How to train best for tennis performance?

Over a series of reviews, I will try to dig dip and provide you with practical takeaways in terms of physiology and the demands of a certain sport. First up is tennis. We will have a look over elite-level male and female tennis as well as how demands change when we analyze lower levels of play. Tennis is a sport based on unpredictability. The unpredictability of point length, shot selection, strategy, match duration, weather, and the opponent all influence the complex physiological aspects of tennis play.

Introduction and analysis of the match demands

Tennis requires players to perform short bursts of high-intensity exercise interspersed with periods of rest or low-intensity activities for a prolonged period (Fernandez, Mendez-Villanueva, & Pluim, 2006; Kovacs, 2007; Mendez-Villanueva, Fernandez-Fernandez, & Bishop, 2007 in Gomes et al., 2011). This variability requires tennis athletes to be highly trained both anaerobically for performance and aerobically to aid in recovery during and after play. The physiological responses to tennis match-play have been reported to be moderate, with factors such as individual playing style, court surface, and game situation all influencing the response (Fernandez et al., 2006; Kovacs, 2007; Mendez-Villanueva et al., 2007 in Gomes et al., 2011).  In general, these previous studies have reported mean heart rates of 60-80% of maximum (Ferrauti, Weber, & Wright, 2003; Kovacs, 2007) and mean blood lactate concentrations of less than 4.0 mmol/l during matches consisting of three sets (Fernandez et al., 2006; Fernandez-Fernandez, Mendez-Villanueva, Fernandez-Garcia & Terrados, 2007 in Gomes et al., 2011). The running activities of players encompass high accelerations and decelerations but low velocities reflecting the intermittent play involved in tennis, which does not allow high velocities to be reached (Hoppe et al., 2014).

Therefore it is important to understand the nature of the sport and train the energy systems that predominate during match play. Below I am presenting an overview of physiological demands accumulated from different studies done on tennis match play. Those reports are conducted on best of 2 sets matches.

Tennis demands in an infographic:

Figure 1: Basic analysis of tennis match play. Adopted from Kovacs (2006) and Fernandez-Fernandez et al. (2009).
Figure 2: Basic analysis of tennis match play. Adopted from Kovacs (2006) and Fernandez-Fernandez et al. (2009).
  • The percentage of the playing time with respect to the total time of the match (on clay courts) has been shown to be approximately 21% for the attacking players, 28.6% for whole court players, and 38.5% for baseline players (Kovacs, 2006). In an earlier study, the percentage of playing time during match, on hard courts, was approximately 20%. From the research, it appears that total playing time is only between 20% and 30% of total match time.
  • When the player in control of the rally was an attacking player (hits the tennis ball hard and attempts to come to the net consistently), the average duration of the points was found to be 4.8 seconds. Rally duration varied between 6 and 11 seconds (mean 8.2s) when the player in control of the rally was a whole court player (who plays from the baseline, but is very comfortable coming to the net). The points lasted on average 15.7s when the player in control of the rally was a baseline player (plays the large majority of points from the baseline, hitting ground strokes, and does not prefer to come into the net) (Kovacs, 2006).
  • It has been reported that VO2 values – both at submaximal and maximal loads – were moderate predictors of players competitive ranking (Brechbühl et al., 2016), and that better aerobic conditioning of male tennis players at international levels were associated with better technical efficiency at higher exercise intensities compared with male tennis players at national levels (Baiget et al, 2016).
Figure 3: Percentage of time spent in low- (open), moderate- (grey), and high-intensity (black) heart rate (HR) zones for each set, and the overall tennis match (adopted from Gomes et al., 2011).
How about best of 5 matches at the elite level?

However, the major international tournaments (e.g. Grand Slam events and Davis Cup) are determined by the best of five sets (the first player to win three sets wins the match) with the longest matches lasting for more than 5 h. One such study was done by Gomes, R. V., Coutts, A. J., Viveiros, L., & Aoki, M. S. (2011) where they recreated a best of 5 tennis match with 2 elite Brazilian tennis players.

What was done?

  • Basic information and measurements of the players:
Figure 4: Information of elite-level ATP players in a best of 3 tennis match analysis from Gomes et al., 2011.


