A New Tool for Analysing the Rhythm of Play in Basketball: BASKETRHYTHM

Adrià Miró

Xavier Iglesias

Marcel·lí Massafret-Marimón

Franc García

Abstract

Systematic observation of the rhythm of play in basketball is essential for the planning of training sessions. This study pursued two objectives: a) to create an observational tool, BASKETRHYTHM, in order to analyse the rhythm of play in basketball that integrates technical and tactical actions, as well as the duration of possessions; and b) to conduct a pilot study so as to evaluate the rhythm of play with this new tool. Expert judgement showed a high validity of BASKETRHYTHM (76% positive agreement). The results of the pilot study indicated that a higher rhythm of play is associated with a higher effectiveness of possessions, and significant differences were observed between quarters and types of competition (domestic vs. international). The results of this study showed that BASKETRHYTHM is a valid and reliable tool for analysing the rhythm of play in basketball, which can contribute to optimising performance in this sport.

Introduction

The analysis of the game, which interprets its internal logic taking into account the players, the ball, the time and the regulatory spatial environment (Sautu Apellániz et al., 2009), allows us to better understand the variables involved in sport and to improve the performance of athletes during competition (Lago, 2022). Thus, game analysis is also useful to accurately plan athletes’ training (Malarranha & Sampaio, 2007; Ortega et al., 2006), as it helps to identify the physical demands required in each sport position (Garcia et al., 2020) and the key factors that contribute to victory  (Sampaio et al., 2010; Sampaio & Janeira, 2003). The concept of rhythm of play was developed at the Barcelona centre of the National Institute of Physical Education of Catalonia (INEFC) in order to study team sports from a qualitative perspective, adapted to the needs of the game. It focused on ball possession and was defined as the sum of technical actions and tactical actions, divided by total possession time. The concept of effective rhythm was also applied by counting only possessions that resulted in scoring (Solé, 2017). This group of teachers from INEFC Barcelona sought to have indicators of game control, such as this rhythm of play, directly linked to the specific coordination structure of each sport that was proposed, together with cognitive work, to seek quality and specificity in the training of team sports (Massafret, 2017).

In basketball, the rhythm of play—understood as the control over the speed at which the game develops across its different phases (Malarranha & Sampaio, 2007),—is a key indicator of the intensity of the playing model (Solé, 2017) and a fundamental element in the analysis of team play.

Previous studies show that a higher rhythm of play provides advantages in decisive moments and contributes to victories (Courel et al., 2014; García et al., 2023; Ortega et al., 2006). Evangelos et al. (2005)  analysed FIBA international games and found that winning teams had more counter-attacking actions (high rhythm of play), accounting for 10.8% of total possessions, compared to 9% for losing teams. These actions were resolved quickly, with high success rates (up to 97% in 1v0 situations) and showed a significant association with winning (χ² = 4.14; p < .05). These data reinforce the idea that a higher rhythm of play, especially through fast and direct actions, clearly contributes to the efficiency and ranking of teams, coinciding with the results of I Ibáñez et al. (2003) and Ortega et al. (2006).

On the other hand,  Bazanov & Rannama (2017)  studied 197 possessions in Estonian junior league players and concluded that offensive possessions with high levels of mechanical activity (> 1.0 g), high heart rates (165-191 bpm) and high collective intensity (0.83-1.55) corresponded with an offensive efficiency coefficient above 0.90. These conditions, typical of situations of high speed and intensity, show a direct relationship between rhythm of play and offensive efficiency and confirm that a high rhythm not only reflects a higher physical demand, but also a better offensive production. Thus, this improved performance referred to is specifically better offensive efficiency, i.e., an increase in points per possession, higher quality finishing and a better ability to take advantage of short or quick possessions to generate points. 

In addition, recent research has shown that physical intensity and workload during quarters can have a direct impact on competitive success. In this vein, the study by Miró et al. (2024) with U-18 players shows how winning quarters have significantly higher values in several external load indicators, such as Player Load (PL), steps per minute and Dynamic Stress Load (DSL). 

