External and Internal Load Response to Different Refereeing Techniques and to Sex of Players in Basketball Games

Featured Application

The primary practical application of this research is to understand how the gender of players in a match and refereeing techniques influence the demands placed on basketball referees. The findings indicate that officiating female competitions imposes greater demands on referees compared to male competitions, and matches officiated by two referees require a higher workload than those officiated with three referees. These results may be valuable in reconsidering refereeing techniques, as well as the processes for promotion and physical evaluations. Until now, officiating female competitions has required a lower level of physical performance in fitness tests compared to officiating male competitions.

Abstract

Basketball referees play a crucial role in game management, yet the physical and physiological demands placed on them during a game remain understudied. This study analyzed the workload of 35 group 1 referees during a U-18 Spanish championship, examining the effects of refereeing technique (two referees vs. three referees) and competition sex (male vs. female) across game quarters. Physical and physiological demands were measured using inertial devices and heart rate monitors during 37 matches (18 men’s and 19 women’s). The results revealed that 2-referee teams experienced significantly greater physical demands, covering approximately 25% more total distance and demonstrating higher values in high-intensity running compared to 3-referee teams. Female competition elicited higher demands in specific variables, particularly in the distance covered above 16 km/h and average speed. Analysis across quarters showed distinct temporal patterns, with the first and third quarters presenting the highest demands, especially for 2-referee teams. These findings suggest that basketball referees’ physical preparation should be tailored to the officiating technique and competition sex, with a particular emphasis on developing specific conditioning programs for the 2-referee technique and implementing targeted recovery strategies between quarters to maintain optimal performance throughout the game.

1. Introduction

The quantification of workload in basketball has become a topic of increasing interest in both sports research and professional practice [1]. As basketball has evolved into a more competitive and physically demanding sport, the necessity of understanding how athletes manage both the physical and psychological load of the game has become paramount [2,3]. While traditional approaches have primarily focused on player performance, referees play a crucial role in the conduct of matches and the integrity of the sport [4].

This quantification process defines basketball as a dynamic sport requiring a high level of concentration, rapid decision-making, and optimal physical conditioning [5]. Referees, although often overlooked, are fundamental to the game’s development, as they are responsible for ensuring compliance with regulations and maintaining order on the court [6]. Their performance can significantly influence the outcome of a match, as well as the perception of the game by players and spectators [7]. Consequently, it is essential that we understand how to manage the workload that referees endure during matches [8].

This quantification can be carried out through various methods, providing valuable insights into the individualization of training. While some studies have evaluated officiating performance using heart rate monitoring [9], in recent years, the use of microtechnology and inertial devices has gained prominence. In this regard, García-Santos et al. [10] demonstrated that an international-level referee officiating a youth competition covered more than 4000 m with an average intensity of 61% of their maximum heart rate when matches were officiated under the 3-referee technique (3 Refs). Similarly, Ibáñez et al. [11], analyzing top-tier Spanish referees, found comparable distances being covered by referees in professional teams competing in the highest national league. However, the average heart rate intensity was 66%. Additionally, there were differences in maximum speed recorded (18 km/h vs. 21 km/h) between both studies. These demands are influenced by the match level, officiating experience, and sex of the competition [11]. Furthermore, contextual competition variables affect all participants, including players and referees [12]. In sports, workload refers to the amount of physical and mental stress that an athlete experiences during training or competition [13]. In the case of referees, competitive demands may be influenced by various psychological factors [12], decision-making processes [14], or error management [15]. The accurate assessment of these loads enables referees and their coaches to design more effective training programs tailored to the specific demands of officiating.

Understanding the workload experienced by referees provides multiple benefits. Firstly, it helps identify areas for improvement in their physical and mental preparation, contributing to more efficient and effective officiating [16]. Secondly, proper workload management can help prevent injuries, which are common in refereeing due to the physically demanding nature of the role [17]. Lastly, a deeper understanding of workload can aid in match and competition planning, ensuring that referees are adequately prepared for the specific challenges of each game [18].

By gaining a more comprehensive understanding of the demands that referees face, the officiating quality can be improved, ultimately enhancing the overall quality of the game. However, after reviewing the existing literature, we believe that while some contributions to sports science have positioned referees as the focal point of research, no studies have evaluated the same group of referees during a single championship under different officiating formats (in terms of the refereeing technique and the sex of the players).

In basketball officiating, the choice between a 2-referee (2 Refs) and a 3 Refs technique significantly impacts game management, decision-making accuracy, and physical demands on officials. Research suggests that the 3 Refs technique enhances game control by distributing responsibilities more effectively, thereby reducing individual workload and improving the accuracy of foul detection [19]. Additionally, the 3 Refs technique provides better coverage of off-ball actions, leading to a more comprehensive assessment of infractions and minimizing missed calls [20]. Conversely, the 2 Refs technique places greater physical and cognitive demands on officials, as they must cover larger areas of the court, often leading to increased fatigue and a higher likelihood of decision-making errors [21]. Despite these differences, some studies indicate that referees experienced in the 2 Refs technique develop compensatory strategies to maintain officiating quality [7]. Understanding these distinctions is essential in optimizing referee assignments, training programs, and officiating policies in competitive basketball. Therefore, the objectives of this study were to conduct a descriptive analysis of the workload experienced by referees in a championship considering the referees technique and sex of the competition and furthermore to examine workload evolution across game quarters.

