One of the most important actions in modern tennis is the serve.1–3 It is one of the most repeated strokes during the game, and it directly influences the outcome of points.4 The effectiveness of this action is determined by several factors such as its speed, impact angle, spin, direction, and precision.3,5 In this sense, serve velocity (SV) has been identified as one of the most determinant factors influencing performance in both men and women’s tennis alongside junior competitors.1,2,6,7 Consequently, SV has boosted in the professional tour together with an increase in aces and a decrease in double faults.8 Concurrently, the existence of sophisticated measurement systems based on high-speed cameras and laser scanners such as Foxtenn5 or Hawk-Eye,8 and the relative ease of application radar guns makes data analysis highly accessible. Therefore, knowledge around mechanisms affecting the capacity of a player to apply speed to the ball is of great interest for tennis players and professionals in terms of developing training programs that improve the action and develop strategies to avoid decrements in performance.
The necessary coordination throughout the entire kinetic chain makes this action a complex motor skill.9 The summation of forces in an optimal time and space during this movement sequence increases the velocity of the different body segments involved in the motion, and ultimately is transferred to the ball.2,3 If any of the links in the chain are not synchronized in an effective way, the result will not be optimal.10 From a technical perspective, serving appears to require a certain execution to achieve not only improved speed, but also reduced injury risk based on less aggressive motions.3 Several studies have aimed at identifying the main biomechanical aspects an effective serve should include, although depending on the model followed and the particular phase of the action (ie, loading, cocking and impact),3 the necessities required toward generating faster serves may vary. Yet, it seems clear that lower leg drive, hip and trunk rotations, and upper arm extension and internal rotations are the major contributors to racquet and ball speed.11–13
Besides technical indications, certain anthropometric characteristics have been found to have a positive relation to SV, making these important factors to consider. The strength of correlations found varies across sexes and playing levels; although, results seem to indicate the importance of obtaining higher peripheral racket velocity at ball impact, which could be increased with greater body height (BH) or arm length (AL).5,14–16 Similarly, body mass (BM) seems to influence SV as the principle of force (mass × acceleration) and torque production directly affect a player’s velocity production capacity.7,17 This positive relation is more evident when analyzing the fastest serves registered in matches.5 Monitoring of these characteristics seems relevant for a close control on growth and maturational status. In addition, the possibility of affecting body composition with appropriate training methods makes a review on how anthropometric traits affect SV of great interest.
Physical capacities and neuromuscular performance variables have also been studied in relation to their influence on SV. Maximal dynamic strength18–20 and maximal isometric strength (MIS),7,15,21,22 rate of force development (RFD), impulse (IMP),15,22 range of motion (ROM),23,24 muscle contractile properties,18 functional measurements of power such as medicine ball throws (MBT), or jumping capacity7,25 have previously been in some way related to higher or lower SV. Nevertheless, although the influence of these traits seems clear, the importance and impact of these variables on SV varies across sexes and playing levels due to interactions with other parameters present in the complex serve motion such as biomechanics and technique. Thus, knowledge around specific physical indicators and how these may vary when assessing different participants seems important for tennis practitioners. In consequence, identifying those parameters that better predict a faster serve in different populations will be reviewed.
Moreover, impairments in these key factors seem to appear following match play or certain training loads. Previous research has mainly focused on competition simulation or data analysis following official events. Although it seems clear that tennis play has the potential to affect key performance factors,26–30 how overall match play loads, volume, intensity of play, calendar, or traveling may influence these characteristics could be of further interest.
Thus, the goal of this investigation was to determine and define the physical factors affecting SV both positively, and negatively, and to approach existing differences regarding sex, level, and age. For this review, search terms included “Tennis Serve,” “Serve Speed or Velocity” and “Anthropometry,” “Biomechanics,” “Physical,” “Strength,” “Power,” “Match,” “Competition,” and “Training.” Studies reviewed included those that examined possible links between physical determinants and SV alongside investigations that registered changes in SV before, and after, training sessions or competitive matches. Criteria regarding participants included players considered as “elite” (belonging to ATP or WTA rankings above 1000 at the time of the study), “competition” (players over 18 y of age participating in competitive events without a ranking above 1000 and collegiate players), and “junior” (players under 18 y old participating in both international and national/regional events). Studies performed with players considered as “amateur” or “recreational” were not considered. We hypothesized that SV would be a multifactorial capacity influenced by biomechanical and ROM abilities (leg drive, hip, trunk, and shoulder rotations), anthropometrics (BH and mass), and strength (applying force in short time frames). In addition, high volume of match play would impair the mentioned variables and decrease performance.