Rhythm of play is linked to tactical efficiency in studies on 3×3 basketball (Sansone et al., 2023). At the same time, training load control has been associated with lower injury incidence and better performance management in professional basketball, a fact that reinforces the importance of reliably measuring variables such as intensity and rhythm of play (Chan et al., 2024).

Systematic observation is widely used in basketball (Arias Estero et al., 2009; Arias Estero, 2012) to analyse elements that influence victories and defeats and to design training adapted to athletes (Ortega et al., 2006). The rhythm of play is defined according to the frequency of possession changes between teams (Charamis et al., 2023; Csátaljay et al., 2011), considering possessions as the period from when one team controls the ball until the opposing team recovers it (Charamis et al., 2023; Sakalidis et al., 2023.; Sampaio et al., 2010; Sampaio & Janeira, 2003). However, there is no methodological consensus as definitions and study contexts vary (Romarís et al., 2016. Some methodologies propose adding technical and tactical actions divided by the total time of possession (Bazanov, 2007; Bazanov et al., 2005, 2006; Solé, 2017), whilst others focus on analysing procedures separately, either by looking at how technical actions (Alamar, 2006; Charamis et al., 2023; Nunes, 2016; Sampaio & Janeira, 2003) or tactical actions (García et al., 2023) are executed. This heterogeneity makes it difficult to contrast and compare data (Romarís et al., 2016). Nonetheless, it seems that integrating tactical and technical actions in the same analysis (Bazanov, 2007; Bazanov et al., 2005, 2006) offers better results, as it is considered that the right balance of tactical and technical actions is the factor where success lies (Bazanov, 2007). Along these lines, recent studies have begun to incorporate tracking technologies to gain a more detailed insight into player behaviour and rhythm of play. Facchinetti et al. (2019) proposed a methodology to automatically identify active phases of play using kinematic data, and Metulini et al. (2017) used spatio-temporal analysis techniques to classify collective movement patterns. Although these studies provide advanced data-based insights, there is a lack of systematised observational tools applicable to real contexts and accessible to coaches and trainers. Nevertheless, tracking technologies offer great practical potential, as they allow for real-time monitoring of the tactical and physical behaviour of players and facilitate decision-making during training and matches, as well as the customisation of specific tasks according to the observed rhythm of play.

The aim of this study was therefore twofold: (1) to create an observational tool for analysing the rhythm of play in basketball (BASKETRHYTHM) that would take into account both technical and tactical actions, as well as the total time of possessions and their context, and to analyse its content validity and the reliability of observers in using it, and (2) to study the rhythm of play using the new tool in a pilot field test. It is hypothesised that the rhythm of play will be significantly higher on effective possessions, and also when the quarter and the match end in a team victory. Likewise, the rhythm of play will be higher when the team is playing at home.  

Methodology

Design

This research was carried out using systematic observational methodology (Anguera & Hernández-Mendo, 2013). The study design was idiographic—focusing on the team as a single unit rather than on individual players; cross-sectional—analysing different matches without considering their progression or development; and multidimensional—examining various levels of response, including technical actions, tactical actions, and the contextual aspects of the rhythm of play. 

Procedure

For the purpose of meeting the objectives of the paper, the following phases were defined: (1) development and validation of the instrument, (2) training in the use of the instrument and observation reliability process, and (3) development of the pilot study. 

Four professionals participated in the design of the observational tool: three Spanish basketball coaches and university lecturers of the degree in Physical Activity and Sport Sciences and a PhD student in Physical Activity and Health. The procedure of creating the new observation tool involved identifying technical and tactical actions and time control of possessions. 

For the content validity analysis, a panel of eight experts was selected according to strict criteria: Spanish basketball coaches with a minimum of ten years of experience and a degree in Physical Activity and Sport Sciences. Using an online questionnaire, the experts assessed the 11 criteria defined in the tool, indicating their agreement with “YES, I agree” or “NO, I do not agree”. This allowed the instrument to be validated in terms of its suitability and relevance for the analysis of rhythm of play.

