Technical–tactical behaviour and rally structure in professional men's doubles pickleball: A combined descriptive,predictive,and sequential analysis

Abstract

Despite the recent expansion of pickleball, empirical evidence on the technical–tactical demands of men's doubles play remains scarce. This study provides the first integrated analysis of match structure and point-resolution mechanisms in elite men's doubles pickleball. A total of 2344 points from fifteen semifinal and final matches of the 2023 PPA Tour were analysed using observational methodology. All actions were coded with LINCE PLUS, and descriptive and inferential analyses were conducted in SPSS. Variables included rally length, striking zone, finishing zone, and final stroke. A CHAID decision-tree model was used to identify multivariate predictors of point outcome, and THEME 6 Edu was applied to detect recurrent T-patterns. Receiving pairs won 55.5% of rallies, consistent with a limited serve advantage characteristic of men's doubles play. Very long rallies (≥13 shots) were the most frequent (35.2%). Short and medium rallies favoured receivers, whereas long and very long rallies favoured servers. Most decisive shots were executed in zone 2, which acted as the primary operational area for rally control. Groundstrokes were closely associated with unforced errors, while volleys (especially in medium and long rallies) showed greater winning potential. The CHAID model identified rally length as the strongest predictor of point outcome, followed by striking and finishing zones. T-pattern analysis revealed recurrent temporal structures characterised by alternations between deep and mid-court zones and terminal volley actions. Men's doubles pickleball exhibits a distinctive internal logic based on cooperative spatial management, prolonged rallies, and incremental serving consolidation. These insights provide objective references for optimising training design and tactical preparation in doubles play.

Keywords

Decision-making match structure performance indicators point-resolution dynamics racket sports

Introduction

Pickleball is a relatively young racket sport, created in 1965 on Bainbridge Island (Washington, USA), combining elements of tennis, badminton and table tennis.¹ Over the past two decades, the sport has experienced exponential growth, becoming one of the fastest expanding disciplines both in the United States and internationally.² This rapid adoption has been attributed to its accessible technical demands, low injury risk, reduced economic barriers and strong social and recreational appeal.^3,4 Although initially more prevalent among adults and older adults due to its low-impact nature,^5,6 recent years have seen increasing participation among younger athletes and highly trained competitors, consolidating pickleball as an emerging competitive discipline.^7,8

Despite its rapid expansion, scientific research on pickleball remains limited. Most studies have examined physiological, psychological or social effects of participation,^3,9 or aspects related to injury mechanisms and prevention.^10–12 However, technical–tactical and structural analyses remain scarce, particularly when compared with the extensive evidence available in established racket sports such as tennis, padel, badminton or table tennis.^13–17 In these disciplines, notational analysis has enabled the identification of key performance indicators, playing styles and training-task optimisation.^18,19 Additionally, recent conceptual work has emphasised the need to develop sport-specific models for emerging net games, including pickleball.²⁰

In competitive pickleball, systematic analyses have only recently been conducted using observational methodology applied to professional tournaments. Studies in men's singles have shown that most points are resolved in short or medium rallies and that the receiver wins a slightly higher proportion of points, with actions near the non-volley zone and unforced errors playing a decisive role.²¹ Research on women's singles has identified distinct technical-tactical patterns, including a slight advantage for the serve and a high incidence of unforced errors, particularly in backhand strokes during medium-length rallies.²² These contributions represent the foundational reference for understanding competitive structure in singles play.

However, men's doubles presents tactical particularities that require specific analysis. Unlike singles, doubles play involves constant interaction between partners, incorporating cooperative dimensions such as spatial coordination, court coverage, motor communication and synchronous offensive–defensive transitions.^23,24 These demands influence the functional distribution of roles (right- and left-side players) and the alternation between parallel and cross-court play, modifying rally length, shot frequency and tactical priorities. Evidence specifically from padel doubles and tennis doubles indicates a greater reliance on volleys and a reduced presence of baseline strokes under these cooperative conditions.^23,25,26

The first activity-profile analysis of professional men's doubles pickleball reported average rally durations of 10.7 ± 8.7 s, an effective playing time of 38.6%, and more than 60% of shots executed from the transition zone, with a predominance of forehands and cross-court patterns.²⁷ These findings highlight the importance of managing the intermediate space between the baseline and the non-volley zone, as well as maintaining technical stability during extended exchanges. However, while this contribution provided valuable temporal and spatial descriptors of competitive activity, it did not examine multivariate technical–tactical predictors of point outcome nor the sequential regularities underlying performance, leaving room for more integrative analytical approaches.

Taken together, this evidence suggests that men's doubles pickleball possesses a distinctive internal logic characterised by longer rallies, lower rates of unforced errors and a greater dependence on coordinated dyadic actions—patterns consistent with those observed in padel doubles^25,28 and in men's professional tennis doubles, where net positioning and partner coordination strongly predict point outcome.²⁹

However, no studies to date have examined men's doubles pickleball using an integrated technical-tactical approach combining descriptive, predictive and sequential analyses. In other racket sports, such approaches have clarified the interaction between spatial and temporal variables through multivariate procedures.^30,31 Decision-tree models such as CHAID, together with the detection of T-patterns using THEME 6 Edu,³² enable the identification of key performance predictors and recurrent behavioural sequences occurring above chance levels.

From a methodological standpoint, observational methodology³³ provides an appropriate framework for analysing behaviour in natural competitive contexts, ensuring ecological validity and high data-quality standards. Its integration with tools such as LINCE PLUS³⁴ and reliability procedures based on Cohen's kappa³⁵ ensures systematic and reproducible coding.

Within this theoretical and methodological framework, it is necessary to deepen the technical–tactical analysis of professional men's doubles pickleball to identify combinations of spatial, temporal and technical variables that determine competitive success. Therefore, the aim of this study is to analyse technical–tactical indicators and effectiveness patterns using a combined descriptive, predictive and sequential approach. Specifically, the objectives are to: (i) describe game structure and the distribution of key observed variables; (ii) identify multivariate predictors of point outcome using CHAID decision trees; and (iii) detect recurrent temporal sequences associated with successful performance using T-pattern analysis. The results aim to provide objective references for optimising training processes and tactical planning, contributing to the development of a robust scientific foundation for this emerging sport.

Method

Design

The research followed a systematic observational methodology,³³ aimed at examining the technical and tactical dynamics of men's doubles pickleball under real competitive conditions.

