Grunting in Tennis Increases Ball Velocity but Not Oxygen Cost : The Journal of Strength & Conditioning Research

Secondary Logo

Journal Logo

Original Research

Grunting in Tennis Increases Ball Velocity but Not Oxygen Cost

Callison, Emily R.; Berg, Kris E.; Slivka, Dusting R.

Author Information
Journal of Strength and Conditioning Research 28(7):p 1915-1919, July 2014. | DOI: 10.1519/JSC.0000000000000333
  • Free



Elite tennis players namely Maria Sharapova,and Venus and Serena Williams and many others are known to grunt as they strike the ball. Players are believed to do this partly as a means of hitting the ball harder. Grunting is also common in weightlifting and martial arts and makes use of the increased stabilization associated with the Valsalva maneuver while lifting or delivering a blow. Force production in tennis and other sports typically involves the transfer of ground reaction forces through the lower extremities and trunk to the upper extremities and finally the hand (14). Grunting may enhance trunk stabilization thereby increasing the transfer of force from the trunk to the limb. Kibler et al. (12) state the importance of core stability for efficient biomechanical function to maximize force generation and to minimize joint loads in all types of activities ranging from running to throwing. However, results of studies examining the effects of breathing on trunk stability and force production of other muscles have been mixed. For example, Hagins et al. (9) reported no influence of various patterns of breath control including a Valsalva maneuver on isometric force of the trunk. In contrast, other studies observed spinal stiffness to peak during maximal expiration (19) or muscle force to be the highest with forced exhalation and Valsalva maneuver (11). Surprisingly, no research to our knowledge has been reported verifying that grunting actually increases ball velocity in tennis. This is a practical question needing to be addressed.

The mean intensity of a singles tennis match is about 50–80% of V[Combining Dot Above]O2max, which corresponds with V[Combining Dot Above]O2 levels ranging from 23 to 40 ml·kg·min−1 (8). These values would classify tennis as being aerobically demanding (14). Matches last from 1–5 hours with most work periods lasting about 5–10 seconds alternated with periods of lower intensity between 20 and 120 seconds (6). Additionally, people of different abilities have shown little difference in V[Combining Dot Above]O2 while playing, 24.5 ml·kg·min−1 for advanced players vs. 23.3 ml·kg·min−1 for recreational players (8). Given these characteristics, the oxygen cost of playing is important. Grunting while playing may increase the oxygen cost. Grunting when hitting each groundstroke may increase muscle activation of accessory breathing muscles and trunk-stabilizing muscles which may in turn increase the oxygen cost of playing. Furthermore, breathing efficiency may be impaired by increasing the volume of air ventilated during a forced expiration. Thus, any gain in velocity of groundstrokes may be associated with increased energy expenditure, but this too has not been verified.

The modern game of tennis is characterized by speed and power, and grunting may facilitate these qualities, but it may also increase the oxygen cost of playing. It is surprising that the possible impact of grunting on stroke velocity and energy cost has not been previously studied. Therefore, the purpose of this study was to determine if grunting increases the velocity of groundstrokes and physiological responses while hitting as indicated by V[Combining Dot Above]O2, HR, VE/V[Combining Dot Above]O2, and RPE. We hypothesized that grunting would increase ball velocity but also increase the associated oxygen cost of playing. These variables were assessed over time to determine possible change that may occur but not be apparent over a short bout of hitting.


Experimental Approach to the Problem

Five male and 5 female members of an National Collegiate Athletic Association (NCAA) Division I tennis team participated in 2 hitting sessions that included wearing a portable metabolic unit while returning forehand and backhand groundstrokes from a ball machine for a total of 10 minutes (five 2-minute periods with a 1-minute rest between periods). Physiological data consisting of V[Combining Dot Above]O2, heart rate, VE/V[Combining Dot Above]O2, and RPE were collected during each hitting session. Concurrently, a radar gun was used to measure ball velocity of each stroke. This design permitted testing the hypotheses dealing with ball velocity and oxygen cost of playing. Values for the grunt vs. non-grunting condition were compared, as were the means of each 2-minute time period in each session to determine the potential effects of fatigue on the data over 5 periods of hitting.