Biomechanics and Movement Competency
As a complex motor skill, an efficient serve (ie, accurate, fast, and nonaggressive to musculoskeletal health) requires the development of certain technical parameters. Previous investigations have aimed at identifying those main biomechanical aspects toward this goal.3,11,12 As a key starting point, an increased speed of the head of the racquet and posterior transfer to the ball is the main aspect players should achieve.12 The height of impact and the amount of momentum and forward rotation applied to the ball seem the principal contributors toward increasing the head of the racquet’s speed.31 Thus, the influence of angular velocity vectors of the upper arm, forearm, and hand in generating this speed seems essential. Specifically, hand and upper arm flexion, trunk rotation, and abduction alongside the internal rotation of the shoulder are of paramount importance to produce fast racquet head speeds.31 This internal rotation gesture has been recognized as the major contributor to speed as it is responsible for accelerating the upper arm and building up angular velocity in the swing to impact. Yet, this explains the moment of impact in the upper arm, while the serve is a multifaceted motor skill involving several body structures and stages. In this line, arm pronation is responsible for racquet orientation while elbow extension has a high influence on impact height, which is an added contributor to head racquet speed.12,31 In addition, authors have recognized rotation and side positioning of the trunk as an enabler of generating extra rotation in the horizontal plane to increase available space and energy storage to transfer to the consequent acceleration phase.12,31 Also, investigations indicate the significance of lower limb and pelvic push as the starting point of the kinetic chain.12,32,33 Although a greater upward vertical drive of the dominant shoulder is associated to a greater SV, this is highly influenced by the drive generated from the hips and lower limbs. Specifically, peak vertical velocity of the hips, alongside an increased drive of the back leg leads to consider extension moments in the legs and internal rotations in the hip as essential toward increasing SV.12,13,32
Because the serve is a skill with several phases, the ideal biomechanical layout for faster serves depends on which specific stage of the stroke is being examined. Kovacs and Ellenbecker3 suggested the use of a multistage analysis of the serve involving 8 stages. Depending on the phase, biomechanical demands vary. Those phases responsible for velocity build-up seem to be the loading, cocking, acceleration, and impact phases. During the loading phase, kinematics mainly refers to lower body positions in which the authors identify 2 main techniques (ie, the footback and the foot-up), depending on distance between feet.3 Although the foot-up seems to generate greater vertical forces, which would be interesting to transfer throughout the kinetic chain, no ball velocity differences are observed between these 2 techniques. The footback style provides greater upward and forward leg drive while the foot-up provides a stable axis of rotation on which players rely to generate momentum.3 Also, during the loading phase, the importance of shoulder and pelvis lateral tilt has been identified, as this specific alignment facilitates the development of angular momentum through lateral trunk flexion during forward swing.3 The cocking phase is known for the importance of driving the racquet down and behind the torso allowing greater storage of elastic energy and an increased path before impact. Maximal shoulder external rotation is reached and a close parallel position between racquet and trunk seems important. Accelerating the racquet from this position until impact is known as the acceleration phase. A rapid rotation force occurs from the lumbar spine and forceful concentric internal rotation movements oversee generating velocity. Trunk rotation, elbow extension, shoulder internal rotation, and hand flexion are the main contributors to momentum in this phase.11,12,31,34,35 During contact, the best kinematic models indicate that the shoulder should be slightly abducted and the elbow, wrist, and lead knee somewhat flexed (24° [14°]).11 It is suggested that optimal impact point should happen at 110° angle of elevation between the upper arm and trunk.3 Following the impact phase, during the deceleration of the movement, all the mentioned structures bear a great level of eccentric forces towards decelerating the motion. Nevertheless, as no velocity is being generated from this phase onward, training programs should focus on providing the player effective tools to manage deceleration.3
In short, depending on the serving model followed and the specific phase of the motion, certain biomechanical and technical needs are present toward generating faster serves; however, it seems clear that literature establishes knee extension and lower leg drive, hip and trunk rotation and elbow extension, shoulder internal rotation, and hand/wrist flexion as the major contributors to angular momentum toward transferring speed to the head of the racquet and to the ball. These motions and body positions seem to be the most correlated to successful fast serves and performed by those players capable to apply speed to the ball effectively.11 Therefore, coaches are encouraged to include these indications in development programs.
Anthropometric Characteristics
