1. Introduction
The quantification of training load is a tool that coaches use more and more, with the aim of improving sports performance, since it helps to control the progression of players [
1]. This is why training monitoring has become one of the most studied research topics in sport in recent years [
2]. This monitoring allows staff to obtain quantitative and qualitative information related to an athlete’s performance, which allows better understanding of their responses to exercise and the restructuring of training plans [
2,
3].
Previous research has established different variables that influence training load in team sports. Among them are the characteristics of the training system [
4] and the individual characteristics of the players [
5]. In relation to this, the player’s experience, position and playing time and their influence on basketball load have been analyzed [
6,
7,
8]. However, there are few studies in which individual characteristics have been related to the weekly training load [
9].
It has been proven that different performance indicators influence the load in different sports, both in training [
6,
10,
11] and in competition [
12,
13,
14]. There is scientific evidence on which performance indicators are the most important to win a game ([
15,
16,
17], and the effect of some situational variables on the game result have been verified. The most studied situational variables are: game location [
18,
19,
20], quality of opposition [
21,
22,
23], season phase [
9,
24,
25], score-line [
3,
19] and recovery cycle [
9,
26]. Generally, situational variables have been related to performance indicators. The only situational variables that have also been related to weekly training load are game schedule, playing time and season phase [
4,
27,
28].
The effect of situational factors on internal and external load have been studied recently, with results that lead to a deeper understanding of the training and competition processes. The place of play, phase of the season, result of the game, and score and quality of the opponent can explain 45% of the subjective internal load measured through RPE. In addition, it has been identified that female players in basketball have a higher RPE when playing away from home and playing against weaker opponents [
29].
Recent studies have shown how loads are affected by variable situations. Fernández-Leo, Gómez-Carmona, García-Rubio and Ibáñez [
30] found that in semi-professional basketball players there were differences in the volume of movements, high intensity actions, impacts and jumps depending on the playing position. The performance of the players decreased as the playing period increased. In addition, in balanced games there was the highest individual technical performance (PIR), while in unbalanced games more high intensity impacts were observed. However, they did not find differences in both physical and technical performance in terms of high difference in points per quarter. The outcome of the competition also influences the load of the players. Players who lose a match present higher values in kinematic variables, while players who win matches have a higher session RPE. The location of the match also influences the load, with the players playing at home presenting higher values in load (PLayerLoad) and in kinematic, objective internal load (HR) and subjective internal load (RPE) variables. This same trend is seen in matches in which the games are more balanced, with a smaller difference in points on the scoreboard [
19]. Along the same lines, Scanlan, Stanton, Sargent, O’Grady, Lastella and Fox [
31] identified that, when an extra time period is played, the internal (RPE, HR) and external (changes of direction, accelerations) load indicators increased with respect to regular matches. On the contrary, the load borne by the players (PlayerLoad) decreased as the quarters progressed.
Sansone, Gasperi, Tessitore and Gómez [
9], after following basketball players for one season, suggest that several individual characteristics (playing experience, playing position and playing time) and contextual factors (recovery cycle, level of upcoming opponent), recovery (playing experience, playing position) and in-game performance (opponent level, weekly training load, pre-game recovery) should be taken into account when monitoring training load. In addition, play actions performed when in possession of the ball and during live playing time in basketball games are affected by playing position [
32].
As can be seen, researchers are trying to understand the interactions between the load that athletes receive during competition with individual and situational factors in order to adjust the training processes, which are still in an incipient phase. These studies employ multiple variables, taking into account the needs and material resources of the researchers. For this reason, it is difficult to group research studies by variables. By way of summary,
Table 1 graphically presents various studies that have attempted to analyze the influence of situational factors on the indicators of internal and external load, physical performance and technical-tactical performance. It is also possible to observe quantitatively which have been the most used variables in the recorded studies.
To control and monitor the training load of the players, it is necessary to have information on performance during the season and how they respond to the training and the competition. For this, it is necessary to use instruments that give us useful information [
4]. Currently, technological development allows monitoring training and competition processes by means of inertial devices (IMUs) based on accelerometry [
39,
40]. The customization and adaptation of external load indicators through inertial devices is necessary to know the real demands of the female players [
41]. In basketball, recovery has been monitored along with physical load, using measures such as RPE, which is related to physiological factors and is a reliable method to assess the degree of fatigue of physical exercise [
42]. There are studies that have taken into account the schedule, playing time, season phase [
4,
27,
28] and the accumulation of matches [
43]. However, there are other situational and individual variables that can influence that training load and that recovery.