Results and changes in a best of 5 match on played on hard court at 26-27.5°C and 66-70% humidity:

Figure 5: Results of a best of 3 tennis match analysis from Gomes et al., 2011.


  • Players may adopt a different playing strategy in an attempt to cope with increasing physiological and perceptual stress – decrease in rally length as the match progressed,
  • In the 3rd and 4th set more than 60% of the rallies had two strokes or less, whereas in the first two sets most rallies were consisting of 2-4 strokes,
  • ‘Cardiovascular drift’ may occur during prolonged tennis match-play and highlight the importance of appropriate hydration strategies when playing in the heat,
  • Blood lactate concentration decreased during the fourth set in both players, despite maintaining blood glucose concentration and having elevated salivary cortisol, higher HR and shorter rallies in 4th set (showing an increase in fatigue),
  • Hypoglycemia does not manifest during match-play,
  • Prolonged tennis match-play is associated with increased fatigue and that these elite players may adjust work rates or tactics to cope with the increased perception of effort,
  • This puts greater demands and focus on pre-, within-, and post-match recovery strategies (fluid and carbohydrate replenishment & recovery interventions)
Energy Systems & Cardiorespiratory Responses
  • Although heart rate is an easily measured index of intensity, it should not be used as the sole measurement of metabolism, as this would not accurately represent the physiological nature of an intermittent sport such as tennis,
  • Heart rate remains significantly increased above pre-exercise levels despite the varying intensity and intermittent nature of the game variability and ranges during a match are rather wide owing to the continual stop/start movements and explosive nature of the sport.
  • It has been suggested that VO2max values >50 ml/kg/min for males and > 42 ml/kg/min for females are generally considered as a minimum standard and preferably a higher value is encouraged for tennis athletes to be able to practice and compete at a high level.
  • During competitive matches, mean HR values range between 60–80% of maximum HR (HRmax), with long and intense rallies eliciting values over 95% of HRmax (Fernandez-Fernandez et al., 2009).
  • Some previous studies have suggested tennis to be an aerobic sport because of the duration and moderate mean heart rate values during play, however, the explosive nature of the serve and ground strokes, the rapid changes of direction which requires a high anaerobic capacity, and the requirement for a high percentage of fast-twitch muscle fibers do not represent typical aerobic focused activities.
  • It might be better to classify it as a anaerobic predominant activity requiring high levels of aerobic conditioning to avoid fatigue and aid in recovery between points and between matches. Interestingly, aggressive attacking players had lower VO2 values during play than baseline players.
  • Based on some studies lactate (LA) production training (drills between 15 and 50 seconds) should also be carried out regularly with the goal of improving the player’s ability to perform high-intensity exercise for longer periods. Specific on-court movements (preferably without the racket) are preferred to ensure the attainment of the desired (maximal) intensities and that local muscle adaptations can be fully transferred to actual match play (Fernandez-Fernandez et al., 2009).

Although heart rate is easily measured, it should not be used as the sole measurement of metabolism, as this would not accurately represent the physiological nature of an intermittent sport such as tennis.

Speed, COD & Agility
  • Tennis is a game of continual emergencies because, with every shot the opponent hits, a ball can have a different velocity, a different type and rate of spin, be placed in many different parts of the court.
  • It requires tennis athletes to have fast reaction times and explosive ‘‘first step’’ speed. Tennis players need to be exceptional movers in linear and lateral (multidirectional) movements.
  • Thus it is important to train tennis players in the specific movement patterns that are encountered during match play. However, one must not confuse game specificity for physiological capabilities, some older studies have suggested that a program consisting of stop-start sprints of no more than 20 meters would be appropriate I strongly disagree since it has been shown multiple times (for reviews see Haugen et al., 2019) that maximal velocity development only occurs when training for, well, maximal velocity sprinting. So the line of thinking that sprint activities that are no longer than the furthest distance that the athlete would run, per shot, during a point are unnecessary are short-sighted and should not make their way into papers, neither training programs of strength and conditioning coaches working with tennis players.
  • Training should consist of both, acceleration focus training and maximal velocity development, especially for younger tennis athletes. An approach that consists of appropriate running/sprinting technique, volume, intensity, and rest is a must if the goal is to develop a player in terms of speed.