2. Materials and Methods

2.1. Design

The design of this research was based on an empirical methodology. Specifically, it was a quasi-experimental study [22] with the objective of understanding the external and internal load requirements of basketball referees who participated in a tournament, analyzing different games and quarters of play. The reason for using a quasi-experimental design was because the objective of this research was to test a hypothesis by manipulating one or more variables, since a random group design could not be carried out.

2.2. Participants

This study included 35 referees belonging to group 1 and 2 (the second and third best at national level, with experience in amateur, semiprofessional, or professional basketball) who were in charge of refereeing the U18 tournament in Spain. In addition, all participants had previous experience in matches of this category and type of championship. All the referees participated in at least two matches.

The inclusion criteria to participate in this study were (i) to be a referee belonging to group 1 of the federation; (ii) not to have had a musculoskeletal injury in the 15 days prior to the competition; (iii) to have previous experience in the championship format; and (iv) to volunteer to participate in this study.

2.3. Sample

The sample for this research consisted of 37 matches from the championship. The matches analyzed were randomly preselected. Upon the completion of the championship, a database was designed, incorporating all the recorded values corresponding to the matches, along with various contextual variables characterizing the sample (the sex of the players and refereeing technique). The distribution of matches was as follows: 18 men’s matches (n = 11 officiated using the 2 Refs technique and n = 7 officiated using the 3 Refs technique) and 19 women’s matches (n = 12 officiated using the 2 Refs technique and n = 7 officiated using the 3 Refs technique). This structure resulted in a database comprising n = 88 unique cases analyzed at the match level and n = 352 unique cases analyzed to examine performance across the quarters of each match (Q1, Q2, Q3, and Q4).

The study was conducted in compliance with the ethical standards established by the university’s bioethics committee (reference requested: 233/2019) and in accordance with the procedures outlined in the Declaration of Helsinki [23].

2.4. Variables

The data analyzed have been classified based on two independent variables. Firstly, the variable Sex of the Match refers to the sex of the players in the analyzed match. Secondly, matches could be officiated using either the 2 Refs or 3 Refs technique.

The remaining variables considered for the various statistical analyses fall under the category of dependent variables. These variables have been previously used in a multitude of scientific investigations and are detailed below [12,24]:

• Total Distance: This is the amount that each referee travels in a specific period of the match (quarter or game), measured in meters (m).
• Total Distance > 16 km/h: Also called explosive distance, this is the amount that each referee travels in a specific period of the match (quarter or game) at a speed equal to or greater than 16 km/h. It is measured in meters (m).
• Maximum Aceleration (Max Acel): This is the maximum capacity to positively increase the speed of the race, measured in m/s².
• Maximum Deceleration (Max Decel): This is the maximum capacity to negatively increase the speed of the race, measured in m/s².
• Average Speed (Avg Speed): This is the average speed at which the referee moves during a specific period of time, measured in kilometers per hour (km/h).
• Maximum Speed (Max Speed): This is the average speed at which the referee moves during a specific period of time, measured in kilometers per hour (km/h).
• Player Load: This is the amount of load that the referee supports during a period of time. It is calculated by adding the accelerations that the subject receives on the 3 different axes and their intensity.
• Impacts > 8 G: This counts the number of times that the referee has endured G forces greater than 8 G. It is measured in count G force.
• Heart Rate Maximum (HRMax): This is the maximum number of heartbeats that a referee experienced, measured in beats per minute (bpm).
• >90% Heart Rate Maximum (>90 HRMax): This is the amount of time that the referee is above 90% of the maximum heart rate reached, measured in beats per minutes (bpm).

2.5. Materials

Each referee included in the sample was equipped with an inertial device from the WIMUPRO™ brand (Wimu, Hudl SL, Lincoln, NE, USA) to measure external load variables, as well as a GARMIN^® device (Garmin, Olathe, KS, USA) to record internal load variables. Data acquisition was conducted using an Ultra-Wideband (UWB) system, in which each playing court was equipped with a set of eight antennas placed around the perimeter, following methodologies employed in previous research [13] (Figure 1).

The collected data were subsequently analyzed using SPRO software (https://spro.cloud/#:~:text=SPRO%20is%20a%20professional%20ERP%20software%20designed%20for,automating%20workflows%2C%20streamlining%20tasks%2C%20and%20optimizing%20business%20efficiency (accessed on 15 July 2025)), developed by the aforementioned company. Each device utilized was capable of capturing both external and internal load variables at a sampling frequency of 18 Hz.

2.6. Procedure

Firstly, the Spanish Basketball Federation was contacted, and we presented the research proposal and the procedures to be carried out. Once the proposal was approved, the designated group of referees for the championship was contacted, and they were informed about the project, relevant details, and potential risks involved. Upon receiving the referees’ approval, the specific dates and matches to be analyzed were selected, and the referees who would be monitored were notified in advance.