Subjects were 10 NCAA Division I male and female tennis players (5 men and 5 women) whose mean ± SD age, height, and mass were 21.5 ± 0.97 years, 173.5 ± 11.0 cm, and 71.8 ± 14.7 kg, respectively. All were experienced players having competed at lower levels of play while in high school and earlier. Subjects were also injury-free at the time of the study. Subjects provided informed consent using the university institutional review board approved form after a thorough explanation of the procedure. Subjects were treated according to the Helsinki Declaration on the rights of human subjects.


Testing occurred after completion of the indoor competitive season, so the athletes were in a trained state. Subjects were instructed to report for testing as they typically would for a practice or match, i.e., hydrated, not having eaten a meal in the previous 2–3 hours. Testing was done at the regular time of practice, which was midafternoon.

Hitting Sessions

Subjects participated in 2 hitting sessions that involved grunting while striking the ball and not grunting while striking the ball. The sequence of the grunt and non-grunting conditions was randomly assigned. Each hitting session consisted of hitting alternating forehand and backhands for five 2-minute periods with a 1-minute break in between each period. A ball machine was used to feed balls to each player, so they hit a ball that was traveling at the same speed, height, and landed in nearly the same spot on the court. This helped to ensure that each player hit the same type of ball, and that balls changed pace or height only minimally during each individual hitting session. Ball velocity was measured with a hand held radar gun held at the net at the edge of the court. The velocity of each hit was recorded unless an errant hit occurred, e.g., ball hit racquet edge. In the grunt condition, subjects were asked to grunt as naturally as possible and so that the sound was audible.

Data Collection

To test our hypotheses, we measured ball velocity of every stroke made as explained in the previous paragraph. The hypotheses dealing with the comparison of physiological responses while grunting or not grunting required measurement of V[Combining Dot Above]O2, HR, VE, and RPE. Each of these measures is explained below.

During each hitting session, a player wore a Medgraphics VO2000 portable metabolic measuring system (RunMan, St. Paul, MN, USA) that measured V[Combining Dot Above]O2 and VE/V[Combining Dot Above]O2. The VO2000 was fitted on the player's body using a standardized procedure determined by investigators in pilot testing and in previous studies. Players were familiarized with wearing the portable system while hitting by having them wear the apparatus during an earlier practice session. V[Combining Dot Above]O2 was measured by the system with a sampling frequency of 0.05 Hz, whereas HR was monitored using a Polar monitor, and RPE was assessed using Borg's original 6–20 scale (Borg, 1982) during each 1-minute rest period between periods. Previous studies confirmed the validity and reliability of the portable metabolic measurement system (Byard and Dengal, 2002). Ball velocity was measured using a radar gun (Pocket Radar, Inc., model 1000, Santa Rosa, CA, USA). The trigger on the radar gun was held during the execution of the stroke; and after the trigger was released, the peak velocity was displayed and recorded. The radar gun emits electromagnetic radiation waves at a specific frequency, which are reflected from the ball at an altered frequency. The shift in frequency is then used to calculate ball velocity (10).

Pilot testing with several players who were not subjects in the study was conducted to make adjustments in the procedures, e.g., location of the radar gun and ball machine. Pilot testing continued until ball velocities over several subjects was consistently within 5 mph of the mean.

Statistical Analyses

Mean and SD were calculated for all variables over each 2-minute hitting period. A series of 2 (grunting or non-grunting condition) × 5 (time period) repeated measures factorial analysis of variance was used to compare each dependent variable across the 5 time periods. The Fisher's protected least significant difference test was used to identify where significance occurs where ratios were found to be significant. Statistical assumptions for normality, linearity, and equality of variance were met. An alpha level of p ≤ 0.05 was used for all tests. Effect size for dependent variables using the method advised by Cohen (4) was used to interpret the magnitude of significant differences. SPSS 19.0 was used for all analyses.

Based on the effect sizes of V[Combining Dot Above]O2 and other physiological data in other studies on tennis players using samples of 8 and 10 athletes (2 and 7, respectively) and in other sports (1,3,17), a sample size of 10 was judged to have a statistical power of about 0.70. Adequate statistical power is supported by the fact that this study was able to detect a number of significant differences.