In order to assess intra-observer and inter-observer reliability, two researchers were specifically trained in the use of the new BASKETRHYTHM observation instrument, integrated in the LINCE PLUS recording software (Soto-Fernández et al., 2022), in which 100 technical and tactical actions of a basketball game were recorded. For intra-observer reliability, the main researcher carried out two independent registrations with a time separation of one week between assessments, while for inter-observer reliability, a second researcher followed the same procedure independently. 

Participants

With the tool validated, a pilot study was carried out to test its applicability. The analysis focused on a professional basketball team from the Spanish first division that also competed in the Euroleague. The sample was selected by convenience sampling, taking into account the availability of high-level equipment and access to the full recording of the matches. The study participants were 14 professional basketball players (age 28.8 ± 4.3 years, height 201.01 ± 7.84 cm, weight 99.94 ± 12.3 kg). Similarly, the team analysed followed a training methodology called “structured training” (Pons et al., 2020; Tarragó et al., 2019). Although the sample is not intended to be representative of the entire population of basketball games, its size was considered adequate for this exploratory study. As this was a pilot and exploratory study to assess a new observation tool, a calculation of the sample size using tools such as G*Power was not applied. However, for future research, it is recommended that these criteria be applied to ensure adequate statistical power.

With a view to have some comparison records, four Liga Endesa (Liga ACB) matchesand four Turkish Airlines Euroleague (Euroleague) matches were chosen. In half of the matches, the team played at home, and in the other half, as the away team. The selected games were evenly split between wins and losses. 

The recordings of the matches, which were private, were provided by the analysed team and met high image quality criteria (1080p, 60FPS). This study did not require the approval of an ethics committee or the informed consent of participants, as it was a non-participant observation, based on videos recorded in natural settings, without any intervention or interaction with the observed participants (American Psychological Association, 2017). Registration took place between 18 January and 2 February 2024. The exclusion criterion was the inability to observe any of the actions during the matches.

Rhythm of Play Observation Instrument

The observational tool designed to analyse the rhythm of play, named BASKETRHYTHM (Table 1), includes four technical actions: ball handling (dribbling), passing communication, different types of finishing, and offensive rebounding; and five tactical actions: 1v1 situations, hand-offs between players, direct and indirect screens, and backdoor cuts. The tool also includes different parameters to understand the context of the rhythm of play: duration of possessions, categorised as short/medium/long, the start and end of these possessions, and the reason for time stopped during play.

Table 1

Criteria and categories of the new observational tool BASKETRHYTHM created to analyse the rhythm of play in basketball

See Table

Thus, rhythm of play is defined as the sum of technical and tactical actions divided by the total time of possession (Solé, 2017), and is expressed in arbitrary units (a.u.). This article aims to provide objectivity and systematisation to this concept and its orientation and therefore, in our methodology, the rhythm of play will be calculated according to the following formula:

A New Tool for Analysing the Rhythm of Play in Basketball: BASKETRHYTHM

See Full Size

a.u. = arbitrary units

n = number of actions

s = seconds

Recording Instrument

LINCE PLUS (Soto-Fernández et al., 2022) has been used both to analyse observer reliability and to record data in the pilot study from the observational instrument BASKETRHYTHM.

Data Quality and Analysis 

The content validity of the tool was established through the responses of eight experts (80% of those invited to participate), all men, selected on the basis of authority and following similar validation models (Soriano et al., 2024). There were 234 positive matches (YES-YES) out of a total of 308 possible matches, for a proportion of 76%. A 95% confidence interval (70.8% – 80.6%) was calculated using the binomial model (binom.test() function in RStudio) (© 2009-2021 RStudio, PBC v.1.4.1717). According to established standards in observational sports studies, this value indicates acceptable validity. The percentage of positive matches was chosen as the main method of analysis because of its simplicity and frequent use in qualitative studies (Watts, 2021).