The design adopted was nomothetic, since it included all points recorded from different matches and tournaments within the same competition level; longitudinal, as it covered several events throughout one season; and unidimensional, because the analysis focused on the sequential order of actions.³⁶

This design framework is especially appropriate for racket and net sports, where the goal is to identify and interpret recurring behavioral patterns emerging naturally during elite competition.

Sample

The dataset comprised every point played in the men's doubles matches corresponding to the semifinal and final rounds of five professional events held during the 2023 Professional Pickleball Association (PPA) Tour: Las Vegas, Cincinnati, Kansas, Seattle, and Denver.

In total, 15 matches were examined, yielding 2344 valid points contested by 16 elite players (8 pairs) who qualified for the closing rounds of each tournament. All athletes competed at the highest professional level within the international PPA circuit. Although official ranking positions at the exact time of competition were not retrospectively accessible, all players had qualified for semifinal and final rounds of PPA Tour events, which ensures representation of the highest competitive level in professional men's doubles pickleball. The sample included both right- and left-handed players; however, handedness was not included as an analytical variable in the present design.

Because the analysis was based exclusively on publicly available recordings officially released by the tournament organization, and the researchers did not intervene in any stage of play or data collection, individual informed consent was unnecessary, consistent with the ethical principles governing observational studies.³⁷ The study protocol was evaluated and authorized by the Ethics Committee of the Faculty of Education and Sport Science (University of Vigo, application 04-090425).

Instruments

An ad hoc observational instrument was developed specifically for this study to analyze the technical and tactical structure of men's doubles pickleball (see Table 1 and Figure 1 in Section 3).

Figure 1.

Visual representation of striking and finish zones on the court.

Table 1.

Structure of the observational instrument, including frequency distributions, chi-square (goodness-of-fit and independence) results, and Cramer's V coefficients for the analyzed variables.

Criteria	Code	Description	N	%	χ² GOF	χ² independence
Service	FS	First service	2324	99.1	χ²= 2256.826	χ²= 17.835; p < 0.001*; V = 0.087
Service	SF	Service fault	20	0.9	p < 0.001*	χ²= 17.835; p < 0.001*; V = 0.087
Rally length (the serve shot is counted)	SH	Short rally (1–4 shots)	530	22.6	χ² = 255.741	χ²= 65.669; p < 0.001*; V = 0.167
	MD	Medium rally (5–8 shots)	684	29.2	p < 0.001*	χ²= 65.669; p < 0.001*; V = 0.167
	LN	Long rally (9–12 shots)	304	13.0
	VLN	Very long rally (13 + shots).	826	35.2
Striking zone (see Figure 1)	SZ1	Non-volley zone	40	1.7	χ²= 5894.823	χ²= 79.326; p < 0.001*; V = 0.184
	SZ2	Mid-court zone	1952	83.3	p < 0.001*
	SZ3	Back court zone, including the baseline	66	2.8
	SZ4	Deep court zone, behind the baseline	258	11.0
	SZ	Service zone	28	1.2
Finish zone (see Figure 1)	FZ1	Left front zone	26	1.1	χ²= 2858.341	χ²= 24.479; p < 0.001*; V = 0.102
	FZ2	Right front zone	20	0.9	p < 0.001*
	FZ3	Left mid-court zone	238	10.2
	FZ4	Right mid-court zone	256	10.9
	FZ5	Left back court zone	101	4.3
	FZ6	Right back court zone	104	4.4
	NT	Net shot	1025	43.7
	OUT	Shot out	574	24.5
Winner	SW	The point is won by the server	1044	44.5	χ²= 27.959	–
Winner	RW	The point is won by the returner	1300	55.5	p < 0.001*	–
Point ending	SWW	Server wins with a winner or a forced error	518	22.1	χ²= 28.027	χ²= 2327.863; p < 0.001*; V = 1.000
	SWUE	Server wins with an unforced error by the opponent	526	22.4	p < 0.001*
	RWW	Receiver wins with a winner or a forced error	652	27.8
	RWUE	Receiver wins with an unforced error by the opponent	648	27.6
Final stroke	ACE	Direct serve	0	0.0	χ²= 2441.631	χ²= 113.555; p = 0.001*; V = 0.220
	FH	Forehand	583	24.9	p < 0.001*
	BH	Backhand	193	8.2
	FHV	Forehand volley	628	26.8
	BHV	Backhand volley	598	25.5
	SM	Smash	192	8.2
	LB	Lob	16	0.7
	DS	Drop shot	10	0.4
	SC	Change of service due to an error in the service	20	0.9
	OT	Other type of stroke	104	4.4

The system was organized according to mutually exclusive and exhaustive (ME&E) categories, designed to record the specific behaviors characteristic of coordinated doubles play. Its conceptual foundation was adapted from the previously validated OI-PICKLEBALL-S23 instrument, originally applied in singles competition.^21,22 To ensure its content validity and precision, the instrument underwent expert review by three specialists in racket and paddle sports, as well as in observational methodology. The panel reached a consensus exceeding 95% agreement regarding the clarity, relevance, and operational coherence of all category definitions. This process guaranteed alignment between the new version and the theoretical framework of observational methodology in sport performance analysis.³³

Behavioral coding was carried out using the software LINCE PLUS, version 2.1.0,³⁴ a tool specifically designed for systematic observation in sport and behavioral sciences. This platform enables researchers to define custom category systems, perform frame-by-frame video analysis, and export the coded data in compatible formats for statistical and sequential analysis. Its use has been consistently validated in the literature for ensuring accuracy and reliability when studying competitive performance in racket sports.¹⁷

All variables, criteria, and operational definitions included in the observational framework are detailed in Section 3 (Table 1 and Figure 1).

Procedure

The observation process was conducted through a multi-stage protocol designed to ensure methodological rigor and data reliability.³³

First, all official recordings of the semifinal and final rounds from the selected tournaments were located and downloaded in high-definition (1080p) quality. The videos were obtained from publicly accessible and officially distributed sources, primarily through the Professional Pickleball Association (PPA) Tour YouTube channel (https://www.youtube.com/@PPAtour, accessed on 15 September 2024). Each match provided a complete court perspective, allowing accurate identification of player positions, stroke execution, and tactical sequences throughout every rally.

Second, two expert observers, both trained in pickleball and observational methodology, completed a specific training program to standardize the use of the coding instrument and software. This training lasted three weeks and included nine sessions of two hours each, during which the observers practiced using the category system and refined their operational consistency through sample matches from the 2023 season.