The main effect for ball velocity was significant favoring the grunting over the non-grunting condition (p = 0.034). The mean ± SD for ball velocity (in kilometer per hour) for the grunting and non-grunting conditions were 83.4 ± 0.6.1 and 80.3 ± 8.7, respectively. The difference represented an effect size of 0.409, which is considered small to moderate. The main effect for change over time was not significant indicating that ball velocity did not significantly change over the 5 time periods. Figure 1 displays ball velocity in the 2 conditions and time periods.

Figure 1:
Summary of ball velocity while grunting or not grunting over five 2-minute periods. *A significant main effect for ball velocity (p ≤ 0.05) occurred favoring the grunt condition, mean ± SD.

The grand mean over the five 2-minute periods of hitting revealed that grunting did not significantly increase V[Combining Dot Above]O2, HR, VE/V[Combining Dot Above]O2, or RPE compared with the non-grunting condition. V[Combining Dot Above]O2 and HR (mean ± SD) for the grunting and non-grunting conditions were 31.1 ± 4.7 and 31.7 ± 4.7 ml·kg·min−1 and 164 ± 14.5 and 159 ± 11.3 b·min−1, respectively. However, a trend toward significance existed with VE/V[Combining Dot Above]O2 higher while grunting in comparison with the non-grunting condition (26.1 ± 3.54 and 25.3 ± 3.15, respectively; p = 0.067). Mean ± SD for RPE in the grunting condition was 14.5 ± 2.02, and for the non-grunting condition, it was 14.0 ± 2.05.

We hypothesized that physiological responses would rise over the 5 time periods because of fatigue. However, V[Combining Dot Above]O2 did not significantly increase while HR did with periods 3, 4, and 5 being significantly higher than period 1 (p = 0.018). VE/V[Combining Dot Above]O2 and RPE were both higher in time periods 2–5 than the first (p = 0.001). The only significant interaction in our findings was that RPE was higher while grunting than non-grunting in the fifth time period (p = 0.49). Physiological responses for the 5 time periods are summarized in Table 1.

Table 1:
Summary of V[Combining Dot Above]O2, VE/V[Combining Dot Above]O2 and heart rate.


The main findings of this study were that grunting increased ball velocity without raising V[Combining Dot Above]O2. This finding provides an evidence base for application to sport. Grunting is widely used in sport, and this study suggests that the common use of forced expiration or grunting in sport may in fact enhance force production and velocity of striking a ball. The proposed mechanism involved may be that grunting increases trunk stability thereby increasing force generation and transfer of energy from the large to small body parts during sport activities (12). Grunting appears to increase intra-abdominal pressure during the respiratory cycle, thereby increasing stiffness of the lumbar vertebrae providing greater trunk stability and increasing force development (15). Supporting the results of our study is an investigation by Ikeda et al. (11) who found that when subjects used forced exhalation during isometric contractions, maximum force increased more than during normal breathing. Results of other studies, however, have produced mixed results with some observing no difference in trunk force during various patterns of breath control including a Valsalva maneuver (9), whereas others have found production to be enhanced with forced exhalation/Valsalva maneuver (19). However, none of these studies used a sport task such as a tennis groundstroke but rather measured only force production. Therefore, no direct comparisons can be made with other studies and, in particular, tennis, since no published research exists to our knowledge. This is surprising given that grunting is common in tennis especially among professional players but in other sports as well such as weightlifting and martial arts.

Trunk stabilization may be critical in dynamic sports activities such as throwing and tennis because of the high velocities achieved by the limbs. For example, the velocity of elbow extension has been measured at 1100–1700° per second and forearm pronation at 900–110° per second 0.1 seconds after impact in professional players (16). The magnitude of the stabilizing forces required in such sports tasks would seemingly be very high. Thus, the commonality of grunting in many sports may reflect the high demand on trunk stabilization required to optimize performance in terms of velocity and accuracy.

Ball velocity increased 3.8% while grunting which produced an effect size of 0.409. This effect size according to Cohen (4) is between small and moderate. An increase of this magnitude may affect tennis play because it gives an opponent less time to prepare for a return shot. The 3.8% improvement in ball velocity is associated with the ball crossing the court 0.05 seconds sooner than without grunting. Having less time to set up for a return shot tends to rush an opponent thereby altering their technique and producing a less accurate and forceful return. The sport of tennis is faster paced than 10–20 years ago, and modern players generally emphasize hitting high velocity shots. Our results show that grunting may be a means of increasing ball velocity. Our players were NCAA Division I level, so it is interesting that with minimal time to practice grunting while hitting groundstrokes, these athletes were able to produce greater velocity. It would be interesting to determine if even greater enhancement might be possible with a longer period of training using the grunt technique. Similarly, it would be interesting to determine if lesser skilled players might yield even greater results.