On the day of the match, the research team set up the antennas and calibrated the systems 60 min before the game’s start. Thirty minutes before the match, the researchers accessed the referees’ locker room and equipped the designated officials with the necessary devices. The referees then completed their warm-up and officiated the entire match while wearing the equipment.

During the match, only the time in which the ball was in play was recorded (10 min per quarter) to ensure data consistency and maintain the validity of comparisons. Upon the conclusion of the match, 10 min after the final whistle, the research team re-entered the referees’ locker room to remove the equipment.

The collected match data were compiled into a database and classified according to the independent variables. At the end of the championship, the results were reported to the relevant stakeholders, including the federation and the refereeing body.

2.7. Statistical Analysis

Descriptive statistics are presented as mean and standard deviation (M ± SD). Prior to inferential analyses, the assumptions of homoscedasticity and normality were tested. Homoscedasticity was confirmed using Levene’s test, and the normality of data distribution was verified using the Shapiro–Wilk test for each variable within each group combination. All variables met these assumptions (p > 0.05), supporting the use of parametric analyses.

For inferential analysis, firstly, a two-way multivariate analysis of variance (MANOVA) was performed, to examine the effects of the refereeing technique (2 vs. 3) and competition sex (male vs. female) on physical and physiological demands, and secondly, to analyze the evolution of demands across game quarters, a repeated measures MANOVA was conducted with the quarter (Q1, Q2, Q3, or Q4) as the within-subjects factor and the refereeing technique and competition sex as between-subjects factors. Finally, when significant multivariate effects were found, univariate analyses were conducted to identify specific differences. Bonferroni adjustment was applied for multiple comparisons to control the Type I error rate.

The magnitude of differences was assessed using effect sizes. For MANOVA results, omega partial squared (ω_p²) was calculated and interpreted using the following thresholds: trivial (<0.01), small (0.01–0.05), moderate (0.06–0.13), and large (>0.14). For post hoc pairwise comparisons, Cohen’s d effect size was calculated and interpreted as trivial (0–0.2), small (0.2–0.6), moderate (0.6–1.2), large (1.2–2.0), or very large (>2.0).

Statistical significance was set at p < 0.05 for all analyses. All statistical procedures were performed using JAMOVI statistical software (Version 2.3.21, The JAMOVI Project, Sydney, Australia). Effect sizes and their confidence intervals were calculated using the ESCI module within JAMOVI. Graphical representations of the results were created using GraphPad Prism software (Version 9.0.0, GraphPad Software Inc., San Diego, CA, USA).

3. Results

During the U-18 tournament, basketball referees covered 4521.86 ± 594.02 m during matches, with 518.77 ± 142.30 m above 16 km/h and 38.24 ± 38.56 above 21 km/h. Their maximum acceleration (Acc_MAX) reached 4.89 ± 1.01 m/s², while their maximum deceleration (Dec_MAX) was −3.51 ± 0.63 m/s. The average speed (Speed_AVG) was 5.01 ± 0.46 km/h, with a maximum speed (Speed_MAX) of 22.69 ± 1.29 km/h. The neuromuscular load included 18.53 ± 29.43 high-intensity impacts (>8 g) and a Player Load (PL) of 47.66 ± 8.74 a.u. Their internal demands based on heart rate averaged 139.49 ± 15.39 bpm, representing 78.53 ± 9.20% of their maximum heart rate (HR_MAX).

3.1. Effect of Refereeing Technique and Competition Sex on Total Game Workload Demands

The number of referees strongly influenced physical demands, showing high effect sizes for total distance (F = 134.71; p < 0.001; ω_p² = 0.61), distance > 16 km/h (F = 76.00; p < 0.001; ω_p² = 0.44), Speed_AVG (F = 58.35; p < 0.001; ω_p² = 0.38), impacts > 8 g (F = 28.19; p < 0.001; ω_p² = 0.24), PL (F = 22.64; p < 0.001; ω_p² = 0.20), and Dec_MAX (F = 20.34; p < 0.001; ω_p² = 0.19). Speed_MAX showed moderate effects (F = 6.91; p = 0.01; ω_p² = 0.06) with higher values in 2-referee teams. No effects were found in Acc_MAX, %HR_MAX, and time above 90% of HR_MAX (F < 1.18; p > 0.28; ω_p² = 0) based on the number of referees.

The sex of games showed small effects on distance > 16 km/h (F = 4.73; p = 0.03; ω_p² = 0.02) and Speed_AVG (F = 4.45; p = 0.04; ω_p² = 0.02), with higher values in the female competition. No effects were found in other variables (F < 2.95; p > 0.15; ω_p² = 0). For this reason, the interaction between the refereeing technique and game sex showed small effects only for distance covered > 16 km/h (F = 6.29; p = 0.01; ω_p² = 0.03) and Speed_AVG (F = 5.34; p = 0.02; ω_p² = 0.03).