Third, a pilot reliability test was carried out on a subset corresponding to approximately 10% of the final sample (around 230 points), selected from matches not included in the definitive dataset. Both observers independently coded this sample following the operational definitions established for the instrument. The Cohen's Kappa coefficient (κ)³⁵ was used to assess intra- and inter-observer reliability, yielding values of 0.97 (intra, Observer 1), 0.96 (intra, Observer 2), and 0.98 (inter), which represent excellent reliability, as values above 0.80 are generally considered strong. Once reliability was confirmed, one observer proceeded to code the entire dataset of 2344 valid points.

Finally, the coded data were exported to a spreadsheet compatible with IBM SPSS Statistics and THEME 6 Edu,³² maintaining the temporal order of each event. This ensured the correct execution of descriptive, predictive, and sequential analyses.

Data analysis

All statistical analyses were performed using IBM SPSS Statistics (version 25.0; IBM Corp., Chicago, IL, USA) and THEME 6 Edu (PatternVision Ltd, Reykjavik, Iceland). A combined descriptive, inferential, predictive, and sequential analytical approach was applied to examine the structure of play and the variables associated with point effectiveness. The variables included in the decision-tree model (rally length, striking zone, finishing zone and final stroke) corresponded to the ME&E categories defined in the observational instrument.

The internal distribution of each categorical variable was first examined using chi-square (χ²) goodness-of-fit tests. Subsequently, χ² independence tests were computed through cross-tabulations to assess relationships among the main performance indicators (rally length, striking zone, finishing zone, and final stroke). Cramer's V was calculated as a measure of effect size, with magnitudes interpreted as trivial (<0.10), small (0.10–0.29), moderate (0.30–0.49), and large (≥0.50).³⁸ Adjusted standardized residuals (|residual| ≥ 1.96) were inspected to identify cells showing significantly higher or lower frequencies than expected. Statistical significance was established at p < 0.05.

The predictive component of the analysis was conducted using a CHAID (Chi-square Automatic Interaction Detection) decision tree to identify the combinations of technical–tactical variables that best predicted point outcome (server win [SW] vs. receiver win [RW]). The dependent variable was the point winner, and predictors included rally length, striking zone, finishing zone, and final stroke. Pearson's χ² was used as the splitting criterion (p < 0.05), and model reliability was evaluated through 10-fold cross-validation. No Bonferroni adjustment was applied in order to preserve model interpretability. Model accuracy and stability were assessed using resubstitution and cross-validation risk estimates.

Finally, T-pattern detection was performed using THEME 6 Edu to identify recurrent temporal and sequential structures of play. This method detects event combinations occurring more frequently than expected by chance, revealing underlying regularities in behavioral sequences.³⁹ Search parameters were set to a minimum of three events per pattern, a significance threshold of p < 0.005, and 90% redundancy reduction to minimize overlapping patterns.

Results

The findings are presented according to the three analytical dimensions applied in this research: (i) a descriptive and inferential examination of the main performance indicators, (ii) a predictive analysis through decision tree modeling, and (iii) a sequential analysis to detect recurrent temporal patterns of play.

Descriptive and inferential analysis

Table 1 and Figure 1 outline the categorical framework of the observational instrument employed in this study, together with the frequency distributions for each criterion and the corresponding chi-square (χ²) goodness-of-fit and independence tests, as well as the Cramer's V coefficients. These results provide a general quantitative characterization of the technical–tactical indicators observed throughout men's doubles competition. Figure 1 illustrates the spatial reference criteria integrated into the instrument, showing the specific zones of stroke execution and point resolution on the pickleball court. The joint interpretation of both elements facilitates comprehension of the coding logic and the statistical relationships analyzed in the following subsections.

A total of 2344 points were analyzed. The great majority of rallies were initiated with a valid first serve (99.1%), whereas service faults accounted for only 0.9% of cases.

Rally-length analysis showed that very long rallies (13 shots or more) were the most common (35.2%), followed by medium (29.2%), short (22.6%), and long rallies (13.0%). The distribution differed significantly from a uniform pattern (χ² = 255.741, p < 0.001), and the association with point outcome was significant (χ² = 65.669, p < 0.001, V = 0.167; small effect). Adjusted residuals indicated that medium (residual = + 3.7) and short rallies (residual = + 5.4) were more frequently won by the receiver, whereas long (residual = + 2.1) and very long rallies (residual = + 6.8) were more often won by the server.

In relation to striking zones, most final strokes were executed from the mid-court area (83.3%), followed by the deep court (11.0%), the back-court zone (2.8%), and the non-volley zone (1.7%). The distribution differed significantly from uniformity (χ² = 5894.823, p < 0.001), and the association with point outcome was significant with a small effect size (χ² = 79.326, p < 0.001, V = 0.184; small effect). Adjusted standardized residuals indicated that mid-court shots were more frequently associated with receiver wins (residual = + 4.4), whereas shots executed from the back-court zone (residual = + 7.2) and the deep court (residual = + 3.3) were more often associated with server wins. No significant deviations were observed for shots played from the non-volley zone, where residuals remained below the ±1.96 threshold.

Finishing-zone analysis showed that most points ended with a ball into the net (43.7%) or out of bounds (24.5%), while the remaining finishing areas—front, mid-court, and back-court zones—presented lower frequencies (1.1–10.9%). The distribution differed significantly from uniformity (χ² = 2858.341, p < 0.001), and the association with point outcome showed a small effect (χ² = 24.479, p = 0.001, V = 0.102; small effect). Adjusted residuals revealed that back-court left endings (FZ5) were more frequent in server wins (residual = + 3.3). Points ending at the net were more often won by receivers (residual = + 2.3), whereas out-of-bounds endings were slightly more frequent in server wins (residual = + 2.2).

At the outcome level, receivers won 55.5% of the points and servers 44.5%, a difference confirmed by the chi-square test (χ² = 28.027, p < 0.001). For the point-ending criterion, frequencies were similar across categories (22.1%–27.8%). Due to the categorical dependency between point ending and point winner, the association produced a large effect (χ² = 2327.863, p < 0.001, V = 1.000).

Finally, the analysis of the final stroke showed that most points were decided through volleys or groundstrokes. Forehand volleys (26.8%) and backhand volleys (25.5%) were the most frequent actions, followed by forehands (24.9%), and (at lower rates) backhands and smashes (both 8.2%). The distribution differed significantly from uniformity (χ² = 2441.631, p < 0.001), and the association with point outcome was significant (χ² = 113.555, p < 0.001, V = 0.220; small effect). Adjusted residuals indicated that backhand groundstrokes occurred more often in points won by the server (residual = + 4.1), whereas forehand volleys were more frequently associated with receiver wins (residual = + 3.7). Smashes did not exceed the ±1.96 threshold and therefore showed no meaningful deviations from expected frequencies.