The intra-observer reliability was calculated on 100 technical and tactical actions of a basketball game, twice, with one week between evaluations, and a Cohen’s Kappa index of .97 was obtained. Inter-observer reliability was analysed on the same 100 actions. A second observer was trained with the observation and recording instruments, and a Kappa value of .98 was obtained. Both values are considered excellent according to the criteria of Landis & Koch, (1977)

Statistical Analysis

Descriptive statistics (mean, standard deviation, minimum and maximum) were calculated to characterise the duration of possessions per game. To compare variables in two groups (games won/lost, home/visitor, effective/not effective possessions, Euroleague/ACB games), Student’s t-test for independent samples was used, applying the Bonferroni correction to adjust p-values for multiple comparisons. Previously, the assumptions of normality (Shapiro-Wilk test) and homogeneity of variances (Levene test) were checked, using non-parametric alternatives such as the Mann-Whitney U test if necessary. In addition, for each comparison, the effect size (Effect Size = ES) was calculated in order to assess the magnitude of the observed differences. Cohen’s d was used for parametric tests, the r coefficient  for Mann-Whitney U comparisons, and partial η² in ANOVA analyses. To compare three or more groups (such as the rhythm of play across quarters), a one-way ANOVA was applied, followed by Bonferroni post hoc correction to identify between-group differences. The statistical package JASP (Version 0.18.3; JASP Team; 2024) was used.

Results

The results of the rhythm of play analysis are presented in Table 2, where it can be seen that effective possessions have a higher rhythm of play (M = 0.68; SD = 0.21) with respect to the non-effective ones (M = 0.60; SD = 0.19), with statistically significant differences (F = 23.281; p < .01). These results can also be seen in Figure 1, which details the average values of the rhythm of play expressed in the arbitrary units described in the methodology.

Table 2

Descriptive values of the rhythm of play by quarter and effectiveness of the possessions

See Table

Figure 1

See Full Size

Note. Mean values and standard deviation by effective and ineffective possessions.

The post hoc analysis revealed a significant difference in the rhythm of play between effective and ineffective possessions. The rhythm of play was significantly higher in effective compared to ineffective possessions (t = 4.825, p < .001, SE = 0.40), suggesting that possessions with positive completion are associated with higher intensity of play.

As shown in Figure 2, the rhythm of play also shows a positive upward progression in the first quarter (M = 0.63; SD = 0.18), in the second (M = 0.64; SD = 0.23), and up to the third quarter (M = 0.68; SD = 0.2) and with statistically significant differences between the third and last quarter (M = 0,64; SD = 0.21) of the match (Figure 2; F = 3.651; p < .05). 

Figure 2

See Full Size

Note. Mean values and standard deviation by quarter.

Significant differences were only observed between the third and the last quarter of the match (t = 2.88, p = .025, SE = 0.34), indicating an increase in the rhythm of play in the third quarter compared to the last quarter. All other comparisons between rooms showed no statistically significant differences (p > .05).

The data analysed show a non-significant trend indicating that the rhythm of play would be higher when playing as an away team (M = 0.66; SD = 0.22) than when playing at home (M = 0.63; SD = 0.2) (Figure 3; F = 3.684; p = .055). 

Figure 3

See Full Size

Note. Mean values and standard deviation at home and away games.

Although the rhythm of play was higher when the team played as the away side (M = 0.66; SD = 0.22) compared to when playing at home (M = 0.63; SD = 0.20), this difference was not statistically significant (p = .055), indicating a non-conclusive trend towards a faster rhythm of play under away conditions.

In the matches of the team studied, the value of rhythm of play was significantly higher in European competition matches (M = 0.66; SD = 0.20), than in the domestic league matches (M = 0.62; SD = 0.21) (Figure 4; F = 5.912; p = .015). It is relevant to note that the majority of Euroleague matches (75%; n = 3) were played as away team.

Figure 4

See Full Size

Note. Mean values and standard deviation in domestic and European competitions.

The rhythm of play was significantly higher in the Euroleague games (M = 0.66; SD = 0.20) en compared to ACB League games (M = 0.62; SD = 0.21) (t = -2.43, p = .015, SE = –0.20). This significant difference suggests that European matches have a higher intensity of play than domestic league matches.

