Professional Documents
Culture Documents
GILLET. A Notational Analysis of Elite Tennis Serve and Serve-Return Strategies On Slow Surface
GILLET. A Notational Analysis of Elite Tennis Serve and Serve-Return Strategies On Slow Surface
T
he hand notation system (5) and computerized recently, strategies based on gender and court surface have
notational analysis (12) are methods for analyzing been investigated in relation to serve speed at Grand Slam
the dynamic and complex situations of sport tournaments. First, the serve is of greater importance in men’s
competition and training. They have been widely singles than in ladies’ singles. On the first serve, men players
applied to racket sports. Among the main functions of serve faster, getting a lower percentage of serves in but
notational analysis (13), tactical evaluation of play has winning a greater percentage of points when the serve is in
become a major area of interest to notational analysts. Match (8,26,27). Second, world tour tennis is characterized by play
on different court surfaces. The ones used at the 4 Grand
Address correspondence to Eric Gillet, eric.gillet@univ-rouen.fr. Slam tournaments can be classified by their coefficients of
23(2)/532–539 friction and restitution, which affect the ball bounce tra-
Journal of Strength and Conditioning Research jectory (2). According to the International Tennis Federation
Ó 2009 National Strength and Conditioning Association (17) court surface classification, surface speed falls into 3
the TM
categories: slow surfaces such as the French Open clay study has specifically examined the serve and serve-return
courts; medium surfaces such as the cement, acrylic, or strategies used by world-class players on slower surfaces
synthetic courts (U.S. Open and Australian Open), and fast using a holistic analysis that integrates the performance
surfaces such as the Wimbledon grass courts. Controversy indicators of location and spin. Therefore, the purpose of the
persists regarding serve dominance in relation to surface present study was to investigate the game patterns and
speed. Furlong (8) and Hughes and Clarke (15) found no strategies of serve and serve-return on a slow surface (clay) to
significant difference between fast and slow surfaces, whereas specify contextual strategic information. This was done to
O’Donoghue and Ballantyne (26) and O’Donoghue and determine whether the point issue is also strongly influenced
Ingram (27) have shown that serves are more effective on by the characteristics and quality of the first strokes produced
faster surfaces. Moreover, although serve speed has been by all opponents. A notational analysis was thus undertaken
found to be negatively correlated with the proportion of for 2 years (2005 and 2006) during the French Grand Slam
serves that are in, the proportion of points won when the tournament.
serve is in has been positively correlated with the serve speed
on both the first and second serves in all Grand Slam METHODS
tournaments (26). Experimental Approach to the Problem
Serve effectiveness depends directly on the opponent’s This study was a quantitative analysis of tennis based on
serve-return skills. Receiving players face high time con- positional play, shot details, stroke selection and execution,
straints (serves up to and greater than 200 kmh21) and winners, and errors. One hundred twenty-six men’s singles
variations of the serve trajectory (spin, direction, and ampli- matches from terrestrial television tournament coverage were
tude). Thus, they must react extremely quickly and are forced recorded and analyzed (Figure 1). The matches included in
to hit in uncomfortable positions because of the serve the current research were from the second round to the final.
speed. The serve-return has become 1 of 2 most important All matches in which fewer than 100 points were played were
shots, along with the serve, in the modern tennis game. excluded from the study to ensure that each match selected
However, few studies have analyzed its tactical character- was representative of Grand Slam tennis (27). This reduced
istics. On average, 15% of all strokes are serve-returns in the number of singles matches analyzed from 126 to 116.
a best-of-3 set tennis match on clay courts against 20% A computerized notational system was developed to collect
on faster courts (30). The serve is more difficult to return data from video recordings of the selected matches. This
on faster courts (3,15). However, the number of points won system was used to produce a spreadsheet for statistical
when returning is higher on fast surfaces (about 40%) than on analysis. It recorded general information (players, score, etc.)
clay courts (about 30%) (30). and performance indicators for each point about serve and
Thus, the beginning of each point in a modern tennis game serve-return strokes: the type (first or second serve; forehand
seems particularly essential and even decisive for increasing or backhand), spin (flat, topspin, or slice), and location. To
the chances of winning the
point. Nevertheless, perfor-
mance indicators other than
serve speed and the type of
surface need to be investigated
to better understand the strate-
gies developed for the serve and
return. Performance indicators
such as the location of the serve
(4) and the return, as well as
their spins, contribute to the
proportion of points won di-
rectly or indirectly from this
serve-return sequence. To our
knowledge, only the study of
Unierzyski and Wieczorek (32),
which describes the game pat-
terns used by the world’s best
players during the final matches Figure 1. Schematic representation of the tennis court for data collection and analysis. Each service box was
at the French Open and Wim- divided into 3 zones: wide, into the receiver’s body, and ‘‘T’’ near the center service line. Concerning return location,
each half of the tennis court was divided into 16 equal rectangles. Eight rectangles located in the middle of the court
bledon events, has taken the formed the central zone; 8 other rectangles located close to the singles sidelines formed the external zone.