Predictive model (decision tree analysis)

To address the predictive component of the analytical approach, a CHAID decision tree model was applied to identify the combination of technical–tactical variables that best predicted the outcome of each point. This procedure recursively partitions the dataset into statistically homogeneous subgroups using χ² tests (p < 0.05), generating a hierarchical structure of the performance indicators most closely associated with server wins (SW) and receiver wins (RW). The model included rally length, striking zone, finishing zone, and final stroke as predictors, with point winner as the dependent variable. Model reliability was assessed through 10-fold cross-validation, which showed a minimal increase in prediction error (resubstitution risk = 0.292; cross-validated risk = 0.307), indicating satisfactory stability.

Figure 2 displays the resulting decision tree. At the root node (Node 0), receiver wins represented 55.5% of all points. The first and most influential split was produced by rally length (χ² = 62.043, p < 0.001), separating short and medium rallies (Node 1) from long and very long rallies (Node 2). In short and medium rallies, receiver wins were more frequent (63.3%), whereas long and very long rallies favored server wins (52.9%).

Figure 2.

CHAID decision tree illustrating the combinations of technical–tactical variables that best predict point outcome (server win [SW] vs. receiver win [RW]) in men's doubles pickleball.

Node 1 (short/medium rallies) was subsequently divided by striking zone (χ² = 141.638, p < 0.001). When the final stroke was executed from mid-court or non-volley areas (Node 3), receiver wins reached 71.8%. This branch was further split by rally length (χ² = 87.124, p < 0.001), producing two terminal nodes: medium rallies (Node 6), with 63.3% receiver wins, and short rallies (Node 7), which showed the highest proportion of receiver wins in the model (95.0%). In contrast, shots executed from the back-court zone (Node 5) strongly favored server wins (93.3%), whereas deep-court shots (Node 4) resulted in a more balanced distribution (46.5% RW; 53.5% SW).

Node 2 (long/very long rallies) was also split by striking zone (χ² = 141.638, p < 0.001). Points ending with shots executed from the back-court zone (Node 5) again favored server wins (93.3%). In cases where the final stroke originated from mid-court or non-volley areas, finishing zone emerged as the next significant predictor (χ² = 34.093, p < 0.001). Points ending out of bounds or in the back-left finishing zone (Node 8) showed 80.2% server wins, whereas points ending at the net or in front finishing areas (Node 9) showed a higher proportion of receiver wins (59.0%).

Overall, the CHAID model revealed a consistent hierarchical structure in which rally length acted as the primary determinant of point outcome, followed by striking zone and finishing zone, leading to progressively more homogeneous performance profiles across terminal nodes.

Sequential pattern analysis (T-pattern detection)

To analyze point resolution dynamics in greater depth, a complementary sequential analysis was performed including only those rallies in which the final stroke was executed from zone 2 (SZ2). This zone accounted for the majority of final-stroke occurrences in men's doubles pickleball and thus offered the most representative basis for examining recurrent game patterns. Table 2 presents the distribution of these patterns according to rally length (short, medium, long, and very long) and type of final stroke (forehand, backhand, forehand volley, backhand volley, and smash). For each combination, the table reports the total number of observed sequences (100%) and the proportions of points won by the serving (SW) or receiving pairs (RW), along with the frequency of winners (W) and unforced errors (UE) within each subgroup.

Table 2.

Distribution of game patterns in men's doubles pickleball for points whose final stroke was executed from zone 2 (n = 1097).