Figure 2 illustrates the pairwise comparisons between the game sex and refereeing technique. Teams with two referees showed very large effects for total distance in both sexes (t > 7.27; p < 0.001; d > 2.27) and distance > 16 km/h and Speed_AVG in the male competition (t = 7.27–8.11; p < 0.001; d = 2.27–2.42). Large effects were found for distance > 16 km/h in the female competition (t = 4.30; p < 0.001; d = 1.34) and impacts >8 g in the male competition (t = 4.30; p < 0.001; d = 1.34). A moderate effect size was observed for Dec_MAX and PL in both competitions (t = 3.15–3.46; p < 0.01; d = 0.94–1.08) and Speed_AVG and impacts > 8 g in the female competition (t = 3.02–3.69; p < 0.02; d = 0.94–1.15). Based on the competition sex, 3-referee teams obtained higher values in the female competition for distance > 16 km/h and Speed_AVG with a moderate effect size (t = 2.97–3.15; p < 0.02; d = 0.95–1.01).

3.2. Effect of Refereeing Technique and Competition Sex in Workload Demands by Game Quarter

Figure 3 shows the internal and external workload demands by game quarter based on the refereeing technique and sex of players in basketball games. Q1 showed significant differences between groups in most external variables. Total distance revealed significant differences (F = 27.30, p < 0.001, ω² = 0.48 large), with high-intensity distance (>16 km/h) also showing notable variations (F = 23.60, p < 0.001, ω_p² = 0.44 large), mainly between 2-referee teams and 3-referee teams, with a very-large-to-large effect size. For acceleration metrics, Acc_MAX displayed differences (F = 4.08, p = 0.009, ω_p² = 0.10 moderate) between 2 Refs–female and 3 Refs–male and between 3 Refs–male and 3 Refs–female with moderate effect, while Dec_MAX showed variations (F = 6.53, p = 0.001, ω² = 0.16 large) between 3 Refs–male and other groups with a moderate effect size. Both Speed_AVG (F = 17.40, p < 0.001, ω_p² = 0.36 large) and Speed_MAX (F = 4.55, p = 0.005, ω_p² = 0.11 moderate) demonstrated differences, with Player Load also showing significant variations (F = 8.31, p < 0.001, ω_p² = 0.20 large) primarily between 2-referee teams and 3 Refs–male teams with large-to-moderate effect sizes.

Q2 maintained similar patterns but with generally lower effect sizes. Total distance continued showing significant differences (F = 15.60, p < 0.001, ω_p² = 0.34 large), as did high-intensity distance (F = 15.90, p < 0.001, ω_p² = 0.34 large). Acc_MAX and Dec_MAX showed no significant differences during this quarter. Speed_AVG maintained significant differences (F = 7.90, p < 0.001, ω_p² = 0.19 moderate), while Speed_MAX showed limited differences (F = 2.93, p = 0.039, ω_p² = 0.06 moderate) only between 2 Refs–male and 3 Refs–female (moderate effect). Player Load showed moderate differences (F = 4.20, p = 0.008, ω_p² = 0.10 moderate) between 2-referee teams and 3 Refs–male teams.

Q3 presented the highest effect sizes in several variables. Total distance showed the largest effect size (F = 39.30, p < 0.001, ω_p² = 0.57 large), with significant differences between all groups, and high-intensity distance maintained significant differences (F = 22.10, p < 0.001, ω_p² = 0.42 large). While acceleration variables showed limited differences, Dec_MAX presented some variation (F = 3.25, p = 0.026, ω_p² = 0.07 moderate) between 2 Refs–female and 3 Refs–male (moderate effect). Speed_AVG showed strong differences (F = 23.40, p < 0.001, ω_p² = 0.44 large), and Player Load displayed significant variations (F = 14.10, p < 0.001, ω_p² = 0.31 large) between 2-referee teams and 3-referee teams.

Q4 maintained significant differences but with generally lower effect sizes compared to previous quarters. Total distance (F² = 13.40, p < 0.001, ω_p² = 0.30 large) and high-intensity distance (F = 17.70, p < 0.001, ω_p² = 0.37 large) continued showing differences between groups. Acceleration metrics showed no significant differences, while Speed_AVG maintained strong differences (F = 14.70, p < 0.001, ω_p² = 0.32 large). Player Load showed moderate differences (F = 4.68, p = 0.005, ω_p² = 0.11 moderate) only between 2 Refs–female and 3 Refs–male teams. Notably, no significant differences were found in heart rate variables or impacts > 8 G across any quarter.

3.3. Evolution Throughout the Game Quarters Based on Refereeing Technique and Competition Sex

Table 1. Descriptive and inferential analysis of the effect of the game quarter and the refereeing technique and sex of the competition on internal and external workload demands.