It is important to note that the increase in ball velocity occurred without an increase in V[Combining Dot Above]O2 and V[Combining Dot Above]O2/VE. However, RPE increased over time and was greater in the fifth period of the grunting condition than the non-grunting condition. Further study over longer periods is needed to determine if an increase in ball velocity can be maintained without increasing V[Combining Dot Above]O2. Our data were limited to five 2-minute periods while match play often lasts 2–3 hours or even longer in professional men's events. The accuracy of shots may also be affected in the later stages of a match if RPE and associated V[Combining Dot Above]O2 levels continued rising. The rise in RPE over time and significantly greater RPE in the fifth period suggest that V[Combining Dot Above]O2 may increase over duration of a match while grunting because RPE and V[Combining Dot Above]O2 are closely associated (5). However, it is interesting that V[Combining Dot Above]O2 did not increase using 2-minute time periods which is far longer than rallies typically last in games. Tennis is played with bursts of some 300–500 periods of high intensity lasting about 5–10 seconds alternated with periods of low intensity lasting about 10–20 seconds (6).

V[Combining Dot Above]O2 and VE/V[Combining Dot Above]O2 did not increase with the grunting trials. We hypothesized that grunting would increase V[Combining Dot Above]O2 because of the forced exhalation which we believed that would increase VE and because of increased activation of the trunk stabilizers. However, results showed this was not the case and even over time no significant differences were found in V[Combining Dot Above]O2 and VE/V[Combining Dot Above]O2 between the 2 conditions. V[Combining Dot Above]O2 may not have increased because the hitting session lasted only 10 minutes of total hitting time. It would be interesting to examine V[Combining Dot Above]O2 in an actual match or in longer hitting sessions.

RPE significantly increased over time across the 5 time periods for both conditions, and the RPE in period 5 for the grunting condition was significantly higher than the RPE in period 5 for the non-grunting condition, as illustrated in Figure 2. These results may reflect the perception of exerting more energy when grunting resulting from increased recruitment of accessory respiratory and trunk-stabilizing muscles. HR also significantly increased over time with periods 3, 4, and 5 being greater than period 1. The increase in HR may be due to cardiovascular drift occurring from sweat loss and a shift of plasma volume into surrounding extracellular fluids (18). Additionally, as with RPE, as time increased, a player would likely recruit more motor units as fatigue mounted thus increasing the HR response (6).

Figure 2:
Summary of RPE in grunt and no grunt conditions over 5 time periods; mean ± SD. *There was a significant interaction with grunt RPE exceeding no grunt in the fifth period (p ≤ 0.05 from grunt).

Several limitations exist in this study. Only 2 of our players typically grunted while hitting and for the others, the test protocol was somewhat unique. However, we provided a familiarization period with grunting in a practice session conducted before the day of data collection. Similarly, players who normally did grunt were asked not to grunt in the practice session to become familiarized with this variation of normal play. Seemingly the familiarization session was adequate to produce greater ball velocity without increasing V[Combining Dot Above]O2 or VE/V[Combining Dot Above]O2. This improvement was observed in all but one of the players. However, possible psychological factors may have influenced the data. The duration of the 2 hitting conditions was only 15 minutes, which consisted of 5 bouts of 2 minutes of hitting alternated with 1-minute rest period. Most recreational tennis matches last on average 1.5 hours with points lasting 3–15 seconds with a 1:2 work to rest ratio (13). Therefore, the protocol used here may not give an accurate representation of actual match play. It may be that with longer play, V[Combining Dot Above]O2 would increase while grunting.

The following conclusions are warranted from the study while hitting tennis groundstrokes: (a) Grunting increases ball velocity without increasing V[Combining Dot Above]O2 or VE/O2; (b) Whether grunting or not grunting, V[Combining Dot Above]O2 does not increase with duration of hitting, whereas VE/V[Combining Dot Above]O2 and RPE do increase. Ball velocity does not change over five 2-minute periods of hitting whether grunting or not. Generalization of our results should be made cautiously given that we studied only 10 NCAA Division I athletes. Further study of players of various ages, performance levels, and during actual match play is needed.