Contrary to our hypothesis, no significant differences were found in the rhythm of play between quarters or matches depending on whether the team won or lost. The analyses showed no significant effects either in the quarters (F = 1.658, p =.198) nor on the match result (F = 0.083, p =.77).

Discussion

The objectives of this study were, firstly, to create an observational tool to analyse the rhythm of play in basketball that considers both technical and tactical actions, the total time of possessions and their context, and to analyse its content validity and observational reliability. Secondly, we wanted to apply this tool in a field study to test its usefulness and to obtain practical data on the rhythm of play. As the main contribution of this work, we can confirm the validation of a new observational instrument, BASKETRHYTHM, to analyse the rhythm of play in basketball.

BASKETRHYTHM is an instrument that allows the technical and tactical actions of the game to be coded in a differentiated manner. The technical actions included are ball handling, passing, shooting, and rebounding, while the tactical actions comprise 1v1 situations, direct and indirect screens, hand-offs, and backdoor cuts. This approach improves the accuracy of the analysis, distinguishing individual contributions from technical skills and collective strategies and providing valuable information for both research and sport practice.

With regard to the first objective, the results confirm that the BASKETRHYTHM instrument has been designed correctly and shows content validity according to the answers provided by the experts. In addition, observer reliability was high, as indicated by the results obtained in the Kappa coefficients for intra-observer and inter-observer reliability. Therefore, this new tool for the analysis of rhythm of play represents a significant contribution for professionals in charge of the design and planning of basketball training sessions. 

With regard to the second objective, the data obtained from the pilot study with BASKETRHYTHMpartially confirm the hypotheses put forward and show that the effectiveness of finishes is greater when the rhythm of play is higher. The observed differences between effective and ineffective possessions are in line with the results of  Bazanov & Rannama (2017), who identified a direct relationship between physical intensity, mechanical load and offensive efficiency. Similarly, the study by Miró et al. (2024) shows how winning quarters are characterised by higher physical activity, with more steps per minute and a higher external load, which can be interpreted as an observable expression of the rhythm of play. These results support the use of BASKETRHYTHM as a useful tool to capture this tactical and physical dimension of the game and provide coaches with a practical tool to identify efficient patterns of play, adjust the intensity of training tasks and adapt strategies according to the phase of the match or observed effectiveness. However, as Solé (2017) points out, an increase in the rhythm of play does not always guarantee a better performance, as it can lead to a loss of efficiency due to an increase in technical or tactical errors, depending on the quality of the players and the opposing team. This underlines the importance of determining the optimal rhythm of play for each team, taking into account its specific characteristics and the competence of the opponent, so as to maximise performance without increasing errors.

On the other hand, the results obtained from the field study show statistically significant differences in the rhythm of play between quarters, specifically between the third and the last quarter of the match, with a positive and incremental trend in rhythm throughout the match, which is significantly reduced in the last quarter. This reduction in rhythm could be related to the team’s strategy of lengthening possessions in order to gain more control over the game, with slower tactical situations, including more stoppages and a longer duration of the quarter (García et al., 2020). Physiological factors, such as accumulated fatigue in the team, could also play a role (Stojanović et al., 2018). 

Contrary to what had been hypothesised, no statistically significant differences were found in the rhythm of play between matches and quarters considering their outcome, whether win or loss. Our results do not coincide with those of previous studies (Conte et al., 2017; Courel et al., 2014; Ortega et al., 2006), which indicated a significant relationship between high rhythm of play and successful outcome. Furthermore, some authors (Calvo et al., 2012; Ruano et al., 2007; Sampaio & Janeira, 2003) suggested that the rhythm of play is generally higher in matches played as home team, and this is also not true in our study, where the rhythm of play was higher in matches played as away team. It should be noted that this is a pilot study, with a limited sample of parties and without the intention of generalising conclusions. Therefore, statements such as these need further and more comprehensive analysis to minimise the influence of contextual factors and to draw more robust conclusions. This difference in results could be explained by differences in the type of matches analysed. For example, in our case, in most of the matches analysed the team acted as away team (75%; n = 3) in a demanding competition such as the Euroleague which, in general, can be considered more demanding (Guerra et al., 2016) than the ACB league. In contrast, in other studies (Calvo et al., 2012; Ruano et al., 2007) the matches evaluated are from domestic competitions and, therefore, perhaps also less demanding. 