The height of the impact location of the ball during the serve and the BH of the player seem to be the most important factors affecting the capacity to produce high-velocity serves16 (Table 1). Biomechanically, hitting the ball in a higher spot increases the available space toward the opponent’s serve box. Because of this, hitting the ball in higher locations allows the player to offer a more optimal trajectory and achieve a higher SV.16 Literature has identified BH, AL, and jumping height as the main conditioning factors affecting the height of ball impact7,16,25 and therefore are highly related to the capacity of a player to serve faster. BH, besides being a key factor allowing to achieve higher ball impact points,7,36 has previously been identified as the anthropometric characteristic that mostly affects SV in male participants.5,7,30 Nevertheless, many investigations have mentioned the positive relationship between BH and first5,7,14–16,24 and second14,16 serves in both, male and female competitors. These studies show that the submaximal nature of the second serve and prioritizing control over attaining a greater SV makes this stroke have a lower relation with BH than the first serve.16 In fact, Baiget et al5 found that BH did not correlate significantly with SV in the second serve in male elite players.5 Besides this, the length of the racquet arm complex has proven to have an influence on impact point height and therefore SV.7,15 Fett et al7 found considerable positive correlations between AL and SV in all groups of ages included in their study. In this case, longer limbs would not only increase the point of impact but would give the opportunity to the player to transfer a greater tangential and achieve greater SV.7,14,15
Anthropometric Characteristics Related to SV
| Reference | Description and goals | Gender | n | Age | Level | Variables tested | Correlation |
|---|---|---|---|---|---|---|---|
| Vaverka and Cernosek16 | Determine the association between BH and SV in elite players during Grand Slams | M | 78–84 | — | Elite | BH, cm: 185.0 (7.0) | SF: r = .52 (.06) |
| S1: r = .55 (.07) | |||||||
| S2: r = .37 (.11) | |||||||
| F | 70–78 | — | BH, cm: 173.0 (7.0) | SF: r = .52 (.06) | |||
| S1: r = .52 (.03) | |||||||
| S2: r = .35 (.08) | |||||||
| Bonato et al14 | Investigate the relationship between anthropometric and functional parameters and SV in professional players | M | 8 | 23.1 (3.9) | Elite | BH, cm: 181.8 (4.1) | r first serve = .78* |
| r second serve = ,80* | |||||||
| BM, kg: 79.7 (4.3) | r first serve = −0.22 | ||||||
| r second serve = −0.15 | |||||||
| Sögut40 | Determine possible relations between SV and BH. | M | 16 | 13.81 (1.11) | Junior | BH, cm: 163.3 (11.42) | r = .69 |
| F | 17 | 13.35 (1.37) | BH, cm: 159.5 (0.08) | r = .49 | |||
| Palmer et al24 | To explore the relation between BH, ROM, strength, motor control, and power and SV | M | 42 | 23.9 (5.82) | Competition | BH, cm: 180.2 (7.23) | r = .46** |
| Hayes et al15 | Determine if a relationship exists between anthropometric measures and SV in elite junior tennis players | M | 12 | 16.5 (2.0) | Elite | BM, kg: 66.5 (10.6) | r = .68* |
| BMI, kg/m2: 21.22 (1.5) | r = .31 | ||||||
| BH, cm: 178.2 (9.9) | r = .80** | ||||||
| F | 9 | 16.0 (2.2) | BM, kg: 63.9 (6.5) | r = .68* | |||
| BMI, kg/m2: 22 (1.3) | r = .31 | ||||||
| BH, cm: 170.3 (4.6) | r = .80** | ||||||
| Fett et al7 | Determine the impact of anthropometric characteristics on SV in elite junior tennis players | M (n = 625) | 124 | U12: 11.3 (0.4) | Junior | BM, kg: 38.8 (5.8) | r = .47** |
| BH, cm: 149.9 (7.7) | r = .40** | ||||||
| BMI, kg/m2: 17.2 (1.4) | r = .36** | ||||||
| 248 | U14: 12.9 (0.5) | Junior | BM, kg: 47.3 (7.9) | r = .55** | |||
| BH, cm: 160.7 (8.3) | r = .52** | ||||||
| BMI, kg/m2: 18.2 (1.7) | r = .38** | ||||||
| 156 | U16: 14.9 (0.5) | Junior | BM, kg: 61.4 (8.7) | r = .57** | |||
| BH, cm: 174.9 (7.5) | r = .51** | ||||||
| BMI, kg/m2: 20.0 (1.8) | r = .40** | ||||||
| 97 | U18: 16.8 (0.5) | Junior | BM, kg: 72.6 (7.0) | r = .44** | |||
| BH, cm: 181.9 (5.8) | r = .31** | ||||||
| BMI, kg/m2: 21.9 (1.5) | r = .32** | ||||||
| F (n = 394) | 78 | U12: 11.4 (0.3) | Junior | BM, kg: 38.7 (6.4) | r = .38** | ||
| BH, cm: 150.0 (6.4) | r = .26 | ||||||
| BMI, kg/m2: 17.1 (1.9) | r = .36** | ||||||
| 171 | U14: 12.9 (0.5) | Junior | BM, kg: 49.0 (7.3) | r = .39** | |||
| BH, cm: 160.9 (7.2) | r = .34** | ||||||
| BMI, kg/m2: 18.9 (1.8) | r = .28** | ||||||
| 90 | U16: 14.8 (0.5) | Junior | BM, kg: 58.2 (6.3) | r = .39** | |||
| BH, cm: 167.6 (6.2) | r = .38** | ||||||
| BMI, kg/m2: 20.7 (1.8) | r = .17* | ||||||
| 55 | U18: 16.7 (0.5) | Junior | BM, kg: 63.5 (6.3) | r = .35** | |||
| BH, cm: 171.5 (6.6) | r = .32* | ||||||
| BMI, kg/m2: 21.6 (1.8) | r = .12 | ||||||
| Sögut41 | Determine various anthropometric and functional attributes and their relationship with SV | F | 12 | 16.4 (1.1) | Junior | 169.6 (5.8) | r = .331* |
| Fernandez-Fernandez et al23 | Establish the relation between physical and anthropometric variables and SV | M | 32 | U13: 12.6 (0.2) | Junior | BH, cm: 154.9 (7.0) | r = .549* |
| BM, kg: 43.5 (6.8) | r = .625* | ||||||
| 36 | U15: 14.6 (0.3) | Junior | BH, cm: 169.0 (5.7) | r = .594* | |||
| BM, kg: 58.4 (7.3) | r = .600* | ||||||
| F | 32 | U13: 12.6 (0.3) | Junior | BH, cm: 159.8 (7.0) | r = .369 | ||
| BM, kg: 49.1 (7.3) | r = .489* | ||||||
| 28 | U15: 14.6 (0.3) | Junior | BH, cm:166.3 (5.7) | r = .319 | |||
| BM, kg: 56.8 (5.4) | r = .066 | ||||||
| Baiget et al5 | To analyze the associations between SV and anthropometric, ball impact and landing location parameters in total and fastest serves in professional tennis players during an ATP Tour event | M | 21 | 26.4 (5.4) | Elite | BH, cm:186.9 (7.4) | r first serve = .503* |
| r second serve = .486* | |||||||
| BM, kg: 81.6 (7.1) | r first serve = .593* | ||||||
| r second serve = .466* | |||||||
| BMI, kg/m2: 23.4 (1.1) | r first serve = .263 | ||||||
| r second serve = .125 |
Abbreviations: ATP, Association of Tennis Professionals; BH, body height; BM, body mass; BMI, body mass index; F, female; M, male; ROM, range of motion; S1, average first SV; S2, average second SV; SF, fastest serve in a match; SV, serve velocity; U12, under 12; U14, under 14; U16, under 16; U18, under 18. Note: Values are presented as mean (SD).