Game pattern FH	n	%	Game pattern BH	n	%	Game pattern SM	n	%	Game pattern FHV	n	%	Game pattern BHV	n	%
SH-FH	46	100	SH-BH	12	100	SH-SM	2	100	SH-FHV	86	100	SH-BHV	60	100
SW-SH-FH	4	8.7	SW-SH-BH	2	16.7	SW-SH-SM	0	0	SW-SH-FHV	4	4.7	SW-SH-BHV	0	0
SW-SH-SWW-FH	0	0	SW-SH-SWW-BH	0	0	SW-SH-SWW-SM	0	0	SW-SH-SWW-FHV	4	4.7	SW-SH-SWW-BHV	0	0
SW-SH-SWUE-FH	4	8.7	SW-SH-SWUE-BH	2	16.7	SW-SH-SWUE-SM	0	0	SW-SH-SWUE-FHV	0	0	SW-SH-SWUE-BHV	0	0
RW-SH-FH	42	91.3	RW-SH-BH	10	83.4	RW-SH-SM	2	100	RW-SH-FHV	82	95.4	RW-SH-BHV	60	100
RW-SH-RWW-FH	10	21.7	RW-SH-RWW-BH	2	16.7	RW-SH-RWW-SM	2	100	RW-SH-RWW-FHV	36	41.9	RW-SH-RWW-BHV	10	16.7
RW-SH-RWUE-FH	32	69.6	RW-SH-RWUE-BH	8	66.7	RW-SH-RWUE-SM	0	0	RW-SH-RWUE-FHV	46	53.5	RW-SH-RWUE-BHV	50	83.3
MD-FH	112	100	MD-BH	38	100	MD-SM	48	100	MD-FHV	224	100	MD-BHV	194	100
SW-MD-FH	30	26.8	SW-MD-BH	20	52.7	SW-MD-SM	6	12.5	SW-MD-FHV	82	36.6	SW-MD-BHV	94	48.5
SW-MD-SWW-FH	6	5.4	SW-MD-SWW-BH	8	21.1	SW-MD-SWW-SM	6	12.5	SW-MD-SWW-FHV	58	25.9	SW-MD-SWW-BHV	58	29.9
SW-MD-SWUE-FH	24	21.4	SW-MD-SWUE-BH	12	31.6	SW-MD-SWUE-SM	0	0	SW-MD-SWUE-FHV	24	10.7	SW-MD-SWUE-BHV	36	18.6
RW-MD-FH	82	73.2	RW-MD-BH	18	47.4	RW-MD-SM	42	87.5	RW-MD-FHV	142	63.4	RW-MD-BHV	100	51.5
RW-MD-RWW-FH	28	25.0	RW-MD-RWW-BH	0	0	RW-MD-RWW-SM	42	87.5	RW-MD-RWW-FHV	80	35.7	RW-MD-RWW-BHV	60	30.9
RW-MD-RWUE-FH	54	48.2	RW-MD-RWUE-BH	18	47.4	RW-MD-RWUE-SM	0	0	RW-MD-RWUE-FHV	62	27.7	RW-MD-RWUE-BHV	40	20.6
LN-FH	42	100	LN-BH	26	100	LN-SM	34	100	LN-FHV	86	100	LN-BHV	88	100
SW-LN-FH	22	52.4	SW-LN-BH	16	61.5	SW-LN-SM	6	17.6	SW-LN-FHV	40	46.5	SW-LN-BHV	56	63.6
SW-LN-SWW-FH	6	14.3	SW-LN-SWW-BH	2	7.7	SW-LN-SWW-SM	6	17.6	SW-LN-SWW-FHV	32	37.2	SW-LN-SWW-BHV	34	38.6
SW-LN-SWUE-FH	16	38.1	SW-LN-SWUE-BH	14	53.8	SW-LN-SWUE-SM	0	0	SW-LN-SWUE-FHV	8	9.3	SW-LN-SWUE-BHV	22	25.0
RW-LN-FH	20	47.6	RW-LN-BH	10	38.5	RW-LN-SM	28	82.4	RW-LN-FHV	46	53.5	RW-LN-BHV	32	36.4
RW-LN-RWW-FH	4	9.5	RW-LN-RWW-BH	0	0	RW-LN-RWW-SM	28	82.4	RW-LN-RWW-FHV	36	41.9	RW-LN-RWW-BHV	24	27.3
RW-LN-RWUE-FH	16	38.1	RW-LN-RWUE-BH	10	38.5	RW-LN-RWUE-SM	0	0	RW-LN-RWUE-FHV	10	11.6	RW-LN-RWUE-BHV	8	9.1
VL-FH	132	100	VL-BH	66	100	VL-SM	94	100	VL-FHV	210	100	VL-BHV	244	100
SW-VL-FH	78	59.1	SW-VL-BH	36	54.6	SW-VL-SM	60	63.8	SW-VL-FHV	106	50.5	SW-VL-BHV	116	47.6
SW-VL-SWW-FH	22	16.7	SW-VL-SWW-BH	12	18.2	SW-VL-SWW-SM	60	63.8	SW-VL-SWW-FHV	68	32.4	SW-VL-SWW-BHV	80	32.8
SW-VL-SWUE-FH	56	42.4	SW-VL-SWUE-BH	24	36.4	SW-VL-SWUE-SM	0	0	SW-VL-SWUE-FHV	38	18.1	SW-VL-SWUE-BHV	36	14.8
RW-VL-FH	54	40.9	RW-VL-BH	30	45.4	RW-VL-SM	34	36.1	RW-VL-FHV	104	49.5	RW-VL-BHV	128	52.4
RW-VL-RWW-FH	22	16.7	RW-VL-RWW-BH	8	12.1	RW-VL-RWW-SM	32	34.0	RW-VL-RWW-FHV	70	33.3	RW-VL-RWW-BHV	94	38.5
RW-VL-RWUE-FH	32	24.2	RW-VL-RWUE-BH	22	33.3	RW-VL-RWUE-SM	2	2.1	RW-VL-RWUE-FHV	34	16.2	RW-VL-RWUE-BHV	34	13.9

Groundstrokes (forehand and backhand)

In short rallies, the receiving pair won the vast majority of points, and most of these were resolved through unforced errors committed by the striking team. This tendency was also evident in medium rallies, where wins by the receiving pair again predominated and unforced errors remained the main mode of point resolution. As rallies became longer, this pattern changed noticeably. In long rallies, wins were more evenly distributed between serving and receiving pairs, although unforced errors continued to account for most outcomes. In very long rallies, the balance shifted further towards the serving pair, which won most of these points, while unforced errors remained a key factor in determining the final outcome.

Volleys (forehand and backhand)

Volleys displayed a distinct pattern compared with groundstrokes. In short rallies, the receiving pair still won most points, but the proportion of winners increased substantially, especially in forehand volleys. In medium rallies, receivers continued to win more points overall, although winners were now more frequent than unforced errors, indicating a greater capacity to finish the point actively at the net. In long rallies, the distribution between serving and receiving pairs became more balanced, accompanied by a clear rise in the proportion of winners compared with groundstrokes. In very long rallies, the serving pair often regained the advantage, with a higher proportion of points finished successfully despite the persistent presence of unforced errors.

Smashes

Although smashes occurred less frequently in zone 2, their behavior was consistent across rally durations. Receiver wins predominated in short and medium rallies, mainly through winners, whereas in long and very long rallies, the serving pair won more points, reflecting an increased likelihood of executing smashes under favorable conditions as rallies extended.

Overall, the sequential analysis of zone-2 points showed a clear temporal structure. In short and medium rallies, receiving pairs won most points, largely due to unforced errors. As rallies increased in duration, the balance progressively shifted toward the serving pair, although unforced errors continued to be the principal means of point resolution across all situations. While this general pattern was observed across shot types, volleys produced more winners than groundstrokes, particularly in medium and long rallies. These results underline the combined effect of rally duration and stroke type on point resolution, as reflected in both the descriptive frequencies and the sequential structures identified through T-pattern detection.

Discussion

Game structure and rally dynamics

The present results reveal a distinctive structural profile in men's doubles pickleball, characterised by a clear advantage for the receiving pair. Across all points analysed, receivers won 55.5% of rallies, confirming the limited benefit of serving and extending earlier evidence from singles pickleball, where the serve has also shown reduced offensive value.^21,22 Recent analyses in doubles racket sports show that serve advantage is frequently neutralised by coordinated positioning and early defensive pressure, a trend also described in men's professional tennis doubles^29,40 and observable in pickleball doubles as well.⁴¹ The double-bounce rule and the rapid capacity of the returning team to apply early pressure likely mitigate any initial advantage derived from serving, particularly in doubles, where cooperative defensive structures are immediately activated.