Variables	Refereeing Technique and Competition Sex	Game Quarter				F (p), ωp2 (Effect)	Post Hoc
		Q1 M ± SD	Q2 M ± SD	Q3 M ± SD	Q4 M ± SD
Total distance (m)	2 Refs–male	1244.32 ± 124.98	1244.32 ± 124.98	1244.44 ± 147.20	1222.06 ± 201.03	0.74 (0.532); 0
	2 Refs–female	1172.50 ± 203.19	1172.50 ± 203.19	1264.04 ± 97.52	1167.91 ± 146.68	2.22 (0.092); 0
	3 Refs–male	977.47 ± 135.18	977.47 ± 135.18	953.81 ± 80.81	996.30 ± 136.30	0.70 (0.554); 0
	3 Refs–female	998.31 ± 155.51	998.31 ± 155.51	1004.72 ± 133.83	976.26 ± 112.59	2.95 (0.039); 0.08 (moderate)	1–4 **
Total distance > 16 km/h (m)	2 Refs–male	147.30 ± 30.77	147.30 ± 30.77	144.86 ± 36.59	134.09 ± 33.54	5.81 (<0.001); 0.12 (moderate)	1–2 **, 1–3 **, 1–4 ***
	2 Refs–female	142.75 ± 43.16	142.75 ± 43.16	160.15 ± 30.15	138.42 ± 30.08	4.25 (0.008); 0.10 (moderate)	1–2 **, 1–4 **
	3 Refs–male	86.51 ± 28.36	86.51 ± 28.36	91.68 ± 32.88	77.89 ± 33.00	2.02 (0.118); 0
	3 Refs–female	114.08 ± 28.06	114.08 ± 28.06	110.14 ± 17.74	109.37 ± 24.98	4.60 (0.005); 0.13 (moderate)	1–2 **, 1–3 **, 1,4 **
AccMAX (m/s2)	2 Refs–male	4.33 ± 1.53	4.33 ± 1.53	4.12 ± 0.66	4.15 ± 1.35	0.38 (0.790); 0
	2 Refs–female	3.97 ± 0.42	3.97 ± 0.42	4.26 ± 0.58	4.06 ± 0.69	3.07 (0.032); 0.07 (moderate)	1–2 **
	3 Refs–male	3.87 ± 0.68	3.87 ± 0.68	3.90 ± 0.60	4.12 ± 1.24	0.42 (0.742); 0
	3 Refs–female	4.51 ± 0.55	4.51 ± 0.55	3.90 ± 0.70	4.40 ± 0.79	3.45 (0.021); 0.09 (moderate)	1–3 **, 2–3 **
DecMAX (m/s2)	2 Refs–male	−4.35 ± 1.38	−4.35 ± 1.38	−4.00 ± 0.70	−4.34 ± 1.24	0.89 (0.447); 0
	2 Refs–female	−3.96 ± 0.56	−3.96 ± 0.56	−4.29 ± 0.65	−4.23 ± 0.75	3.28 (0.025); 0.07 (moderate)	1–2 **
	3 Refs–male	−4.04 ± 0.76	−4.04 ± 0.76	−3.72 ± 0.55	−3.99 ± 1.55	0.80 (0.496); 0
	3 Refs–female	−3.93 ± 0.60	−3.93 ± 0.60	−3.78 ± 0.71	−4.13 ± 0.95	2.60 (0.059); 0.01 (trivial)
SpeedAVG (km/h)	2 Refs–male	5.15 ± 0.39	5.15 ± 0.39	5.20 ± 0.36	5.08 ± 0.38	8.22 (<0.001); 0.17 (large)	1–2 **, 1–3 **, 1–4 ***
	2 Refs–female	5.21 ± 0.37	5.21 ± 0.37	5.34 ± 0.25	5.23 ± 0.28	2.48 (0.067); 0.01 (trivial)
	3 Refs–male	4.83 ± 0.23	4.83 ± 0.23	4.73 ± 0.22	4.56 ± 0.30	14.20 (<0.001); 0.32 (large)	1–2 **, 1–3 ***, 1–4 ***, 2–4 **
	3 Refs–female	4.83 ± 0.30	4.83 ± 0.30	4.77 ± 0.30	4.89 ± 0.42	1.63 (0.190); 0
SpeedMAX (km/h)	2 Refs–male	21.59 ± 1.61	21.59 ± 1.61	21.90 ± 1.63	21.56 ± 1.50	2.50 (0.064); 0.01 (trivial)
	2 Refs–female	21.34 ± 1.46	21.34 ± 1.46	21.91 ± 1.58	21.73 ± 1.61	1.43 (0.240); 0
	3 Refs–male	21.26 ± 1.64	21.26 ± 1.64	21.00 ± 1.82	20.48 ± 1.63	1.42 (0.242); 0
	3 Refs–female	20.25 ± 1.33	20.25 ± 1.33	21.50 ± 1.52	20.68 ± 1.31	3.08 (0.033); 0.08 (moderate)	2–3 **
PL (a.u.)	2 Refs–male	12.43 ± 2.80	12.43 ± 2.80	12.46 ± 2.90	11.76 ± 3.01	0.92 (0.435); 0
	2 Refs–female	12.68 ± 3.13	12.68 ± 3.13	13.80 ± 1.83	12.58 ± 2.26	1.64 (0.186); 0
	3 Refs–male	10.48 ± 1.53	10.48 ± 1.53	10.07 ± 1.00	10.19 ± 1.85	0.32 (0.814); 0
	3 Refs–female	10.86 ± 2.01	10.86 ± 2.01	10.91 ± 1.60	10.71 ± 1.42	3.70 (0.016); 0.10 (moderate)	1–2 **, 1–3 **, 1,4 **
Impacts >8 g (n)	2 Refs–male	5.62 ± 9.60	5.62 ± 9.60	4.73 ± 8.47	4.31 ± 7.05	0.12 (0.946); 0
	2 Refs–female	7.91 ± 14.91	7.91 ± 14.91	6.64 ± 11.20	5.91 ± 9.45	0.12 (0.949); 0
	3 Refs–male	4.00 ± 4.67	4.00 ± 4.67	3.02 ± 4.55	3.26 ± 4.11	0.82 (0.485); 0
	3 Refs–female	3.78 ± 5.04	3.78 ± 5.04	3.17 ± 3.82	3.00 ± 3.74	0.17 (0.915); 0
HRMAX (%)	2 Refs–male	77.61 ± 11.36	77.61 ± 11.36	76.67 ± 9.33	75.15 ± 11.25	1.02 (0.389); 0
	2 Refs–female	77.11 ± 11.26	77.11 ± 11.26	77.10 ± 8.27	76.29 ± 10.50	0.51 (0.677); 0
	3 Refs–male	81.72 ± 6.22	81.72 ± 6.22	79.19 ± 6.81	77.28 ± 6.92	3.18 (0.028); 0.07 (moderate)	1–4 **
	3 Refs–female	79.47 ± 8.87	79.47 ± 8.87	77.22 ± 8.69	78.68 ± 8.56	0.67 (0.574); 0
>90% HRMAX (%)	2 Refs–male	15.75 ± 24.67	15.75 ± 24.67	11.24 ± 22.35	10.28 ± 20.03	1.54 (0.209); 0
	2 Refs–female	15.45 ± 24.53	15.45 ± 24.53	12.51 ± 19.76	12.65 ± 21.80	0.89 (0.451); 0
	3 Refs–male	13.49 ± 22.04	13.49 ± 22.04	8.71 ± 16.60	6.92 ± 12.39	1.86 (0.143); 0
	3 Refs–female	15.73 ± 24.32	15.73 ± 24.32	11.11 ± 18.33	12.73 ± 22.45	0.42 (0.737); 0