Traning should consist of both, acceleration focus training and maximal velocity development, especially for younger tennis athletes.

Strength & Power

  • Strength is required in muscles and joints both for performance enhancement  (ball velocity, quality, and speed of movement) and to reduce injuries (protection of joints, ligaments, tendons, etc.).
  • Shoulder: In the tennis serve the greatest contribution to the final speed of the racket head was (in order of importance): upper arm internal rotation, wrist flexion, upper arm horizontal adduction, forearm pronation, and forward movement of the shoulder. The shoulder region is highly involved in all tennis strokes, and it has been shown that shoulder internal, external, and diagonal peak torques contribute substantially to service ball velocity. Eccentric muscular contractions play a role in functional activities, but in the tennis shoulder—specifically, the rotator cuff muscles (infraspinatus and teres minor) are of major importance during the follow-through of the service motions as well as ground strokes. These two muscles undergo high decelerative eccentric muscle contractions to preserve healthy joint movement.  Adequate strength and range of motion (ROM) in the rotator cuff muscles are essential in preventing overhead overuse injuries as they are vital in stabilizing and movement throughout the extreme ROM experienced during the serve and ground strokes. The speed of the serve or throwing motion depends partly on a rapid and forceful concentric internal rotation in the acceleration phase of the serve. The eccentric phase of training may specifically affect the decelerative phase, which may determine the trajectory and velocity components of performance. It is recommended that tennis athletes include both concentric and eccentric shoulder training in their training programs for performance improvement (Kovacs, 2006).
  • Grip: Solid contact between the racket and the ball is required for optimum stroke execution, and this is influenced by grip strength. A firm wrist is necessary to prevent the racket head from straying from its intended path under the influence of high angular speeds and torques. A maximum grip strength of 600N has been reported in elite-level tennis players, as well as greater grip endurance compared with non-players, although grip strength and grip endurance were not well correlated and should be trained separately.
  • Lower Body: The majority of tennis injuries have been reported to occur in the lower body. Unlike the asymmetrical differences seen in upper body strength, lower body strength measures have been shown to be symmetrical in tennis players. Players should undertake both bilateral and unilateral strength exercises to improve performance and reduce the risk of injury (Kovacs, 2006).
  • What we can conclude from this information is that a tennis player should be involved in progressive strength training from a young age, at first to develop movement proficiency and technique in a plethora of movements. Later on, with maturation, the focus turns on the progressive overload of the involved joints and musculature, both for the lower (quads, hamstrings, glutes, hip muscles; calf, ankle, and foot muscles) as well as the upper (upper back, chest, arms, rotator cuff) body. The involved muscles and their functioning should concurrently develop in terms of force as well as velocity (i.e., muscle mass gained is capable of executing high velocity movements).

A tennis player should be involved in progressive strength training from a young age, at first to develop movement proficiency and technique in a plethora of movements. Strength and power training should develop muscles in terms of force as well as velocity concurrently.

  • Both male and female tennis athletes have been shown to have a smaller range of internal shoulder rotation and a greater range of external shoulder rotation in their dominant arm than other athletes. The major reason for this is probably the repetitive service action which increases the external ROM—a possible performance benefit. There has been speculation that if the imbalance is not improved, itmay lead to muscle and joint injury in the medium to long term. With the asymmetries and imbalances being a hot topic still today, there is no proof that such detriments in fact do occur. Just to be on the safe side, a good rule of thumb is to try to stay inside the 10-15% difference.
  • Playing tennis alone, without any external shoulder ROM training, is not enough to improve shoulder ROM, which could increase performance. Thus tennis athletes should undertake a shoulder ROM program (Kovacs, 2006). I believe that ROM improvements should come with a good, well-rounded, and individualized training program.
  • Lower back pain and injury are common complaints among elite tennis players, and this correlates with poor lower back and hamstring ROM (Kovacs, 2006). Tennis players have been shown to have a smaller ROM in both hamstrings than other athletes, but an even poorer ROM in their back leg, while serving. This poor hamstring ROM may be explained by the need for tennis players to be in the typical ‘‘low ready position.’’ This is the most efficient starting position for explosive movement, because of the lowered center of mass, but it does require the athlete to have the hamstring in a shortened contracted position for long periods. I have no proof, but I don’t believe that this is actually the case, since correlation does not mean causation, the reason for back pain is most likely somewhere else (load management, quality of movement, quality of rest & recovery, psychological factors, etc.).
  • Flexibility needs to be individualized. If the ROM is sufficient, excessive flexibility training may induce negative benefits (reduced power output). Thus training time may be spent productively in maintaining flexibility and focusing on other training variables than continually trying to improve ROM. Like I have mentioned in the strength part, a quality, well-rounded training program should cover all the bases and it should be done both during the pre- and/or off-season as well as during the main competitive season, with proper adjustments that is.