Rally length emerged as a central determinant of game dynamics. Very long rallies (≥13 shots) were the most frequent category, followed by medium and short rallies, a pattern consistent with the activity profiles previously described for men's doubles pickleball.²⁷ This distribution suggests a stable exchange system in which pairs maintain prolonged sequences of coordinated movements and controlled tactical responses. Compared with singles, which typically presents shorter and more abrupt exchanges,^21,22 the higher proportion of extended rallies in doubles highlights the importance of shared spatial occupation and anticipatory collaboration. A similar tendency toward longer exchanges has been reported in recent elite padel research, where rally duration reflects enhanced tactical synchronisation and shared spatial management,⁴² as well as in professional tennis doubles, where coordinated behaviour between both partners facilitates sustained rallies and structured point development.⁴⁰

The association between rally length and point outcome, although small in effect size (V = 0.167), revealed a meaningful tactical distinction. Short and medium rallies predominantly favoured the receiving pair, possibly due to the immediate capacity to neutralise the serve and pressure the server's second shot. In contrast, long and very long rallies were won more frequently by the serving pair, suggesting that extended exchanges provide time for the serving team to reorganise, advance toward advantageous zones, and gradually regain tactical control. Comparable trends have been described in doubles racket sports such as padel, where the progression of the rally favours positional consolidation and incremental offensive pressure,^25,31,43 and in tennis doubles, where more experienced teams exhibit a greater ability to control extended rallies and accelerate point finishing when advantageous.⁴⁰

Collectively, these findings indicate that men's doubles pickleball is structured around a recurrent balance between early receiving advantage and progressive serving consolidation, with rally duration acting as a central modulator of tactical momentum and point outcome.

Technical–tactical indicators related to point resolution

The technical–tactical indicators analysed provide deeper insight into how points are resolved in men's doubles pickleball. The final-stroke zone showed a strong association with point outcome, with most decisive actions executed in zone 2. This area functioned as the main operational space for controlling exchanges, a pattern consistent with previous pickleball research highlighting the relevance of intermediate court positions for maintaining rally stability and enabling offensive transitions.^21,22,27 The strong concentration of actions in zone 2 observed in the present study reinforces its role as the tactical hub of doubles play, where both partners can coordinate interceptive movements and constrain opponent options, similar to the spatial dominance patterns described in tennis doubles.^29,44

Shot type also differentiated point outcomes. Groundstrokes were strongly associated with unforced errors across all rally lengths, especially in short and medium exchanges, where receivers capitalised on technical instability. This behaviour is consistent with individual pickleball analyses showing higher error rates and reduced winning probability for shots executed from deeper areas due to positional disadvantage and increased temporal pressure.^21,22 Evidence from other racket sports supports this tendency, with studies reporting that deep-court actions are associated with greater technical difficulty, reduced preparation time, and higher error likelihood.^45–48 Comparable patterns have also been documented in elite women's doubles tennis, where unforced errors from baseline positions constitute a major determinant of point outcome.⁴⁴

In contrast, volleys were associated with a higher proportion of winning actions, particularly in medium and long rallies. Their effectiveness highlights the tactical value of early interception, reduced opponent reaction time and proactive control of the exchange. This pattern parallels findings in pickleball singles, where offensive actions from intermediate areas (particularly volleys) have been identified as strong predictors of point success.^21,22 Research in other racket sports similarly emphasises the importance of interceptive shots during stabilised phases of play, particularly in padel,^25,49 and in tennis doubles, where coordinated net play and proactive interception have been shown to limit opponent variability and increase point-ending opportunities.^29,50

In very long rallies, serving pairs displayed higher success regardless of shot type, indicating that extended exchanges allow tactical reorganisation, positional advancement and access to more favourable striking opportunities. This progressive shift from receiving advantage to serving recovery has been documented in both singles and doubles pickleball.^21,27 Overall, the interaction between shot zone, shot type and rally duration reflects a coherent tactical logic: unstable phases favour the receiving side, whereas stabilised and prolonged rallies progressively increase the serving team's probability of success.

Discussion on temporal and sequential parameters by periods and categories

The CHAID model identified rally length as the main factor explaining point outcome in men's doubles pickleball. Short and medium rallies favoured the receiving team, whereas long and very long rallies progressively increased the serving team's chances of success. Although previous studies in pickleball have focused on singles, they similarly show that rally stabilisation allows players to recover initial disadvantage and reorganise tactically.^21,22 This temporal evolution is comparable to dynamics observed in other racket sports, where prolonged exchanges facilitate positional consolidation and more controlled tactical decisions.^40,49,51

Striking zone was the second most influential predictor. Final shots executed from deeper areas tended to benefit the serving team in extended rallies, likely because longer exchanges provide time to regain structure and advance into more favourable positions. In contrast, actions originating from intermediate zones favoured receivers in short rallies, where early pressure and reduced preparation time often provoke server errors. Finishing zone added further discrimination: net and advanced-court endings were more likely to result in receiver wins, whereas deep-court or out-of-bounds endings were more frequently associated with server success. This spatial logic aligns with evidence showing that controlling forward areas and reducing the opponent's reaction time are key determinants of effectiveness in racket sports,^46,47 and in tennis doubles formations that prioritise proactive net positioning.⁵⁰

The T-pattern analysis confirmed that doubles pickleball rallies do not evolve randomly: instead, players follow recurrent temporal sequences occurring above chance levels. Most structures included alternations between deep and intermediate zones, followed by volleys or overheads in the later stages of long rallies. According to Magnusson's framework,^32,39 these temporal regularities reflect the underlying constraints and preferred tactical pathways characteristic of interactive sports, supporting recent proposals emphasising anticipatory coordination and spatial co-management in doubles racket play.^29,44

Together, both analytical approaches reinforce rally duration and spatial progression as fundamental tactical axes in men's doubles pickleball, integrating temporal tendencies observed in singles with the cooperative dynamics specific to doubles play.

Practical implications, limitations, and future directions

Practical implications

The present findings have direct applications for coaching and training design. The persistent advantage of the receiving team in early rally phases suggests the need to strengthen the serving team's second shot and subsequent positional recovery. Coaches should integrate drills that emphasise coordinated progression into zone 2, structured routines for neutralising early pressure, and high-percentage placements designed to delay opponent initiative.

Given that zone 2 and volley actions were central to point resolution, training should prioritise approach sequences, volley-control exercises, interception under reduced reaction times and tactical patterns involving “progress–intercept–finish.” Extended rally drills replicating medium and long exchanges may help serving teams develop stability and transition capacity, reflecting the scenarios in which they perform best.

The identification of hierarchical predictors (CHAID) and recurrent temporal structures (T-patterns) suggests incorporating pattern-based training into practice. Reinforcing common temporal chains, such as “reset from deep zone → stabilise in zone 2 → volley resolution”, may strengthen automatised behaviours that increase competitive success. Finally, given the high frequency of unforced errors from groundstrokes, specific technical work on back-court stability could reduce early-rally vulnerability.