placement of winning serves S = server; R = receiver.
and returns into account. No
Figure 3. Location of first and second serves in relation to the serve spins used and the half of the court where the serve stroke was executed. Significantly more
flat first serves down the ‘‘T’’ near the center serve line (*). Significantly more topspin (+) and slice ($) first serves into wide location. In the deuce court, significantly
more flat (*) and topspin + first serves and topspin second serves (+) down the center ‘‘T’’ and slice first serves ($) into the wide location. In the advantage court,
significantly more topspin (+) and slice ($) first serves and topspin (+) second serves into the wide location.
won significantly more points when they sent the ball into p , 0.001). Receivers performed more winners after the
the central zone than into the external zone (Figure 6). second serve (54.7%) than the first serve (45.3%). Nonethe-
Moreover, as presented in Figure 6, this association less, they performed more errors after the second serve
depended on the type of serve (x21 = 26.7, p , 0.001). (78.6%) than the first serve (21.4%).
proportion of backhand serve-returns was around 60%). A has run out (1). This may imply, in return, a reduction in
high bounce on the serve, together with a short angle, helps ‘‘actual available time’’ (20) assigned to servers, disrupting
to gain initiative and is followed with a pressure stroke. Then, their motion coordination at the time of the split step (i.e., the
servers play their second shot inside the baseline, which can first step after landing). Servers indeed seem to have greater
be hit for a winner or as a preparatory point-winning stroke. difficulty in turning around the ball than in moving toward
Serving players can use their best shot, such as an inside-out the ball in response to an external serve-return. Last, return-
forehand, to the weak side of the receiver. Finally, the topspin ing to the central zone also reduces the tactical response
first serve strategy enables players to obtain high serve possibilities of the server’s second shot. This game situation is
efficiency during a complete match by maintaining a high less comfortable for opening up the court or winning the
percentage of points won from the serve (52.4%). This point by this second shot. This interpretation seems to be
topspin strategy is classically used on the second serve corroborated by the present results, which show that the
(91.6%) to limit aggressive and offensive serve-returns. percentage of serve-returns played in the external zone was
The results show that a high proportion of serve-returns significantly higher after second serves (28.0%) than first
were in (80%). This percentage reflects the specificities of the serves (21.2%), indicating that this location allowed receivers
clay game in relation to the characteristics of the surface. The to win more points after second serves (34.1%). The external
height and relative slowness of the bounce (2), associated serve-returns are thus performed either with the intention of
with the receivers’ ready position on the court far from the winning the point directly (i.e., put-away shots) or dominat-
baseline, increases the available time budget for hitting the ing the rally with the receiver’s second shot. According to
ball. However, servers always dominate over receivers. Schonborn (30), 6% of all points on clay courts end with this
Moreover, our results show that the bounce of serve-return fourth shot. Last, this serve-return strategy may be subject to
strokes was significantly placed into the central zone (75.5%). the specificity of the second serve on a clay surface.
This location was the most beneficial for winning the point, Indeed, more than 90% of the second serves were
as much after a first serve (73.3%) as after a second serve performed with topspins toward 2 main locations: down
(65.9%). These data seem to contradict the findings of the ‘‘T’’ in the deuce court (48.0%) and to the wide zone in the
Unierzyski and Wieczorek (32), who have noted that advantage court (46.0%). Knowledge of this game situation
returning into the external zone gave only a slightly better (current and past) and use of this knowledge during the match
chance of winning a rally than returning to the central zone. may influence receivers’ performances by enabling them to
Nevertheless, this discrepancy in the data can be explained by develop tactical solutions for the second serve and thus
the difference in the location of the central zone in the facilitating the decision-making process. Declarative and
aforementioned and current studies (i.e., the middle of the procedural knowledge, sport-specific memory adaptations
court and from the net to the baseline, respectively). Given and structures, and domain-related strategies probably allow
the high speed of the flat first serve and the high bounce receivers to select the best response and carry it out (21, 23).