Flexibility training needs to be individualized and based on needs. If ROM Is sufficient, training time may be spent productively focusing on other training variables (strength & power, speed, technique, tactics, etc.). A quality, well-rounded training program should cover all the bases and it should be done both during the pre- and/or off-season as well as during the main competitive season, with proper adjustments that is.


Physical components, age, and rankings

  • Research looking at links between physical fitness components, age, and tennis ranking in competitive junior players. Correlation studies have been undertaken to determine which physical components have a strong relation with match results and ranking (Kovacs, 2006).
    • Only one correlation was found between the results of certain athletic performance test and the sectional rankings of the junior tennis players—the hexagon test (r = 0.23, p=0.05).
    • No correlations were found between the athletes’ tennis stroke rating (rated by highest level USPTA tennis coaches) and physical performance tests.
    • Each stroke was significantly correlated with sectional and national rankings, except for the backhand stroke, which did not have a significant correlation with sectional rankings.
  • The most remarkable finding of this investigation was that the physical performance tests in advanced young male tennis players (8 to 12 years old) did not predict their ability to play tennis at a competitive level. Agility, however, was the physical ability that most influenced the competitive level of young tennis players. Data from this study did indicate that skills related to tennis strokes may be used to predict success at this age (Kovacs, 2006).
  • Basically, there is no proof or point in making any kind of decision for a player’s potential level based on the physical performance tests done or the tests that are being done are not indicative of their future ability. Since the hexagon test is mentioned as the test for agility and the current definition is “agility is ‘‘a rapid whole-body movement with change of velocity or direction in response to a stimulus’’, we can see that the hexagon test does not measure actual agility. What in fact does predict success is, guess what, the quality of tennis strokes itself. So play that has a better stroke is a better player. What a concept…
  • As technical stroke production appears to influence rankings more than physical ability in the younger players, their training should concentrate on effective and efficient stroke mechanics, improving technique and ball placement, with less emphasis on physical conditioning until they reach puberty and beyond (Kovacs, 2006). This one is important because we must ask ourselves what is the goal? Is the goal to have a high national ranking as a youth player, or to produce a quality player on a professional level? If the goal is the latter, then the emphasis must be on both, the physical conditioning part can just be done in less time or as part of the tennis session itself. My suggestion is that up to age 12, 1-2 per week of 60 minutes is more than enough. Playing another sport might be in place of physical conditioning as well.
  • As junior competitive tennis players age, there seems to be an increase in shoulder problems. As age increases, there is a continual decrease in internal ROM (Kovacs, 2006).

As tennis players mature, it seems that the major factors affecting their performance alter.

Power vs Placement

  • The investigators surmised that the players showing the greatest isokinetic strength and perhaps the greatest ball velocity had the most difficult time with ball placement. This received support from a study that found no significant correlations between stroke velocity and accuracy. As would be expected, correlations have been found between shoulder strength and the velocity of tennis strokes (r2 = 0.68) (Kovacs, 2006).

No significant correlations between stroke velocity and accuracy, but there are correlations between shoulder strength and the velocity of tennis strokes.