Player performance during games should be a variable to control. It is common in basketball to analyze individual performance according to the characteristics of the player and situational variables [
44,
45]. Recently, Fernández-Cortés, Mandly, García-Rubio, and Ibáñez [
24] identified a change in the intervention of female players according to the specific position in each phase of competition. During the regular league, guards (#1) have a greater contribution in field goals and 2-point shots, while power forwards (#4) contribute more free throws. During the play-offs, the point guards (#2) and power forwards (#4) increase their efficiency in 3-point shots. Finally, the offensive game during the play-offs is favored by the contribution of the guards through assists, while the defensive game is increased by the contribution of more rebounds by the centers (#5) and the defensive intensity through steals by the forwards (#3). There is little scientific literature in which this individual performance is related to weekly physical load [
9]. It would be interesting to know if there is any relationship between weekly load and the subsequent performance in the game.
Finally, the menstrual cycle has been shown to be a variable to be taken into account in the performance of female athletes [
46,
47]. Therefore, more and more sports professionals are paying special attention to the monitoring of this variable, which affects both the performance of female basketball players [
48] as well as the occurrence of injuries [
49,
50]. This is a variable that should be taken into account in studies that analyze the processes of training and sports performance in women, with the active and voluntary participation of female athletes being necessary.
Controlling the training load in competition will provide information on its demands [
9,
39]. Knowing these demands will allow coaches to make better decisions, reduce the risk of injury and improve performance [
9,
19]. Although it is considered that situational variables can condition performance during competition, most studies in basketball have focused only on some of these, such as the schedule, playing time and season phase. For this reason, it would be beneficial to include more specific ones like weekly load, game load and pre-game recovery.
As highlighted in the review of the state of the art, there is a predominance of studies that address the influence of situational factors on male basketball players, and it is necessary to analyze specifically how these factors affect female basketball players. This is one of the first studies of this topic in professional female players. For this reason, it is generally hypothesized that situational and individual factors will have an influence on the training load and game performance of the female players. Therefore, the aim of this study is to know the influence of situational and individual variables on training load (weekly load, game load and pre-game recovery) and individual performance in a team of Liga Femenina 2, LF2, during competition.
2. Materials and Methods
2.1. Subjects
The study was carried out on a team from group B of LF2 of the Spanish Basketball Federation during the 2020/2021 season. The team consisted of 13 semi-professional players aged between 19 and 46 (mean = 25.2 ± 7.3 years) of heights between 166 and 195 centimeters (mean = 178.2 ± 8.8 cm); all have at least 10 years of experience in this sport (mean = 15.8 ± 6.6 years).
The 28 games played by the team under study during the 2020/2021 season were analyzed.
2.2. Design
The design is non-experimental, retrospective, and cross-correlational. The study sought to identify relationships between the variables during the 2020/2021 season.
The variables under study were divided into dependent and independent. The dependents are as follows:
Weekly load (Session RPE): Borg’s adaptive scale was used with values from 1 to 10 [
51]. Multiplying the RPE value by the duration of the session gives the session RPE load in arbitrary units [
52]. Adding the load of each training session of the week, the weekly training load is obtained, not including that of the game [
12,
37,
53].
Game load: It is calculated by multiplying the game RPE by the playing time of each player [
4,
19].
Pre-game recovery: The Total Quality Recovery (TQR) scale was used, with values from 1 to 10 [
54]. The players had to mark their sensations of recovery.
Performance Index Rating (PIR): The PIR is a performance metric that is calculated from traditional basketball statistics [
55]. Its formula is: (Points + Rebounds + Assists + Steals + Blocks + Fouls received) – (Missed field shots + Missed free throws + Turnovers + Blocks received + Fouls committed). To normalize this metric, the value obtained for each player is divided by the playing time [
9,
56].
The independent variables are as follows:
Perceived individual performance: A scale from 1 to 10 is used [
51].
Playing position: Defined by the first coach that divided the players into 3 groups: (a) guards, (b) forwards and (c) centers [
9,
36].
Previous experience: 3 groups according to the years practicing basketball: (a) a lot of experience (more than 17 years), (b) medium experience (between 14 and 17 years) and (c) little experience (less than 14 years) [
9].
Game location: (a) local or (b) away [
19].
Score-line: (a) equal game (final difference was a maximum of 8 points) or (b) unequal (final difference was more than 8 points) [
19].
Playing time: 3 groups according to the average number of minutes per game: (a) high (more than 22 minutes), (b) medium (between 15 and 22 minutes), or (c) low (less than 15 minutes) [
9,
36,
57].
Season phase: Period of the year in which the game was played: (a) preseason (4 weeks before the start of the competition), (b) first round (between matchdays 1 and 13), or (c) second round (between matchdays 14 and 26) [
9,
37].