*P < .05. **P < .01.
Further anthropometric characteristics such as BM and BM index (BMI) have also been studied in relation to SV and have found certain correlation between these parameters. Fett et al,7 Hayes et al,15 and Baiget et al5 found important relationships between BM and SV. Also, Fett et al7 and Wong et al17 found positive correlations between BMI and SV, while no significant relations between this variable and SV were found in elite players in either the first or the second serve in Baiget et al.5 In terms of an athlete’s capacity of being able to produce strength levels, and following allometric theory,17 an increment in BM in accordance with BH is traduced into an increment in torque production. Consequently, greater BM or BMI may assist in the capacity of producing faster strokes and increasing SV, always considering that an increment in these variables without a close control on lean mass and fat ratios could negatively affect agility and change of direction speed.5 Contrary to BH, BM has shown strong relations to SV in female participants,7,23 In this line, and considering the advantages of producing greater strength levels, it seems tennis demands of female competition (ie, lower stroke frequency)37 would tolerate profiles to shift toward players with a tendency to endomorph body types. Male competitors on the other hand may rely more thoroughly on BH and other physical factors influencing SV.7,15 A factor to consider is that BM and BMI are modifiable parameters from training,19,20 and it is suggested that the optimization of programs could have positive effects on performance, always considering the detrimental effect on speed and agility a nonoptimal program of these characteristics could have.
Strength, Power, and ROM
Beyond the importance of anthropometric parameters in achieving greater SV, knowledge around strength factors affecting this stroke has also received thorough attention in literature. Studies have established different strength aspects as determinants of SV (Table 2). Initially, maximal dynamic strength needs during strokes seem to be low,20 as the weight of the implement (ie, racquet) ranges from 200 to 400 g and in this line, studies assessing maximal dynamic strength via bench press or overhead press have not found strong associations between this variable and SV.16 Because of this, investigations have typically aimed at analyzing MIS values at specific joint angles observed throughout the kinetic chain, involving upper and lower body structures. Most of these studies concluded that the main contributor to a greater SV is shoulder internal rotation,31,38 although positive relations were found between MIS and SV in most arm positions tested involved in the serve kinetic chain,15,21,22 being the wrist flexion, extension, and shoulder flexion the movements with stronger associations. Notwithstanding, Baiget et al21 considered these positive correlations present in specific positions and involving MIS of few muscle groups would not be a strong predictor of SV by themselves but only accounted for one piece of the puzzle. In this line, authors perform a multiple regression analysis indicating a 55% of SV variability could be explained by the combination of shoulder internal rotation and shoulder flexion MIS. Added to this, besides the combination of different joint positions and movements involved in the serve, certain strength levels regarding RFD and IMP may also positively influence SV. Baiget et al22 investigated the influence of RFD at different time intervals (ie, 0–250 ms) alongside IMP on SV. Authors conclude that the ability to produce force rapidly (RFD) and the accumulation of force over a given period (IMP), especially in rotational movements, seem to be more determinant than MIS to generate high-velocity serves. As the authors point out, although the early phases (<250 ms) of RFD in the shoulder internal rotation account for roughly 50% of SV variability, the multiple regression analysis showed other shoulder positions (ie, shoulder and wrist flexion) and MIS as important contributors to faster serves. Therefore, while all mentioned aspects seem important contributors to velocity production, the combination and interaction of these variables, alongside anthropometric characteristics and technical capabilities seem to determine the capacity of a player to produce fast serves.
Strength, Power, and ROM Variables Related to SV Performance
| Reference | Description and goals | Gender | n | Age | Level | Variables tested | Correlation (r) |
|---|---|---|---|---|---|---|---|
| Cohen et al38 | Determine the relation between strength variables, ROM, and SV in competition players | M | 40 | 33.7 (7.1) | Competition | Elbow extension torque | .474** |
| Dominant wrist flexion ROM | .338* | ||||||
| Shoulder internal rotation ROM | .324* | ||||||
| 60° shoulder internal rotation eccentric contraction | 361* | ||||||
| 60° shoulder internal rotation concentric contraction | .372* | ||||||
| 180° shoulder internal rotation eccentric contraction | .310* | ||||||
| 180° shoulder internal rotation concentric contraction | .335* | ||||||
| Pugh et al39 | Study the relation between lower body, shoulder, and grip strength and SV in college players | M | 15 | 20.8 (2.0) | Competition | Knee extension strength | .36 |
| Shoulder internal rotation strength | .29 | ||||||
| Grip strength | .41 | ||||||
| Signorile et al6 | Examine the correlations between isokinetic peak torque and SV | M (n = 23) and F (n = 10) | 33 | 14.97 (1.36) | Junior | Diagonal throwing peak torque | .69** |