Limitations

This study focused exclusively on elite men's doubles and relied on video-based analysis, limiting generalisability and restricting the inclusion of biomechanical, physiological or contextual variables (e.g., score, match phase, player roles). The CHAID model is constrained by categorical data, and T-pattern sensitivity depends on event density and parameter selection. Additionally, player handedness was not incorporated as an analytical variable in the present observational design. Although both right- and left-handed players were represented in the analysed sample, lateral dominance was not systematically coded or examined in relation to spatial distribution, role allocation, or shot directionality. Given the potential influence of handedness on cooperative positioning and cross-court dynamics in doubles play, future research should integrate this variable to better understand its impact on tactical organisation and performance outcomes.

Future research

Future studies should incorporate tracking or biomechanical data to better characterize spatial dynamics and physical loads. Comparative analyses across competitive levels, sexes or modalities (singles vs doubles) may provide deeper insights into adaptive patterns. Machine-learning models and network approaches could refine tactical prediction beyond CHAID structures. Longitudinal designs would help capture the evolving tactical trends of a rapidly expanding sport.

Conclusions

This study provides the first comprehensive examination of technical–tactical behaviour in professional men's doubles pickleball using a combined descriptive, predictive and sequential approach. The findings reveal a distinctive internal logic characterised by (i) an early advantage for the receiving team, (ii) the central role of rally length as the primary determinant of point outcome, and (iii) the strategic importance of intermediate court spaces (particularly zone 2) as the operational hub for controlling exchanges.

Groundstrokes were strongly associated with unforced errors, especially in short and medium rallies, whereas volleys showed a superior capacity to produce winning actions in stabilised and extended exchanges. The CHAID model established a robust hierarchy of performance predictors (rally length - striking zone - finishing zone), while T-pattern analysis confirmed the presence of recurrent temporal sequences involving transitions between deep and intermediate zones followed by volley- or overhead-based resolutions.

Together, these results demonstrate that men's doubles pickleball is governed by cooperative spatial management, progressive consolidation throughout the rally, and a dynamic balance between receiving dominance in early phases and serving recovery in prolonged exchanges. These performance regularities are consistent with the spatial and temporal tendencies observed in the results section, particularly the dominance of zone-2 actions and the progressive consolidation of serving effectiveness in extended rallies. These insights provide not only theoretical implications but also concrete decision-making references for elite coaching in doubles pickleball.

Footnotes

Acknowledgments

This publication was made possible thanks to the research stays during the year 2025 at the Instituto Politécnico de Viana do Castelo [IPVC] - Escola Superior de Desporto e Lazer.

ORCID iDs

Iván Prieto-Lage

Xoana Reguera-López-de-la-Osa

Antonio José Silva-Pinto

Alfonso Gutiérrez-Santiago

Ethical considerations

This study was approved by the ethics committee of the Faculty of Education and Sport Science (University of Vigo, application 04-090425).

Consent to participate

Not applicable.

Consent for publication

Not applicable.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by the Ministerio de Ciencia, Innovación y Universidades (MCIU), the Agencia Estatal de Investigación (AEI), and the (EU) under the Project LINCE PLUS: Multimodal platform for data integration, synchronization and analysis in physical activity and sport [PID2024-156051NB-I00] (2025–2027), awarded to Iván Prieto-Lage and Alfonso Gutiérrez-Santiago, (grant number PID2024-156051NB-I00).

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

The data presented in this study are openly available in FigShare at 10.6084/m9.figshare.31220833

References

USAPICKLEBALL. History of the Game, https://usapickleball.org/what-is-pickleball/history-of-the-game/ (2021, accessed 26 August 2024).

Gupta

. Understanding the fundamental reasons for the growth of pickleball. J Student Res 2024; 13. DOI: 10.47611/jsrhs.v13i2.6795.

Buzzelli

Draper

. Examining the motivation and perceived benefits of pickleball participation in older adults. J Aging Phys Act 2020; 28: 180–186.

Ryu

Yang

Kim

ACH

, et al. Understanding pickleball as a new leisure pursuit among older adults. Educ Gerontol 2018; 44: 128–138.

Cerezuela

Lirola

Cangas

. Pickleball and mental health in adults: a systematic review. Front Psychol 2023; 14: 1137047.

Casper

Bocarro

Drake

. Physical activity associated with older adult pickleball participation: a pilot study. Recreat Sport J 2023; 47: 49–56.

Terrell

Ficquette

. Exploring training strategies to optimize court performance in older pickleball athletes. Strength & Conditioning Journal 2023; 45: 1–12.

Stroesser

Mulcaster

Andrews

. Pickleball participation and the health and well-being of adults—A scoping review. J Phys Act Heal 2024; 21: 847–860.

Heo

Ryu

Yang

, et al. Serious leisure and depression in older adults: a study of pickleball players. Leis Stud 2018; 37: 561–573.

10.

Casals

Jimenez

Caparros

, et al. Scoping review and quality of studies on the epidemiology of pickleball injuries. Apunt Sport Med 2023; 58: 100403.

11.

Forrester

. Pickleball-Related injuries treated in emergency departments. J Emerg Med 2020; 58: 275–279.

12.

Weiss

Dougherty

DiMaggio

. Non-fatal senior pickleball and tennis-related injuries treated in United States emergency departments, 2010–2019. Inj Epidemiol 2021; 8: 34.

13.

Abian-Vicen

Castanedo

Abian

, et al. Temporal and notational comparison of badminton matches between men’s singles and women’s singles. Int J Perform Anal Sport 2013; 13: 310–320.

14.

Malagoli Lanzoni

Di Michele

Merni

. A notational analysis of shot characteristics in top-level table tennis players. Eur J Sport Sci 2014; 14: 309–317.

15.

Sánchez-Alcaraz

Courel-Ibáñez

Muñoz

, et al. Analysis of attacking actions in professional men's padel. Apunt Educ Física y Deport. 2020; 142: 29–34.

16.

Escudero-Tena

Sánchez-Alcaraz

García-Rubio

, et al. Analysis of game performance indicators during 2015–2019 World Padel Tour seasons and their influence on match outcome. Int J Environ Res Public Health 2021; 18: 4904.

17.

Prieto-Lage

Paramés-González

Torres-Santos

, et al. Match analysis and probability of winning a point in elite men’s singles tennis. PLoS One 2023; 18: 1–16.

18.

Fitzpatrick

Stone

Choppin

, et al. A simple new method for identifying performance characteristics associated with success in elite tennis. Int J Sports Sci Coach 2019; 14: 43–50.

19.