Note: Q1, Quarter 1; Q2, Quarter 2; Q3, Quarter 3; Q4, Quarter 4; F, F-value of MANOVA; p, significance; ω_p², omega partial squared; post hoc, 1–2 statistical differences between Q1 and Q2 (p < 0.05), 1–3 statistical differences between Q1 and Q3 (p < 0.05), 1–4 statistical differences between Q1 and Q4 (p < 0.05), 2–3 statistical differences between Q2 and Q3 (p < 0.05), 2–4 statistical differences between Q2 and Q4 (p < 0.05). Cohen’s d effect-size: ** moderate, *** large.

shows the internal and external workload demands throughout the game quarters based on the refereeing technique and competition sex. The results showed different patterns according to the analyzed variable. For total distance, only the 3 Refs–female group showed significant differences between quarters (F = 2.95, p = 0.039, ω_p² = 0.08 moderate), between Q1 and Q4, with moderate effect. In total distance >16 km/h, significant differences were found in three groups: 2 Refs–male (F = 5.81, p < 0.001, ω_p² = 0.12 moderate), with differences between Q1 and Q4 with a large effect and between Q1 and Q2 and Q1–Q3 with moderate effect; 2 Refs–female (F = 4.25, p = 0.008, ω_p² = 0.10 moderate), with differences in Q1–Q2 and Q1–Q4 with moderate effect, and 3 Refs–female (F = 4.60, p = 0.005, ω_p² = 0.13 moderate), showing differences in Q1–Q2, Q1–Q3, and Q1–Q4 with moderate effect.

Regarding Acc_MAX and Dec_MAX, only female groups showed significant differences; in Acc_MAX, this included both 2 Refs–female (F = 3.07, p = 0.032, ω_p² = 0.07 moderate), with differences in Q1–Q2 (moderate effect); and 3 Refs–female (F = 3.45, p = 0.021, ω_p² = 0.09 moderate), with differences in Q1–Q3 and Q2–Q3, with moderate effect. In Dec_MAX, only 2 Refs–female showed differences (F = 3.28, p = 0.025, ω_p² = 0.07 moderate) between Q1 and Q2, with moderate effect.

Speed_AVG revealed the most notable differences in 2 Refs–male (F = 8.22, p < 0.001, ω_p² = 0.17 large) with differences between Q1 and Q4 with large effect, and between Q1 and Q2 and Q1 and Q3 with moderate effect) and in 3 Refs–male (F = 14.20, p < 0.001, ω_p² = 0.32 large), showing differences in Q1-Q3 and Q1-Q4 with large effect, and in Q1-Q2 and Q2-Q4 with moderate effect. In Speed_MAX, only 3 Refs–female showed significant differences (F = 3.08, p = 0.033, ω_p² = 0.08 moderate) between Q2 and Q3, with moderate effect.