| Wong et al17 | Investigate the effects of kinematics on SV in elite players | M | 12 | 20.5 (3.8) | Elite | Knee ROM during phases I and II of the serve | .705* |
| Knee extension velocity during phase II of the serve | .751** | ||||||
| Peak hip extension speed during phase II of the serve | .657* | ||||||
| Shoulder ROM during phase III of the serve | .616* | ||||||
| Peak elbow extension velocity during phase II of the serve | .708** | ||||||
| Baiget et al21 | Investigate the relation between maximal isometric strength and SV in competition players | M | 12 | 17.2 (1.0) | Junior | Shoulder internal rotation maximum isometric strength | .67* |
| Shoulder internal rotation ± shoulder flexion maximum isometric strength | .76* | ||||||
| Hayes et al.15 | Determine if there is a relation between IMTP, CMJ, BH, shoulder internal and external rotation strength and SV in elite adolescent players | M (n = 12) and F (n = 9) | 21 | M: 16.5 (2) F: 16.0 (2.2) | Junior | IMTP peak strength | .87** |
| CMJ height | .77** | ||||||
| IMP at 300 ms | .71** | ||||||
| IMP at 200 ms | .58** | ||||||
| IMP at 100 ms | .64** | ||||||
| 90° Shoulder internal rotation | .63** | ||||||
| <90° Shoulder external rotation | .63** | ||||||
| Dossena et al25 | Investigate the relationship between jumping capacity and SV in professional tennis players | M | 8 | 20 (3) | Competition | Maximal jumping height during first serve | .71* |
| Maximal jumping height during second serve | .71* | ||||||
| Palmer et al24 | Determine if upper and lower body power variables are predictive of SV in elite players | M | 42 | 23.9 (5.82) | Elite | Hip external rotation ROM | .39** |
| Single leg hop (ipsilateral) | .36* | ||||||
| Single leg hop (contralateral) | .31* | ||||||
| Dominant arm seated shot put throw | .30* | ||||||
| Eriksrud et al40 | Determine the relationship between power, strength, and dynamic balance and SV in competition players | M | 12 | 28.3 (10.3) | Competition | CMJ | .715** |
| Dominant arm vertical press | .650* | ||||||
| Bilateral arm overhead anterior push | .643* | ||||||
| Fett et al7 | Determine the relationship between strength and power variables and SV in elite junior players | M (n = 625) | 124 | U12: 11.3 (0.4) | Junior | Grip strength | .43** |
| MBO | .55** | ||||||
| MBF | .49** | ||||||
| MBB | .55* | ||||||
| 248 | U14: 12.9 (0.5) | Junior | Grip strength | .59** | |||
| MBO | .52** | ||||||
| MBF | .63** | ||||||
| MBB | .58** | ||||||
| 156 | U16: 14.9 (0.5) | Junior | Grip strength | .59** | |||
| MBO | .60** | ||||||
| MBF | .58** | ||||||
| MBB | .60** | ||||||
| 97 | U18: 16.8 (0.5) | Junior | Grip strength | .57** | |||
| MBO | .52** | ||||||
| MBF | .55** | ||||||
| MBB | .51** | ||||||
| F (n = 394) | 78 | U12: 11.4 (0.3) | Junior | Grip strength | .37** | ||
| MBO | .20* | ||||||
| MBF | .29* | ||||||
| MBB | .21 | ||||||
| 171 | U14: 12.9 (0.5) | Junior | Grip strength | .36** | |||
| MBO | .39** | ||||||
| MBF | .56** | ||||||
| MBB | .50** | ||||||
| 90 | U16: 14.8 (0.5) | Junior | Grip strength | .34** | |||
| MBO | .54** | ||||||
| MBF | .59** | ||||||
| MBB | .60** | ||||||
| 55 | U18: 16.7 (0.5) | Junior | Grip strength | .27* | |||
| MBO | .48** | ||||||
| MBF | .51** | ||||||
| MBB | .38* | ||||||
| Fernandez-Fernandez et al23 | Analyze the functional profile of the shoulder and establish relations between the tested variables and SV | M | 32 | U13: 12.6 (0.2) | Junior | MBO | .557 |
| MBF | .638* | ||||||
| MBB | .442* | ||||||
| 36 | U15: 14.6 (0.3) | Junior | MBO | .418* | |||
| MBF | .582* | ||||||
| MBB | .532* | ||||||
| F | 32 | U13: 12.6 (0.3) | Junior | MBO | .433* | ||
| MBF | .295 | ||||||
| MBB | .307 | ||||||
| 28 | U15: 14.6 (0.3) | Junior | MBO | .202 | |||
| MBF | .413* | ||||||
| MBB | .098 | ||||||
| Colomar et al18 | Study the influence of strength, power, and muscle stiffness on SV in junior tennis players | M | 21 | 17.0 (0.8) | Junior | Gastrocnemius stiffness | .45* |
| Infraspinatus stiffness | .42* | ||||||
| Baiget et al22 | To analyze the associations between SV and various single-joint upper limb isometric force time parameters (IF, RFD, and IMP) | M (n = 12) and F (n = 5) | 17 | 16.8 (1.1) | Junior | IF 30 ms | .01–.49 |
| WF IF 50 ms | .54* | ||||||
| WE, WF, SHF IF 90 ms | .49–.56* | ||||||
| WE, WF, SHF, SHIR IF 100 ms | .52–.58* | ||||||
| WE, WF, SHF, SHIR IF 150 ms | .5–.67** | ||||||
| WE, EE, SHE, WF, SHF, SHIR IF 200 ms | .51–.7** | ||||||
| WE, EE, SHE, WF, SHF, SHIR IF 250 ms | .54–.72** | ||||||
| Peak IF | .54–.7** | ||||||
| WE, RFD 0–30 ms | .66** | ||||||
| WE, SHE, SHIR RFD 0–50 ms | .52–.69** | ||||||
| WE, SHE, SHF, SHIR RFD 0–90 ms | .5–.69** | ||||||
| WE, EE, SHE, SHF, SHIR RFD 0–100 ms | .49–.69** | ||||||
| WE, EE, SHE, WF, SHF, SHIR RFD 0–150 ms | .5–.7** | ||||||
| WE, EE, SHE, WF, SHF, SHIR RFD 0–200 ms | .58–.69** | ||||||
| WE, EE, SHE, WF, SHF, SHER, SHIR RFD 0–250 ms | .5–.71** | ||||||
| IMP 30 ms | .04–.48 | ||||||
| WF IMP 50 ms | .5* | ||||||
| WF IMP 90 ms | .58* | ||||||
| WF IMP 100 ms | .59* | ||||||
| WE, SHE, WF, SHF, SHIR IMP 150 ms | .52–.64** | ||||||
| WE, EE, SHE, WF, SHF, SHIR IMP 200 ms | .5–.66** | ||||||
| WE, EE, SHE, WF, SHF, SHIR IMP 250 ms | .52–.66** |
Abbreviations: CMJ, countermovement jump; EE, elbow extension; F, female; IMP, impulse; IMTP, isometric midthigh pull; M, male; MBT, medicine ball throw; MBB, MBT backhand; MBF, MBT forehand; MBO, MBT overhead; RFD, rate of force development; ROM, range of motion; SHE, shoulder extension; SHF, shoulder flexion; SHER, shoulder external rotation; SHIR, shoulder internal rotation; SV, serve velocity; U12, under 12; U14, under 14; U16, under 16; U18, under 18. Note: Values are presented as mean (SD).