resulting from the topspin second serve, the strategy to return The knowledge structure developed over many hours of
into the central zone allows receivers to minimize the error ‘‘deliberate practice’’ (7) and competition on clay may enable
number of the stroke direction while increasing their chances receivers to accurately anticipate upcoming second serves
of winning the point. Receivers are then able to find an while maintaining other responses on alert for possible
individual compromise between the velocity and the selection once the game situation demands it. This was also
precision of their racket movement (19). When receiving confirmed by the percentage of serve-return winners (54.7%)
rapid first serves, this issue may be expressed by the execution and errors (78.6%) after second serves in comparison with
of counterstrokes with a short backswing reduced to a simple first serves, indicating greater risk taking. Receivers can thus
hip and shoulder rotation (i.e., blocked return). Players return perform an offensive serve-return by running around their
with low racket and segment velocities while maintaining backhand and hitting an inside-out topspin or flat forehand
good coordination and synchronization within the kinematic to the external zone or to the weakest side of the server. As
chain from the proximal to distal segments (19). This the action unfolds, situational probabilities integrated with
backswing shortening has the advantage of reducing the game contextual knowledge would provide a confirmation or
‘‘movement time’’ of receivers and, consequently, decreases modification of the anticipated response (22,29). The player’s
the time constraints to which they have been subjected. knowledge of the serve proportion (i.e., general situational
Faced with topspin second serves, receiving players may probabilities) and the specific server’s tendencies (i.e., specific
reach maximum racket velocity at impact. The reduction in situational probabilities) result in a decrease in situational
serve speed then enables receivers to execute offensive uncertainty, reaction time, and anticipatory movement time.
strokes with a backswing similar to that used for regular This implies an increase in the time required to perform an
ground strokes. Thus, the compromise between velocity and offensive return whose longer movement time is determined
precision is shifted to the velocity side (19). Both these by the backswing amplitude.
strategies may be used to increase the ‘‘time required’’ to This notational analysis has allowed us to confirm and
build an effective striking action before the ‘‘available time’’ extend our knowledge of clay tennis games by highlighting
the indices of performance of the first strokes of each player. 5. Downey, JC. The use of notational analysis in determining optimal
Even on the clay court surface, serves and serve-returns are strategies in sport. In: Proceedings of the Notational Analysis of Sport I
and II. Hughes, M, ed. Cardiff: UWIC, 1992. pp. 3–18.
decisive for the positive or negative outcome of the rally and,
6. Durey, A. Physique Pour les Sciences du Sport. Paris: Masson, 1997.
consequently, for winning matches. Knowledge of the
7. Ericsson, KA, Krampe, RT, and Tesch-Römer, C. The role of
performance indicators of this game situation should provide deliberative practice in the acquisition of expert performance. Psychol
players and coaches with objective information so that they Rev 100: 363–406, 1993.
can improve their acquisition of the specific skills needed for 8. Furlong, JDG. The serve in lawn tennis: how important is it? In:
the clay game. Finally, the above results could be made even Proceedings of the Science and Racket Sports. Reilly, T, Hughes, M, and
Lees, A, eds. London: E & Fn Spon, 1995. pp. 266–271.
more meaningful by incorporating biomechanical indicators
such as hitting speeds and segment velocities. 9. Gale, D. Optimal strategy for serving in tennis. Math Mag 44:
197–199, 1971.
PRACTICAL APPLICATIONS 10. Georges, SL. Optimal strategy in tennis: a simple probabilistic
model. J R Stat Soc Ser C Appl Stat 22: 97–104, 1973.
Modern clay court players should use, or at least try to play 11. Howell, DC. Fundamental Statistics for the Behavioral Sciences (5th ed.).
with, an attacking game based around their serve and serve- Brussels: De Boeck University, 2002.
return strokes to force their opponents to start the rally 12. Hughes, M. Computerised notation of rackets sports. In: Proceedings
aggressively and thus facilitate the setting up of their powerful of the Science and Racket Sports. Reilly, T, Hughes, M, and Lees, A, eds.
London: E & Fn Spon, 1995. pp. 249–256.
groundstroke game. Clay court tactics combining powerful
serves on the ‘‘T’’ and topspin serves into wide location should 13. Hughes, M. The application of notational analysis to racket sports.
In: Proceedings of the Science and Racket Sports II. Lees, A, Maynard,
place opponents under more time pressure and allow players IW, Hughes, M, and Reilly, T, eds. London: E & Fn Spon, 1998. pp.
to better open up the court with sharp angles. Receiving 211–220.
strokes should be played as aggressively as possible with 14. Hughes, M and Bartlett, R. The use of performance indicators in
a combination of optimal power and control. Only aggressive performance analysis. J Sports Sci 20: 735–737, 2002.
serve-returns can save players from being placed on the 15. Hughes, M and Clarke, S. Surface effect on elite tennis strategy.