Practical Applications

The results of this study provide an evidence base for using grunting as a means of enhancing sport performance. Our study shows benefits for hitting groundstrokes in tennis with greater velocity It may be worthwhile for players and coaches in tennis and other sports to experiment with grunting to determine possible improvement in performance. In tennis, an increased ball velocity of 3.8% should benefit players during match play because one's opponent has less time to set up and hit accurate strokes. Being rushed to do so usually results in less effective return shots. Improvement of ball velocity of the 10 players in this study is equivalent to reducing the time to react to a ball by 0.05 seconds. Another possible advantage is that grunting may also increase a player's serve velocity. These potential benefits may be worthy of further study and experimentation by individual athletes in a variety of sports.


1. Berg K, Narazaki K, Latin R, Vincent W, Meisinger M, Sjoberg C, Kaufman C. Oxygen cost of and energy expenditure of racquetball. J Sports Med Phys Fitness 47: 395–400, 2007.
2. Botton F, Hautier C, Eclache JP. Energy expenditure during tennis play: A preliminary video analysis and metabolic model approach. J Strength Cond Res 25: 3022–3028, 2011.
3. Chaffin M, Berg K, Zuniga J, Hanumanthu S. Pacing pattern in a 30 minute maximal cycling test. J Strength Cond Res 22: 2011–2017, 2008.
4. Cohen J. Statistical Power Analysis for the Behavioral Sciences (3rd ed.). Hillsdale, NJ: Erlbaum, 1988.
5. Eston RG, Williams JG. Reliability of ratings of perceived effort regulation of exercise intensity. Br J Sports Med 22: 153–155, 1988.
6. Fernandez J, Mendez-Villanueva A, Pluim BM. Intensity of tennis match play. Br J Sports Med 40: 387–391, 2006.
7. Fernandez-Fernandez J, Kinner V, Ferrauti A. The physiological demands of hitting and running on different surfaces. J Strength Cond Res 24: 3255–3264, 2010.
8. Fernandez-Fernandez J, Sanz-Rivas D, Sanchez-Munoz C, Pluim BM, Tiemessen J, Mendez-Villanueva A. A comparison of the activity profile and physiological demands between advanced and recreational veteran tennis players. J Strength Cond Res 23: 604–610, 2009.
9. Hagins M, Pietrek M, Sheikhzadeh A, Nordin M. The effects of breath control on maximum force and IAP during a maximum isometric lifting task. Clin Biomech (Bristol, Avon) 21: 775–780, 2006.
10. Hewitt P. Conceptual Physics: A New Introduction to Your Environment (2nd ed.). Boston, MA: Little, Brown, and Company, Inc., 1974. pp. 365–367.
11. Ikeda ER, Borg A, Brown D, Malouf J, Showers KM, Li S. The Valsalva maneuver revisited: The influence of voluntary breathing on isometric muscle strength. J Strength Cond Res 23: 127–132, 2009.
12. Kibler WB, Press J, Sciascia A. The role of core stability in athletic function. Sports Med 36: 189–198, 2006.
13. Kovacs MS. Applied physiology of tennis performance. Br J Sports Med 40: 381–386, 2006.
14. Kovacs MS. Tennis physiologly training the competitive athlete. Sports Med 37: 189–198, 2007.
15. Li S, Laskin JJ. Influences of ventilation on maximal isometric force of the finger flexors. Muscle Nerve 34: 651–655, 2006.
16. Mayle R, Ellenbecker T, Safran M. Tennis. In: Netter's Sports Medicine. Philadelphia, PA: Saunders Elsevier, 2010.
17. Narazaki K, Berg K, Stergiou N, Chen B. Physiological demands of competitive basketball. Scand J Med Sci Sports 19: 425–432, 2009.
18. Powers SK, Howley ET. Exercise Physiology: Theory and Application to Fitness and Performance (7th ed.). New York, NY: McGraw-Hill, 2009.
19. Shirley D, Hodges PW, Eriksson AE, Gandevia SC. Spinal stiffness changes throughout the respiratory cycle. J Appl Phsiol (1985) 95: 1467–1475, 2003.

breathing; Valsalva maneuver; force; groundstrokes; energy cost

Copyright © 2014 by the National Strength & Conditioning Association.