*P < .05. **P < .01.
As greater upper body strength and power levels seem to positively influence SV, the role of lower body values is not as clear. The elevation the body experiences with respect to the floor when extending ankles, knees, and hips affects the height of the ball impact spot.3,25 Following this idea, it could be considered beneficial to have greater strength and power levels in the lower body that could derive into higher impact points and therefore increase SV. In any case, the low relationship between SV and countermovement jump assessments, or leg extension maximal isometric contractions18,21,25 indicates that the influence of this variable may be relatively low. Authors emphasize the differences between both motions and suggest the introduction and use of more specific jumping tests that include both upper and lower body (ie, sergeant jump). Literature seems to agree to grant the lower body a coordinating role in the serve motion, most likely linked to coordination and technique rather than affecting SV by themselves. However, some studies have found a positive effect of lower body strength and power parameters and SV, showing knee extension velocity,17 knee extension strength,39 isometric mid-thigh pull test,15 jumping height in a countermovement jump,15,42 hop tests,24 or even the level of stiffness of the gastrocnemius muscle18 as predictors of velocity in this stroke. These studies give importance to the role of ground reaction forces (GRFs) and the ability to transfer energy to the upper segments of the body. As higher power levels in the lower limbs seem to relate to generating greater GRF,9,24 these parameters would also be beneficial for SV. Because of these reasons, although the lower body seems to have a more coordinative role than a velocity generator, greater strength and power levels could favor an appearance of GRF of greater magnitude and, if the transfer throughout the kinetic chain is effective, SV would be enhanced. Regarding the stabilization functions and transfer of the generated GRF, the trunk is considered essential toward effective serving. Some studies43,44 agree in granting this region not only a coordinative role, but also as a force transfer link in the kinetic chain. Although literature has generally not investigated the relationship between trunk power and strength levels and SV, Wong et al17 found that peak velocity of hip extension positively influenced SV.
Alongside isometric force–time curve values that positively correlate to SV, literature considers power and the effective use of the stretch-shortening cycle as highly specific indicators of velocity production capacity.7,23 The technical execution of a serve implies a prestretching of most of the muscle groups involved in the motion, being the elastic energy storage and rebound capacity of the muscle of great importance for the action. MBTs have been proven to be a useful tool to assess upper-body power.7,23 This type of assessment allows the summation and transfer of forces throughout the entire kinetic chain and is considered an interesting method to obtain values of power in tennis-specific motions. A great number of studies have found positive correlations between MBT distance or speed and SV or even other tennis strokes.7,23,45,46 Fernandez-Fernandez et al23 point out that MBT distance is an important predictor of SV in male tennis players. Fett et al7 show that power values established from MBT is one of the best predictors for SV, in both male and female competitors, especially as age advances. In younger players, although distance in MBT could be useful to predict SV, this would present a stronger interaction with SV in male competitors.7,23 Added, it has previously been hypothesized that the influence of these abilities seems to rise in importance as the players level and age increase.7,18 These investigations theorized that technical and coordinative aspects seem more relevant in young inexperienced players, as physical factors might become more important as technical capacity is solid in all performers.
Besides strength and power values and measurements, ROM of joints involved in the serve motion have shown important relationships with SV.17,24,38 Similar to strength and power variables, an increased ROM around joints that are greatly involved in the serve kinetic chain seems relevant to enhanced velocity production. Shoulder internal rotation, wrist flexion, and hip external rotation have previously shown positive associations with SV,17,24,38 establishing the ability to achieve necessary movement degrees as relevant to improve this variable. In this line, coaches are encouraged to guarantee high levels of ROM in the mentioned joints toward performance increases.
