significantly
correlated to serve speed (11), with trunk extension and ipsi-
lateral rotation o↵er the best representation of joint movement
combinations, or represent significant boundary conditions as-
sociated with the preparation and acceleration phases of the
tennis serve. The ability to utilize the combination of these
joint movements of a certain magnitude might allow players
to produce greater linear and angular momentum and thereby
increasing serve speed (2). This has been corroborated in a
study in which elite players with high serve speeds were ca-
pable performing backswings of greater magnitude (20). In
addition, it is important to consider that the HSEBT reaches
assess maximum reach capacity, which might not be necessary
during a serve. In fact, it could be that the subjects had suffi-
cient capacity in the non-significant reaches, which raises the
question of what portion of maximum reach capacity was uti-
lized during the serve and the margins of error between safe
and possible at-risk movements.
Practical Applications
Power tests that target established mechanical determi-
nants of serve speed can be a valuable addition to the test
batteries of tennis players.
Considering the lack of dynamic postural control tests that
target full kinetic chain movements in tennis players, the
HSEBT can be a valuable assessment tool.
sportperfsci.com 3 SPSR - 2018 |Dec |42 |v1

Upper and lower body power tests predict serve performance
Limitations
Longitudinal studies or larger cohort studies consisting of
players of di↵erent sex, age and performance level should be
evaluated to better understand the implications of current
findings on serve performance.
The sample size is small (n=12), but comparable to other
cross-sectional studies (8 to 15 participants) (15, 16, 19,
31).
Conflict of interest.
Ola Eriksrud and Ali Ghelem are shareholders in 1080 Motion
AB.
Acknowledgments. All athletes and the Royal Tennis Club of Stockholm for allowing
us to use their facilities and line judges.
Dataset
Dataset available on SportPerfSci.com
References
1. Elliott BC, Marshall RN, No↵al GJ. Contributions of
Upper-Limb Segment Rotations during the Power Serve in
Tennis. Journal of Applied Biomechanics. 1995;11(4):433-42.
2. Elliott B, Marsh T, Blanksby B. A 3-Dimensional Cin-
ematographic Analysis of the Tennis Serve. Int J Sport
Biomech. 1986;2(4):260-71.
3. Kibler WB. Biomechanical analysis of the shoulder during
tennis activities. Clin Sports Med. 1995;14(1):79-85.
4. Fleisig G, Nicholls R, Elliott B, Escamilla R. Kinematics
used by world class tennis players to produce high-velocity
serves. Sports Biomech. 2003;2(1):51-64.
5. Elliott B. Biomechanics and tennis. Br J Sports Med.
2006;40(5):392-6.
6. Elliott B, Fleisig G, Nicholls R, Escamilia R. Technique
e↵ects on upper limb loading in the tennis serve. Journal of
science and medicine in sport / Sports Medicine Australia.
2003;6(1):76-87.
7. Bahamonde RE. Changes in angular momentum during
the tennis serve. J Sports Sci. 2000;18(8):579-92.
8. Martin C, Kulpa R, Delamarche P, Bideau B. Professional
tennis players’ serve: correlation between segmental angular
momentums and ball velocity. Sports Biomech. 2013;12(1):2-
14.
9. Sweeney M, Reid M, Elliott B. Lower limb and trunk
function in the high performance tennis serve. Asian J Ex-
erc Sports Sci. 2012;9(1):13-20.
10. Signorile JF, Sandler DJ, Smith WN, Stoutenberg M,
Perry AC. Correlation analyses and regression modeling be-
tween isokinetic testing and on-court performance in com-
petitive adolescent tennis players. J Strength Cond Res.
2005;19(3):519-26.
11. Cohen DB, Mont MA, Campbell KR, Vogelstein BN,
Loewy JW. Upper extremity physical factors a↵ecting tennis
serve velocity. Am J Sports Med. 1994;22(6):746-50.
12. Reid M, Schneiker K. Strength and conditioning in
tennis: current research and practice. J Sci Med Sport.
2008;11(3):248-56.
13. Kraemer WJ, Triplett NT, Fry AC, Koziris LP, Beuer JE,
Lynch JM, et al. An in-depth sports medicine profile of women
college tennis players. J Sport Rehabil. 1995;4(2):79-98.
14. Perry AC, Wang X, Feldman BB, Ruth T, Signorile J.
Can laboratory-based tennis profiles predict field tests of ten-
nis performance? J Strength Cond Res. 2004;18(1):136-43.