  • Fatigue has been shown to have a detrimental effect on a player’s mechanics, thereby reducing ball velocity (performance), possibly in a protective mechanism to avoid injury by limiting the large ranges of motion and forces in a compromised biomechanical position. Fatigue has been shown to decrease proprioceptive ability, which may lead to protective mechanisms being too slow in response to prevent injury. Fatigue affects the sensation of joint movement, decreases athletic performance, and increases fatigue-related shoulder dysfunction. Fatigue has been shown to reduce shoulder external rotation, which has been suggested as the possible reason for the performance and force decrements found with extended tennis play.
  • Metabolic and physiological functioning is also reduced. The duration of recovery, as well as the duration of the intensity of work, is important for the regulation of physiological strain during intermittent exercise. Power decrements in the course of high-intensity intermittent exercise, as in tennis, have been related to a continuous degradation of phosphocreatine, thus placing greater demand on glycogenolysis and glycolysis, with increasing muscle and blood lactate concentrations resulting in large reductions in muscle pH.
  • The quality of movement patterns and coordination of specific actions in tennis is dependent on the physiological strain produced during short term intermittent exercise. Small changes in the recovery time can produce large changes in the performance of the exercise.
  • Ferrauti et al. in Kovacs, 2006, suggest that a decrease in running speed results in inaccurate stroke preparation, leading to a decrease in stroke speed (performance), as well as possible stroke intention (avoiding errors v hitting winners). It is very important when structuring drills on the practice court that the intention of the drill is understood. When working on technical issues, it is essential to give appropriate rest. It is imperative to use work/rest ratios that provide the coach and athlete with the right environment for ‘optimum’ outcomes. When working on technical skills, it is important to have greater rest than when working on tennis-specific movement or energy-system specific training.
  • Hitting accuracy is reduced by as much as 81% when a tennis player is nearing volitional fatigue. It has been reported that after a two-hour strenuous training session, an increase in ground stroke errors during defensive rallies and an increase in errors on first serves were observed. Fatigue from maximal tennis hitting has resulted in a 69% deterioration in hitting accuracy of ground strokes and a 30% decline in accuracy of the service to the right-hand court. After a fatiguing test, the serve was the most obvious tennis stroke to deteriorate in skill.
  • In addition to the serve, ground strokes place additional stress on the shoulder, though to a lesser degree. Our results showed that for service games topspin ground strokes were the second most frequently hit strokes, while for return games there were more topspin ground strokes and service returns than all other strokes. While muscle activity during the preparation phase of ground strokes is minimal, the acceleration and follow-through phases yield much higher activity. Electromyography during the forehand yields high activity in the subscapularis, biceps brachii, pectoralis major, and serratus anterior. The serratus anterior, subscapularis, infraspinatus, and biceps are also moderately active during the follow-through. With regard to the backhand, the middle deltoid, supraspinatus, and infraspinatus show a high degree of activity during acceleration. These muscles are also active during the follow-through, along with the biceps, though to a lesser degree (Fernandez-Fernandez et al., 2009).
  • The greater number of strokes on clay may contribute to earlier fatigue and possibly to a higher prevalence of injury, especially if players are forced to compete on consecutive days (Johnson et al., 2006).
  • As tennis competition has average points lasting less than 10 seconds, with rest periods of approximately 20 seconds between points and 90 seconds after every second game, the physiological variables are unlikely to lead to a large accumulation of lactate. Thus accumulating lactate levels are not a major cause of fatigue in tennis match play.

When working on technical skills, it is important to have greater rest than when working on tennis-specific movement or energy-system specific training. 

Differences between junior and professional players

The current population of junior competitors is twice the size of the professional tour and lacks the competitiveness and depth of the professional game. Thus, in the transition from the junior to professional level, players must become accustomed to an environment with a deeper and higher-quality group of athletes. These findings substantiate the perceived loss of competence and confidence that has been reported by transitioning junior players (Newman, 2009 in Kovalchik & Reid, 2007).

Some of the largest differences we found between juniors and professionals were in the observed physical demands of play. Professional players play with more power and accuracy than junior players, this is, at least in part, due to physical capacity maturation. The largest differences were observed in serve speed (men’s pro 20kph faster than boy’s, women’s 10 kph faster than girl’s serve) and serve placement (fewer deliveries hit near to lines at junior levels) which shows that serve is becoming a clear advantage at the professional level. Professional players produced nearly double the rally shots and serve shot production, total distance run and work performed when comparing a senior Grand Slam to a junior GS. Same was not observed for female players, most likely due to differences in match formats (best of 5 for male vs best of 3 for female). However, speed and changes of direction were actually higher for junior players (both boys and girls), showing that juniors have the potential to play at the same intensity as the professionals, where they lack is in the ability to keep those efforts at a high enough level throughout the course of a match and/or tournament. That is especially the case for transitioning male players. This study as long as some previous ones reaffirm the view that female players are capable of reaching their peak performance in tennis at earlier ages than male players (Kovalchik et al., 2017; Otis, et al., 2006 in Kovalchik & Reid, 2017).