For PL, only 3 Refs–female presented significant differences (F = 3.70, p = 0.016, ω_p² = 0.10 moderate) between Q1 and Q2, Q1 and Q3, and Q1 and Q4 with moderate effect. Regarding impacts > 8 g, no group showed significant differences between quarters. Finally, for heart rate variables, only 3 Refs–male showed significant differences in HR_MAX (F = 3.18, p = 0.028, ω_p² = 0.07 moderate) between Q1 and Q4 with moderate effect, while no differences were found in >90% HR_MAX for any group.

4. Discussion

The present study aimed to analyze the physical and physiological demands on basketball referees during official games, examining the influence of the refereeing technique (2 vs. 3 referees) and game sex (male vs. female) on their workload demands, as well as investigating the evolution of these demands throughout game quarters. The main findings revealed that (a) 2-referee teams covered significantly greater distances and exhibited higher physiological demands compared to 3-referee teams, particularly in high-intensity activities; (b) the female competition generally elicited higher demands in specific variables, such as distance covered >16 km/h and average speed; and (c) the workload demands showed varying patterns across game quarters, with the most notable differences observed in Q1 and Q3, particularly for 2-referee teams.

4.1. Effect of Refereeing Technique on Physical and Physiological Demands

The greater physical demands observed in 2-referee teams compared to 3-referee teams represent a significant finding with important implications for basketball officiating. Teams with two referees covered approximately 25% more total distance and demonstrated significantly higher values in high-intensity running (>16 km/h), aligning with previous research in basketball officiating [12]. This substantial difference in physical demands can be attributed to the increased court area that each referee must cover to maintain optimal positioning for decision-making, a crucial factor in basketball officiating quality [25,26].

The analysis revealed particularly pronounced differences in high-intensity activities, with 2-referee teams covering significantly more distance above 16 km/h. This finding is consistent with research in soccer officiating, where fewer officials result in increased high-intensity running demands [27]. The greater demands placed on 2-referee teams are further evidenced by their higher average speed and increased Player Load, suggesting more intensive movement patterns throughout the game [17].

Interestingly, despite these marked differences in external load measures, the physiological response, as measured by heart rate parameters, showed similar patterns between 2- and 3-referee teams. This finding suggests that despite the increased physical demands, officials in both techniques maintained comparable cardiovascular stress levels, possibly indicating efficient adaptation to their respective workloads [28]. This cardiovascular response pattern aligns with research by Nabli et al. [29], who found that basketball referees develop specific fitness adaptations to meet the demands of their role, regardless of the officiating technique employed.

The implications of these findings are particularly relevant in referee preparation and management. The substantially higher physical demands placed on 2-referee teams suggest the need for more specific conditioning programs when officials are scheduled to work in this configuration [30]. Additionally, the similar physiological responses despite different external loads might indicate that referees in 2-referee technique develop more efficient movement patterns to cope with the increased spatial demands of their role [4]. These findings could have a direct influence on the positioning of both referees, affecting the movements and displacements that they perform under the two-referee system, which differ from those executed under the three-referee system due to their initial positioning. However, these types of movements were not analyzed.

4.2. Impact of Game Sex on Referees’ Performance

The analysis of game sex influence on referees’ performance revealed several noteworthy findings that contribute to our understanding of officiating demands in basketball. The female competition elicited higher demands in specific variables, particularly distance covered >16 km/h and average speed. This finding presents an interesting contrast with previous research by García-Santos et al. [10], who reported generally lower physical demands in female basketball games, and Vaquera et al. [31], who found reduced movement patterns in women’s competition officiating.

The higher demands observed in female competition could be attributed to different tactical approaches and game dynamics. As noted by Conte et al. [32], women’s basketball often involves more frequent transitions and a different spatial distribution of play, which might require referees to cover more ground to maintain optimal positioning. This interpretation is supported by the work of Allegretti Mercadante et al. [33], who documented distinct tactical patterns in women’s basketball that could influence officiating movement requirements.

The interaction between refereeing technique and game sex showed small but significant effects for distance covered >16 km/h and average speed. These findings align with research by Morocco et al. [34], suggesting that the impact of refereeing technique might be more pronounced in female competition. Particularly in the 3 Refs technique, officials in female games demonstrated higher values in these variables compared to male competition, a finding that deserves attention in referee preparation protocols.

The absence of significant differences in other variables between male and female competitions indicates that the fundamental demands of basketball officiating remain relatively consistent regardless of players’ sex. This observation supports Nabli et al. [35]’s assertion that core referee preparation and fitness requirements should maintain similar baseline standards across competitions. However, as pointed out by García-Santos et al. [17], the specific differences observed in high-intensity running patterns suggest the need for some tailored aspects in referee preparation based on the competition type.

These findings have important implications for referee training and assignment strategies. The higher demands in specific movement patterns during female competitions, particularly in the 3 Refs technique, suggest the need for targeted conditioning programs that address these unique requirements [36,37]. Additionally, the interaction between refereeing technique and game sex indicates that referee assigners should consider these factors when planning official rotations and workload management [38].