Factors Negatively Affecting SV
The previously discussed physical parameters positively related to greater SV may be altered by tennis match play. As these variables are directly linked to the multifactorial nature of the tennis serve, fatigue is considered as a triggering aspect negatively influencing SV. Metabolic exhaustion, muscle impairment, soreness, and functionality are directly related to a descent in muscular strength29 and have the potential to negatively affect SV. More specifically, literature indicates that the main performance aspect negatively affected by fatigue is precision.47–49 Davey et al47 and Rota et al49 found reductions in serve accuracy after performing a maximal intermittent activity (−30% and −11.7%, respectively), attributed to lactate accumulation. Added to effects on precision, fatigue in certain regions and on determinant strength and power variables seem to be main contributors to decreases in SV.47 Notwithstanding, this fatigue does not seem to affect all players in the same way and is most likely determined by match load, experience, and playing level. Terraza and Baiget50 found no reductions in accuracy or SV following a resistance training or MBT protocol, suggesting although impairments could have appeared in strength and power levels, players may rely on different neuromuscular parameters to maintain performance during the serve. Maquirrain et al (2016) and Moreno-Pérez et al30 did not observe reductions in precision or speed of elite tennis players after 5-set matches. On the contrary, studies have found reductions in SV (3.9%–4%) in competition players of lower level49–52 or age,53 suggesting experienced athletes could be able to find strategies to avoid the reduction of SV in fatiguing situations. Nevertheless, investigations are limited when examining the influence of fatigue on SV in young competitors, making of great interest further studies on the topic. Research has not uniquely focused on fatigue caused by the direct outcome of tennis practice or competition but has investigated the effect of prolonged play or repetitive bouts of play on SV (Table 3). In this line, the organization model tennis follows has proven to negatively affect SV.26,48 Gallo-Salazar et al26 found reductions in SV attributed to playing 2 tennis matches in one same day. One of the main reasons these decreases happen is the loss of functionality around the shoulder region caused by activities maintained and repeated in short periods of time.30,51 These studies show shoulder strength deficits and ROM impairments in internal and external rotation values after performing a certain volume of tennis play. Authors agree and recommend the application of intervention programs including strategies to reestablish values before competition or practice, especially in players without a sufficient experience and level to take advantage of technical proficiency or tactical decisions to replace reductions in SV.
Factors Negatively Affecting SV
| Reference | Description and goals | Gender | N | Age | Level | Fatiguing condition | Findings |
|---|---|---|---|---|---|---|---|
| Davey et al47 | Examine the effect of fatigue on specific sporting abilities | M (n = 9) and F (n = 9) | 18 | 20.7 (0.9) and 21.7 (0.6) | Competition | High-intensity intermittent activity | SV was not reduced but precision declined 30%* |
| Ojala and Hakkinen54 | To examine changes in selected physiological and performance variables during a 3-day tennis tournament | M | 8 | 23 (3.8) | Elite | 3-d tennis tournament | SV was significantly lower before the third match compared with the first match (−2.72%*) |
| Rota et al49 | Examine the effect of fatigue on upper body muscular activity and tennis performance | M | 10 | 23.8 (4.0) | Competition | High-intensity intermittent-specific activity | SV is reduced 4.5%* and precision 11.7%* |
| Gescheit et al48 | Determine how playing matches in consecutive days affects performance, physiological and perceptual responses | M | 7 | 21.4 (2.2) | Competition | Matches on consecutive days | SV is moderately increased day by day. Precision decreases during consecutive days |
| Maquirrain et al55 | Analyze SV and accuracy in prolonged male professional matches played on grass courts | M | 30 | — | Elite | Wimbledon 5 set matches | No significant changes were registered |
| Martin et al51 | Examine changes in shoulder ROM and SV during a 3-h tennis match | M | 8 | 20.4 (2.8) | Competition | 3-h match | 1.8 m/s (−3.9%*) reductions after 3 h of play. No reductions at 90 min |
| Gallo-Salazar et al50 | Analyze how playing 2 consecutive matches on the same day affects performance in young tennis players | M | 12 | 14.4 (0.9) | Junior | Playing 2 matches in the same day | Trivial reductions in SV from 151.7 (13.94) to 149 (15.09) km/h |
| Moreno-Pérez et al30 | Determine the acute effects of a tennis match on SV and shoulder ROM | M | 26 | 20.4 (4.4) | Elite | One tennis match | No significant decreases in SV |
| Tooth et al49 | Assess the influence of scapular muscle fatigue on tennis performance | M | 15 | 22.8 (3.45) | Competition | Elastic band exercise until exhaustion on racquet velocity | Significantly decreased racquet velocity (4%*) and accuracy (55%**) |
| Terraza-Rebollo and Baiget (2021) | Acute and delayed effect of strength training on SV and accuracy | M (n = 4) and F (n = 6) | 10 | 15.3 (3.45) | Competition | MBT or resistance training exercises | No significant reductions in SV or accuracy |
Abbreviations: F, female; M, male; MBT, medicine ball throws; ROM, range of motion. Note: Values are presented as mean (SD).
*P < .05. **P < .01.
Practical Applications
- •Many serving models are available toward optimizing SV. However, knee extension and lower leg drive, hip and trunk rotation, and elbow extension, shoulder internal rotation, and hand/wrist flexion seem the major contributors to angular momentum and transferring speed to the head of the racquet and to the ball. These indications should be encouraged by coaches toward technical proficiency.
- •The BH and AL seem highly important for the tennis player, as the capacity of reaching higher ball impact locations seems to correlate strongly with SV. Nevertheless, more trainable aspects such as shifting body composition toward a greater lean BM may have positive influence on SV, making interesting training options aiming at this goal. However, individual needs should be considered.
- •Force–time curve parameters (MIS, RFD, and IMP) around the shoulder joint are good predictors of SV across sexes and especially as age and level increase. Coaches are encouraged to include strength training programs that cover the whole load–velocity curve spectrum. In any case, special attention to maximal velocity intention while performing the program seems essential to achieve desired gains.