While juniors have the potential to play at the same intensity as the professionals, where they lack is the ability to keep those efforts at a high enough level throughout the course of a match and/or tournament. That is especially the case for transitioning male players.

Key points

  • Junior players transitioning to the professional level must adapt to a field of deeper and higher-quality athletes.
  • Junior players rise in the professional ranks, they can expect to compete in more events, matches sets, and games throughout the year.
  • The margins differentiating winners and losers of matches at the professional level are significantly narrower than at the junior level.
  • Some of the largest differences between junior and professional tennis are in its physical demands.


  • Improving tennis performance is the goal of every tennis scientist, coach, and athlete. Age, sex, style of play, physical components, technical components, tactical components, and psychological components will all determine the success of the tennis athlete. Effective planning and training programs will help in designing a safe, effective, and productive program design to help optimize performance.
  • Most training drills should simulate the time requirements experienced during match play (5– 20 seconds) with appropriate work to rest ratios (1:3 to 1:5). As speed, change of direction and agility, as well as maximum velocity movements respond to specific and individualized training, it is important that tennis players focus on all of those aspects with drills combining linear, lateral, and multidirectional movements.
  • Having good aerobic capacity is important for recovery during play and between sessions. It is recommended that tennis athletes strive for VO2max values greater than 50 ml/ kg/min (42 ml/kg/min+ for female). Having adequate strength levels in all muscles and joints is important, but specific areas of focus should be the lower body for speed and quality of the movement and shoulder, forearm, and wrist for performance (stroke velocity and accuracy).
  • While tennis rehabilitation/prehabilitation programs do exist they are not based on objective data but rather on an expert’s knowledge of the sport and often aren’t modified according to a player’s skill level. Second, as serves, service returns, and topspin ground strokes are the pre-dominant strokes, coaches should emphasize proper mechanics and training of these stroke types.
  • Identifying the ways in which the junior and professional levels differ in competitiveness, play demands, and the physical characteristics of shot and movement can be useful for young tennis players and their coaches to set realistic expectations as they transition. Another possible application is to set benchmarks for juniors aiming for a professional career.

It appears that the all-court player is the player’s role that coaches would like to train to be competitive in tomorrow’s tennis.


Full articles available through links:

Fernandez-Fernandez, J., Sanz-Rivas, D., & Mendez-Villanueva, A. (2009). A review of the activity profile and physiological demands of tennis match play. Strength & Conditioning Journal, 31(4), 15-26.

Gomes, R. V., Coutts, A. J., Viveiros, L., & Aoki, M. S. (2011). Physiological demands of match-play in elite tennis: A case study. European Journal of Sport Science11(2), 105-109.

Kovacs, M. S. (2006). Applied physiology of tennis performance. British journal of sports medicine40(5), 381-386.

Kovalchik, S. A., & Reid, M. (2017). Comparing matchplay characteristics and physical demands of junior and professional tennis athletes in the era of big data. Journal of sports science & medicine, 16(4), 489.

Johnson, C. D., & McHugh, M. P. (2006). Performance demands of professional male tennis players. British journal of sports medicine40(8), 696-699.

Hoppe, M. W., Baumgart, C., Bornefeld, J., Sperlich, B., Freiwald, J., & Holmberg, H. C. (2014). Running activity profile of adolescent tennis players during match play. Pediatric exercise science, 26(3), 281-290.

Brechbuhl, C., Girard, O., Millet, G. P., & Schmitt, L. (2016). On the use of a test to exhaustion specific to tennis (TEST) with ball hitting by elite players. PloS one, 11(4), e0152389.

Baiget, E., Corbi, F., Fuentes, J. P., & Fernández-Fernández, J. (2016). The relationship between maximum isometric strength and ball velocity in the tennis serve. Journal of human kinetics, 53(1), 63-71.


Domen Bremec