4.3. Workload Evolution Throughout Game Quarters

The analysis of workload demands across game quarters revealed distinct temporal patterns that provide valuable insights into the dynamics of basketball officiating. Q1 showed significant differences between groups in most external variables, with total distance revealing significant differences and high-intensity distance (>16 km/h) showing notable variations. These findings align with research by García-Santos et al. [17], who identified the opening quarter as a period of heightened physical demands for basketball referees.

The temporal analysis revealed that Q3 presented the highest effect sizes in several variables, with total distance showing the largest effect size. This pattern supports previous findings by Nabli et al. [35], who observed increased physical demands at the beginning of each half. As suggested by Vaquera et al. [26], this phenomenon might be attributed to both referees and players returning to the competition with renewed energy after the half-time break, resulting in higher movement intensities.

Speed-related variables showed particularly interesting patterns across quarters. Speed_AVG demonstrated strong differences, especially in Q3, while Player Load displayed significant variations between 2-referee and 3-referee teams. These findings support research by Leicht et al. [37], who documented similar temporal patterns in referee movement demands. The decline in physical performance parameters observed during Q4, particularly in 2-referee teams, is in concordance with studies by Ahmed et al. [14], suggesting cumulative fatigue effects on officials’ movement patterns.

The maintenance of heart rate responses across quarters, despite variations in physical demands, indicates efficient cardiovascular regulation throughout the game. This finding corresponds with research by García-Santos et al. [13], who observed similar cardiovascular stability patterns in elite basketball referees. The consistent heart rate response is particularly noteworthy for 2-referee teams, who must sustain higher physical demands while maintaining their decision-making capabilities, as highlighted by Nabli et al. [25].

Q2 maintained similar patterns to Q1 but with generally lower effect sizes, showing significant differences in total distance and high-intensity distance. This finding supports work by Ruiz et al. [30] suggesting that the second quarter represents a period of relative stabilization in game intensity. The observed patterns in Player Load and high-intensity activities across quarters align with recent research by Ibáñez et al. [39], indicating the need for specific conditioning strategies to address the temporal demands of basketball officiating.

These quarter-by-quarter variations have important implications for referee preparation and performance optimization. As suggested by Leicht and Connor [36], understanding these temporal patterns can inform more effective pre-game warm-up protocols and between-quarter recovery strategies. Additionally, the distinct demands of each quarter, particularly for 2-referee teams, highlight the importance of targeted conditioning programs that prepare officials for the varying intensities that they will encounter throughout a game [34].

4.4. Limitations and Future Research Directions

Several limitations should be considered when interpreting these results. The study did not account for the competitive level or importance of the games, factors that could influence the physical and physiological demands on referees. These limitations highlight the need for future research incorporating these variables to provide a more comprehensive understanding of basketball officiating demands. A final limitation was that all the referees analyzed were part of the same tournament. The sample was very large, and no previous research has included such a large number of referees. However, only one tournament was analyzed, and it is unknown whether the values obtained are repeatable patterns or represent an isolated event.

Future research should investigate the relationship between physical demands and decision-making accuracy across different officiating technique and competition types, particularly focusing on the impact of specific game contexts (e.g., close games, playoff matches) on referee demands. Studies examining the specific movement patterns and positioning strategies employed by referees in different team techniques, along with measures of cognitive load and decision-making performance, would help optimize officiating techniques and training protocols. These insights would be valuable in developing evidence-based guidelines for referee preparation and management, particularly for tournaments and congested fixture periods.

5. Conclusions

This study examined the physical and physiological demands on basketball referees, revealing three key findings that can advance our understanding of officiating requirements. The results showed that 2 Refs teams experienced substantially higher physical demands compared to 3 Refs teams, with greater total distances covered and more high-intensity running activities throughout the game. Additionally, female competition elicited higher demands in specific variables, particularly in distance covered at high speeds and average speed, with these differences being more pronounced with the 3 Refs technique. Finally, the analysis of workload evolution revealed distinct patterns across game quarters, with the first and third quarters presenting the highest demands, particularly for 2-referee teams, suggesting critical periods that require special attention in referee preparation and management.

The findings of this study have important implications for basketball referee preparation and management. Referees assigned to the 2 Refs technique require specific conditioning programs focused on high-intensity running and repeated sprint ability to meet the increased physical demands. These programs should emphasize the following:

• The development of aerobic capacity to sustain higher total distances;
• High-intensity interval training to cope with increased demands in critical game periods (Q1 and Q3);
• Recovery strategies between quarters, particularly for 2-referee teams;
• Specific movement pattern training based on competition sex differences;
• Position-specific conditioning drills that simulate actual game demands.

Moreover, referee assigners and sports organizations should consider these demands when planning referee rotations and rest periods, particularly in tournaments or congested fixture periods. The development of individualized training programs based on specific officiating techniques and competition types could help optimize referee performance and reduce fatigue-related decrements in decision-making. This could be accompanied by specific training for referees, where interval training of a similar intensity and duration to that found in competitions would be proposed.