Regarding the novelty of the research, this is the first study to develop a specific tool to observe the rhythm of play combining technical and tactical actions and temporal context in basketball from an observational perspective. This combination makes it possible to analyse collective behaviour in an integrated and applicable way. Many of the studies on basketball analysis are based on monitoring systems or physiological variables (Facchinetti et al., 2019; Metulini et al., 2017), where there is a lack of consensus on both analysis concepts (definition of intensity zones, load thresholds…) and methods (video analysis, microsensors, local positioning systems), a fact that evidences the need for effective tools to measure the rhythm of play in a valid and applicable way (Tuttle et al., 2024). The validated BASKETRHYTHM proposal allows a direct application by coaches, without high technological requirements, thus facilitating the methodological transfer to training contexts, semi-professional or without advanced digital resources.

This work has a number of limitations that need to be taken into consideration when interpreting the results. Firstly, the sample analysed was limited to one men’s team and eight matches, which limits the generalisability of the results. In the future, the sample analysed needs to be expanded and extended to women’s basketball, amateur basketball and training, as these data will help to develop better preparation programmes for male and female players, regardless of gender, sporting level and competition. Furthermore, a limited number of matches were considered in this pilot study, sufficient to assess the applicability of BASKETRHYTHM, but insufficient to generalise the results obtained. Finally, the analysis of technical and tactical rhythm is insufficient to understand and explain sporting success. 

Thus, the association between rhythm of play and match result is one of the elements of a very complex reality, which includes cognitive (players’ attitude towards the game), emotional (players’ degree of tension or anxiety), and contextual (venue/visitor, type of competition, etc.) elements. In fact, if this hypothesis were confirmed in future studies, our work would have contributed to the accumulated evidence indicating that the final outcome of a basketball game and the performance of players depends on many factors (Csátaljay et al., 2011; Csátaljay et al., 2012; Fox et al., 2021; Ortega et al., 2006; Ruano et al., 2007, among which high rhythm of play would only be one. 

Despite its limitations, the results of this study suggest—albeit preliminarily—that the new tool provides valid and reliable information for studying the rhythm of play in basketball.  

Conclusions

In this study, a new observation instrument, called BASKETRHYTHM, has been created and validated to analyse the rhythm of play in basketball, which considers both technical and tactical actions, as well as the total time of possessions. The results of the pilot study indicate that a high rhythm of play is associated with a higher effectiveness of possessions, although no direct relationship was found between rhythm of play and team victory or defeat. In addition, this varies significantly between match rooms, competitions and pitch location. 

The new BASKETRHYTHM instrument is presented as a useful tool both for future research and for tactical planning and analysis of sports performance in basketball teams. Coaches can use it to systematically assess the quality of play, identify effective patterns or moments of mismatch and, from that, design more specific training tasks. The tool makes it possible to adjust the intensity and rhythm of training situations according to the needs detected during the competition, as well as to prepare strategies adapted to the rhythm patterns of the opponents observed in previous matches.

Acknowledgements

The authors would like to acknowledge the support of the Spanish Government project entitled “Optimización del proceso de preparación y rendimiento en competición en deportes de equipo basada en integración de datos multimodales y multinivel mediante modelos inteligentes”  [PID2023-147577NB-I00], funded for the period 2024–2027 under the 2023 call for proposals for “KNOWLEDGE GENERATION PROJECTS”, within the framework of the State Programme to Promote Scientific and Technical Research and its Transfer, part of the State Plan for Scientific, Technical and Innovation Research of the Ministry of Science, Innovation and Universities (MCIU).