In: Proceedings of the Science and Racket Sports. Reilly, T, Hughes, M,
defensive from the outset. This forcing return should be and Lees, A, eds. London: E & Fn Spon, 1995.
a deep ball into the central zone against a power first serve and pp. 272–277.
a deep or cross-court into the external zone against a weak 16. Hughes, M and Franks, IM. Notational analysis—a review of
first serve or a second serve. These serve-return strategies literature. In: Notational Analysis in Sport (2nd ed.). Hughes M and
should allow the player to recover his or her position and keep Franks, IM, eds. London: Routledge, 2004. pp. 59–106.
the server behind the baseline. Although players hit 17. International Tennis Federation. Approved Tennis Balls & Classified
Court Surfaces. London: ITF Licensing Ltd, 2007.
considerably more forehands and backhands than other
18. King, HA and Baker, JAW. Statistical analysis of serve and
strokes during a match on clay surface, the time spent match play strategies in tennis. Can J Appl Sport Sci 4: 298–301,
practicing the serve and serve-return in terms of volume and 1979.
quality are areas that players and coaches need to address. 19. Kleinöder, H and Mester, J. Strategies for the return of 1st and
Surprisingly, knowledge of these game patterns may allow the 2nd serves. In: Proceedings of the Tennis Science and Technology.
use of other unpredictable and effective strategies. Haake SJ and Coe, AO, eds. Oxford: Blackwell Science, 2000.
pp. 401–408.
20. Leveau, C. Notion de pression temporelle dans la relation duelle: un
ACKNOWLEDGMENTS exemple en badminton. In: Sports de Raquettes: Entre Pratiques et
The authors would like to thank the French Tennis Federation The´ories, Dossier Eps n°53. Louis, E. ed. Paris: Revue Eps, 2000.
pp. 77–88.
and Roland Garros Institution for providing funding for this
21. McPherson, SL. The development of sport expertise: mapping the
research program. Special thanks are extended to Patrice tactical domain. Quest 46: 223–240, 1994.
Hagelauer, the French Davis Cup coach, Jean-Claude Massias 22. McPherson, SL. Expert-novice differences in performance skills and
and Marc Renoult (National Technical Direction), and problem representations of youth and adults during tennis
Alexandre Loth (Innovation Technology Director) for their competition. Res Q Exerc Sport 70: 233–251, 2000.
support and cooperation. The authors gratefully acknowl- 23. McPherson, SL and Kernodle, MW. Tactics, the neglected attribute
edge Cathy Carmeni for linguistic revision of the manuscript. of expertise: problem representations and performance skills in
tennis. In: Expert Performance in Sports: Advances in Research on Sport
Expertise. Starkes JL and Ericsson, KA, eds. Champaign: Human
REFERENCES Kinetics, 2003. pp. 137–167.
1. Alain, C and Proteau, L. Etudes des variables relatives au traitement 24. Norman, JM. Dynamic programming in tennis—when to use a fast
de l’information en sports de raquette. Can J Appl Sport Sci 3: 27–33, serve. J Oper Res Soc 36: 75–77, 1985.
1978.
25. O’Donoghue, PG. The advantage of playing less sets than the
2. Brody, H, Cross, R, and Lindsey, C. The Physics and Technology of opponent in the previous two rounds of Grand Slam tennis
Tennis. Solana Beach, CA: Racquet Tech Publishing, 2002. tournaments. In: Proceedings of the Science and Racket Sports III. Lees,
3. Collinson, L and Hughes, M. Surface effect on strategy of elite A, Khan, JF, and Maynard, IW, eds. London: Routledge, 2003. pp.
female tennis players. J Sports Sci 21: 266–267, 2002. 175–178.
4. Croucher, JS. Developing strategies in tennis. In: Stat Sport. 26. O’Donoghue, PG and Ballantyne, A. The impact of speed of serve in
J. Bennett, ed. London: Arnold, 1998. pp. 157–171. Grand Slam singles tennis. In: Proceedings of the Science and Racket
the TM
Sports III. Lees, A. Khan, J.F. and Maynard, I.W. eds. London: 30. Schonborn, R. The return of serve. In: Proceedings of the 11th ITF
Routledge, 2003. pp. 179–184. Worldwide Coaches Workshop: Top Tennis Coaching. London: ITF
27. O’Donoghue, PG and Ingram, B. A notational analysis of elite tennis Licensing Ltd, 1999. pp. 38–43.
strategy. J Sports Sci 19: 107–115, 2001. 31. Underwood, G and McHeath, J. Video analysis of the energy system
in elite and competitive singles tennis. J Hum Mov Stud 28: 73–86,
28. O’Donoghue, PG and Liddle, SD. A match analysis of elite tennis
1977.
strategy for ladies’ singles on clay and grass surface. In: Proceedings
32. Unierzyski, P and Wieczorek, A. Comparison of tactical solutions
of the Science and Racket Sports II. Lees, A, Maynard, IW, Hughes, M,
and game patterns in the finals of two grand slam tournaments in
and Reilly, T, eds. London: E & Fn Spon, 1998. pp. 247–253. tennis. In: Proceedings of the Science and Racket Sports III. Lees, A,
29. Paull, G and Glencross, DJ. Expert perception and decision making Khan, JF, and Maynard, IW, eds. London: Routledge, 2003.
in baseball. Int J Sport Psychol 28: 35–56, 1997. pp. 169–174.