- •Intense match play performed regularly have the capacity to reduce SV and accuracy. Elite and experienced players seem to be able to maintain SV relying on other aspects involved in the execution of an optimal serve (ie, ROM, technique, or tactical decisions), but repetition of competitive bouts or intense match play will most likely end up negatively influencing SV. Thus, effective recovery strategies to reestablish initial strength and power values as soon as possible should be implemented, especially in younger and inexperienced populations in which the negative outcome could be more evident.
Conclusions
Fast serving certainly needs a well-developed technical ability. Toward this goal, coaching methods and literature have provided specific serving models intending to achieve greater velocity production and efficient motion. However, depending on which model followed and the specific phase of the action, technical requirements may vary. Yet, aspects such as lower leg drive, hip and trunk rotations, and upper arm extension and internal rotations seem the major contributors to racquet and ball speed. Regarding anthropometric characteristics favoring SV, a higher impact point achieved by BH or AL, alongside a greater lean BM seem to aid faster serves. Strength and power levels such as MIS and RFD in joint positions involved in the serve kinetic chain, alongside upper body power have previously been positively correlated to faster SV and should be developed in a velocity production enhancement program. Notwithstanding, the effects of continued or repetitive competition loads may impair the above-mentioned key physical factors and negatively influence SV, especially in younger unexperienced players.
References
- 1.↑
Fitzpatrick A, Stone JA, Choppin S, Kelley J. Important performance characteristics in elite clay and grass court tennis match-play. Int J Perform Anal Sport. 2019;19(6):942–952. doi:10.1080/24748668.2019.1685804
- 3.↑
Kovacs MS, Ellenbecker TS. A performance evaluation of the tennis serve: implications for strength, speed, power, and flexibility training Strength Cond J. 2011;33(4):22–30. doi:10.1519/SSC.0b013e318225d59a
- 5.↑
Baiget E, Corbi Soler F, López J. Influence of anthropometric, ball impact and landing location parameters on serve velocity in elite tennis competition. Biol Sport. 2022;40(1):273–281. doi:10.5114/biolsport.2023.112095
- 27.
Gescheit DT, Cormack SJ, Duffield R, et al. A multi-year injury epidemiology analysis of an elite national junior tennis program. J Sports Sci Med. 2019;22(1):11–15. PubMed ID: 29958840 doi:10.1016/j.jsams.2018.06.006
- 28.
Girard O. Changes in exercise characteristics, maximal voluntary contraction, and explosive strength during prolonged tennis playing. Br J Sports Med. 2006;40(6):521–526. PubMed ID: 16720888 doi:10.1136/bjsm.2005.023754
- 31.↑
Elliott B, Marshall R, Noffal G. Contributions of upper limb segment rotations during the power serve in tennis. J Appl Biomech. 1995;11(4):433–442. doi:10.1123/jab.11.4.433
- 32.↑
Sweeney M, Reid M, Elliott B. Lower limb and trunk function in the high performance tennis serve. Asian J Exerc Sports Sci . 2012;9(1):9.
- 36.↑
Roetert EP, Ellenbecker TS, Reid M. Biomechanics of the tennis serve: implications for strength training. Strength Cond J. 2009;31(4):35–40. doi:10.1519/SSC.0b013e3181af65e1
- 40.↑
Söğüt M. all speed during the tennis serve in relation to skill level and body height. Pamukkale J Sport Sci. 2016;7(2):13.05.2016.
- 41.↑
Söğüt M, Altunsoy K. Physical and morphological characteristics of Turkish national adolescent tennis players and their association with serve speed. Spor Hekimligi Dergisi. 2018;54(1):64–70. doi:10.5152/tjsm.2019.117
- 42.↑
Eriksrud O, Ghelem A, Henrikson F, Englund J, Brodin N. Upper and lower body power tests predict serve performance in national and international level male tennis players. Sport Perform Sci Rep. 2018;42(1):1–5.
- 46.↑
Terraza-Rebollo M, Baiget E, Corbi F, Planas Anzano A. Efectos del entrenamiento de fuerza en la velocidad de golpeo en tenistas jóvenes/Effects of strength training on hitting speed in young tennis players. Rev Int Med Cienc Act Fis Deporte. 2017;17(66):349–366. doi:10.15366/rimcafd2017.66.009
- 50.↑
Terraza-Rebollo M, Baiget E. Acute and delayed effects of strength training in ball velocity and accuracy in young competition tennis players. PLoS One. 2021;16(12):e0260825. Published December 9, 2021. doi:10.1371/journal.pone.0260825
- 51.↑
Martin C, Bideau B, Delamarche P, Kulpa R. Influence of a prolonged tennis match play on serve biomechanics. PLoS One. 2016;11(8):e0159979. doi:10.1371/journal.pone.0159979
- 52.↑
Tooth C, Schwartz C, Fransolet C, et al. Influence of scapular dyskinesis, kinesiotaping and fatigue on tennis serve performance. Int J Perform Anal. 2020;20(3):456–469. doi:10.1080/24748668.2020.1761672
- 54.↑
Ojala T, Häkkinen K. Effects of the tennis tournament on players' physical performance, hormonal responses, muscle damage and recovery. J Sports Sci Med. 2013;12(2):240–248.
- 55.↑
Maquirriain J, Baglione R. Cardey M. Male professional tennis players maintain constant serve speed and accuracy over long matches on grass courts. Eur J Sport Sci. 2016;16(7):845–849. doi:10.1080/17461391.2016.1156163