Prieto-Lage

Reguera-López-de-la-Osa

Durán-Rodríguez

, et al. Assessing the probability of winning a point in men’s padel: a comprehensive analysis. Appl Sci 2024; 14: 6642.

20.

Krizkova

Tomaskova

Tirkolaee

. Sport performance analysis with a focus on racket sports: A review. Appl Sci 2021; 11: 9212.

21.

Prieto-Lage

Reguera-López-de-la-Osa

Juncal-López

, et al. Notational analysis of men’s singles pickleball: game patterns and competitive strategies. Appl Sci 2024; 14: 8724.

22.

Prieto-Lage

Reguera-López-de-la-Osa

Vázquez-Estévez

, et al. Technical and tactical performance in women’s singles pickleball: a notational analysis of key match indicators. J Funct Morphol Kinesiol 2025; 10: 20.

23.

Courel-Ibáñez

Sánchez-Alcaraz

Cañas

. Effectiveness at the net as a predictor of final match outcome in professional padel players. Int J Perform Anal Sport 2015; 15: 632–640.

24.

Lupo

Condello

Courel-Ibáñez

, et al. Effect of gender and match outcome on professional padel competition. RICYDE Rev Int Ciencias del Deport 2018; 14: 29–41.

25.

Mellado-Arbelo

Baiget Vidal

Usón

. Analysis of game actions in professional male padel. Cultura_Ciencia_Deporte 2019; 14: 191–201.

26.

Martínez-Gallego

Ramón-Llin

Crespo

. A cluster analysis approach to profile men and women’s volley positions in professional tennis matches (doubles). Sustainability 2021; 13: 6370.

27.

Lozano

Casals

Ampuero

, et al. Match activity profile analysis during professional men’s double pickleball tournaments. Int J Perform Anal Sport 2025; 25: 868–880.

28.

Courel-Ibanez

Sánchez-Alcaraz Martinez

Muñoz Marín

. Exploring game dynamics in padel: implications for assessment and training. J Strength Cond Res 2019; 33: 1971–1977.

29.

Martínez-Gallego

Crespo

Ramón-Llin

, et al. Men’s doubles professional tennis on hard courts: game structure and point ending characteristics. J Hum Sport Exerc 2020; 15: 633–642.

30.

Martínez-Gallego

Guzmán

James

, et al. The relationship between the incidence of winners/errors and the time spent in different areas of the court in elite tennis. J Hum Sport Exerc 2013; 8: –5.

31.

Ramón-Llín

Guzmán

Muñoz

, et al. Analysis of shots patterns finishing the point in padel through decision-tree analysis. Rev Int Med y Ciencias la Act Física y del Deport 2022; 22: 933–947.

32.

Magnusson

. Discovering hidden time patterns in behavior: t-patterns and their detection. Behav Res Methods, Instruments, Comput 2000; 32: 93–110.

33.

Anguera

Blanco-Villaseñor

Losada

, et al. Guidelines for designing and conducting a study that applies observational methodology. Anu Psicol 2018; 48: 9–17.

34.

Soto

Camerino

Iglesias

, et al. LINCE PLUS: research software for behavior video analysis. Apunt Educ Física I Esports 2019; 137: 149–153.

35.

Cohen

. Weighted kappa: nominal scale agreement with provision for scaled disagreement of partial credit. Psychol Bull 1968; 70: 213–220.

36.

Anguera

Blanco-Villaseñor

Hernández-Mendo

, et al. Observational designs: their suitability and application in sports psychology. Cuad Psicol del Deport 2011; 11: 63–76.

37.

American Psychological Association. Ethical principles of psychologists and code of conduct. Am Psychol 2002; 57: 1060–1073.

38.

Cohen

. Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Hillsdale, NJ: Lawrence Erlbaum Associates, 1988.

39.

Magnusson

. Why search for hidden repeated temporal behavior patterns: T-pattern analysis with theme. Int J Clin Pharmacol Pharmacother 2017; 2: 128.

40.

Martínez-Gallego

Vives

Guzmán

, et al.

Time structure in men’s professional doubles tennis: does team experience allow finishing the points faster?

Int J Perform Anal Sport 2021; 21: 215–225.

41.

Martín-Miguel

Muñoz

Sánchez-Alcaraz

, et al. Final 2-shot sequence performed by the same pair during the world padel tour Madrid indoor 2023 master: sex differences in shot type and effectiveness. Int J Sports Physiol Perform 2025; 20: 1590–1599.

42.

de Ossó

Pecci

Sánchez-Trigo

, et al. Differences between genders and competitive levels on technical-tactical parameters in padel: implications for training. Int J Sport Sci Coach 2024; 19: 1349–1356.

43.

Escudero-Tena

Girard

Reid

, et al. Research and prediction of winning smashes in professional padel: men’s versus women’s matches and winning versus losing pairs. Int J Sports Physiol Perform 2025; 20: 840–847.

44.

Borderias

Crespo

Martínez-Gallego

, et al. Comparison of the game structure and point ending during grand slam women’s doubles tennis. Int J Sports Sci Coach 2023; 18: 1248–1255.

45.

Abdullahi

Coetzee

. Notational singles match analysis of male badminton players who participated in the African badminton championships. Int J Perform Anal Sport 2017; 17: 1–16.

46.

Delgado-García

Vanrenterghem

Muñoz-García

, et al. Probabilistic structure of errors in forehand and backhand groundstrokes of advanced tennis players. Int J Perform Anal Sport 2019; 19: 698–710.

47.

Barreira

Chiminazzo

JGC

. Who, how and when to perform winner points and unforced errors in badminton matches? An analysis of men’s single matches in the 2016 Olympic games. Int J Perform Anal Sport 2020; 20: 610–619.

48.

Sánchez-Pay

Ortega-Soto

Sánchez-Alcaraz

. Notational analysis in female grand slam tennis competitions. Kinesiology 2021; 53: 154–161.

49.

Sánchez-Alcaraz

Jiménez

Muñoz

, et al. Effectiveness and distribution of attacking strokes to finish the point in professional padel. Rev Int Med y Ciencias la Act Física y del Deport 2022; 22: 635–648.

50.

Kocib

Carboch

Cabela

, et al. Tactics in tennis doubles: analysis of the formations used by the serving and receiving teams. Int J Phys Educ Fit Sport 2020; 9: 45–50.

51.

Escudero-Tena

Martín-Miguel

Conde-Ripoll

, et al. Technical–tactical actions of continuity in professional and amateur padel: a comparative analysis by sex and level. Int J Sports Physiol Perform 2025; 20: 899–906.