15. Pugh SF, Kovaleski JE, Heitman RJ, Gilley WF. Upper
and lower body strength in relation to ball speed during a
serve by male collegiate tennis players. Percept Mot Skills.
2003;97(3 Pt 1):867-72.
16. Baiget E, Corbi F, Fuentes JP, Fernandez-Fernandez J.
The Relationship Between Maximum Isometric Strength and
Ball Velocity in the Tennis Serve. J Hum Kinet. 2016;53:63-
71.
17. Roetert EP, McCormick TJ, Brown SW, Ellenbecker TS.
Relationship between isokinetic and functional trunk strength
in elite junior tennis players. Isokinetics and Exercise Science.
1996;6(1):15-20.
18. Ulbricht A, Fernandez-Fernandez J, Mendez-Villanueva
A, Ferrauti A. Impact of Fitness Characteristics on Tennis
Performance in Elite Junior Tennis Players. J Strength Cond
Res. 2016;30(4):989-98.
19. Bonato M, Maggioni MA, Rossi C, Rampichini S,
La Torre A, Merati G. Relationship between anthropomet-
ric or functional characteristics and maximal serve velocity
in professional tennis players. J Sports Med Phys Fitness.
2015;55(10):1157-65.
20. Girard O, Micallef JP, Millet GP. Lower-limb activity
during the power serve in tennis: e↵ects of performance level.
Med Sci Sports Exerc. 2005;37(6):1021-9.
21. Ellenbecker TS, Roetert EP. An isokinetic profile of trunk
rotation strength in elite tennis players. Medicine and science
in sports and exercise. 2004;36(11):1959-63.
22. Eriksrud O, Federolf P, Sæland F, Litsos S, Cabri J. Re-
liability and Validity of the hand reach star excursion balance
test. J Funct Morph Kinesiol. 2017;2(3).
23. Eriksrud O, Federolf P, Anderson P, Cabri J. Hand
reach star excursion balance test: An alternative test for dy-
namic postural control and functional mobility. PloS one.
2018;13(5):e0196813.
24. Tubez F, Forthomme B, Croisier JL, Cordonnier C, Bruls
O, Denoel V, et al. Biomechanical analysis of abdominal in-
jury in tennis serves. A case report. Journal of sports science
& medicine. 2015;14(2):402-12.
25. Wagner H, Pfusterschmied J, Tilp M, Landlinger J,
von Duvillard SP, Muller E. Upper-body kinematics in team-
handball throw, tennis serve, and volleyball spike. Scand J
Med Sci Sports. 2014;24(2):345-54.
26. Kovacs M, Ellenbecker T. An 8-stage model for evaluating
the tennis serve: implications for performance enhancement
and injury prevention. Sports Health. 2011;3(6):504-13.
27. Gordon BJ, Dapena J. Contributions of joint rotations to
racquet speed in the tennis serve. J Sports Sci. 2006;24(1):31-
49.
28. Hoaglin DC, Iglewicz B. Fine-Tuning Some Re-
sistant Rules for Outlier Labeling. J Am Stat Assoc.
1987;82(400):1147-9.
29. Portney LG, Watkins MP. Foundations to clinical re-
search applications to practice. Stamford, CT, USA: Appleton
& Lange; 1993.
30. Behringer M, Neuerburg S, Matthews M, Mester J. Ef-
fects of two di↵erent resistance-training programs on mean
tennis-serve velocity in adolescents. Pediatr Exerc Sci.
2013;25(3):370-84.
31. Durand S, Ripamonti M, Beaune B, Rahmani A.
Leg ability factors in tennis players. Int J Sports Med.
2010;31(12):882-6.
Copyright: The articles published on Science Performance and Science
Reports are distributed under the terms of the Creative Commons Attribu-
tion 4.0 International License (http://creativecommons.org/licenses/by/4.0/),
which permits unrestricted use, distribution, and reproduction in any medium,
provided you give appropriate credit to the original author(s) and the
source, provide a link to the Creative Commons license, and indicate if
changes were made. The Creative Commons Public Domain Dedication
sportperfsci.com 4 SPSR - 2018 |Dec |42 |v1

Upper and lower body power tests predict serve performance
waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the
data made available in this article, unless otherwise stated.
sportperfsci.com 5 SPSR - 2018 |Dec |42 |v1