Recovery cycle: Days between games were counted: (a) short cycle (less than 7 days) or (b) long cycle (7 or more days) [
9,
37].
Previous game outcome: Result of the previous game: (a) victory or (b) defeat [
9,
19,
37].
Next game outcome: Result of the next game: (a) victory or (b) defeat [
9,
19,
37].
Quality of previous opposition: (a) high-level team (15 wins or more), (b) medium-level team (between 9 and 15 wins), or (c) low-level team (less than 9 wins). The level of the last opponent that had been played was considered [
9,
36,
37].
Quality of next opposition: (a) high-level team (15 wins or more), (b) medium-level team (between 9 and 15 wins), or (c) low-level team (less than 9 wins). The level of the next opponent to be played was taken into account [
9,
36,
37].
2.3. Procedures and Materials
Data recording was carried out throughout the 2020/2021 season. In the first session of the preseason there was an informative meeting with the players and the coaching staff to familiarize themselves with the scales to be used during the season and the operation of the mobile application “Quanter”, which was used to collect the data and the protocol to follow throughout the year. The participants signed the corresponding informed consent.
To obtain the recovery data, the mobile app simply sent them a notification in the morning so they could fill out the corresponding questionnaire. While, for the rest of the data, at the end of the session or the game, the mobile application sent them a notification so that they could fill out the corresponding questionnaire; this was done during the 15–30 minutes after the end of the session.
The team, in addition to 1 or 2 weekly games, had a total of 4 training sessions on the court and 2 or 3 in the gym. For home games, the players were summoned 2 h before the game time. When playing away, they also arrived at the court 2 h before the game to follow the same protocol as at home.
All data, collected through the Quanter mobile app, through M400 polar heart rate monitors and from the Spanish Basketball Federation, were stored in Microsoft Excel 2019 for subsequent statistical analysis with the IBM SPSS Statistics 21 program. Only the data of the players who played the game were taken into account.
2.4. Statistical Analysis
The variables that were classified into 3 categories by a k-means cluster analysis were: previous experience, playing time, quality of previous and next opposition, perceived individual performance, pre-game recovery, weekly load, game RPE and PIR.
The normality of the data was confirmed using the Kolmogorov–Smirnov test, and the homogeneity of the variances using the Levene test. Four linear mixed models for repeated measures were performed to evaluate the individual effect of each individual and situational variable on each dependent variable.
Subsequently, a pairwise comparison was made that was evaluated using the Bonferroni test, and the effect sizes were calculated using Cohen’s d. The values for effect size were interpreted according to the following scale: d ≤ 0.2 trivial, 0.2 < d ≤ 0.5 small, 0.5 < d ≤ 0.8 moderate and d > 0.8 large [
58].
Descriptive data were expressed as mean ± standard deviation (M ± SD). The level of significance was set at p ≤ 0.05. In the linear mixed models, the results were expressed with their p and F values.
3. Results
Table 2,
Table 3,
Table 4 and
Table 5 show the main significant effects of the individual and situational variables on each dependent variable.
Pairwise comparisons are shown in
Table 6,
Table 7,
Table 8 and
Table 9, as well as descriptive data (M ± SD).
Table 6 shows that the weekly load is higher in the preseason compared to the first round (
p = 0.012). In addition, when the recovery cycle was short, the accumulated weekly load was significantly lower (
p < 0.001).
In relation to the game load (
Table 7), this was significantly higher when the pre-game recovery values were high than when they were medium (
p < 0.001) or low (
p = 0.002). The players with a high PIR also accumulated more load in the games (
p < 0.001), and the more experienced players accumulated more load in relation to those with medium (
p = 0.008) or low (
p < 0.001) experience. Regarding playing time, the players who played more minutes had higher loads (
p < 0.001).
Table 8 shows that during the second round, the pre-game recovery data were significantly worse than those during the preseason (
p = 0.003) and the first round (
p < 0.001). When the recovery cycle was short, the pre-game recovery values were also significantly lower (
p = 0.018).
Table 9 shows that the players who perceived that their individual performance had been high also had a significantly higher PIR (
p < 0.001). The centers had a better PIR than the guards (
p = 0.025) and the forwards (
p < 0.001). The less experienced players had a lower PIR than the players with medium (
p = 0.003) and high (
p = 0.017) experience. The players who spent the most minutes on the court were the most valued (
p ≤ 0.001). When the game was won, the players’ PIR was also higher (
p = 0.010).
5. Conclusions
After analyzing the results, it can be concluded, for the sample studied, that:
- (a)
With respect to the weekly load, the more it accumulates, the more days there are between games and after a victory in the previous game. In addition, after a game against a low-level opponent, the weekly load should be higher.
- (b)
The weekly load is directly related to the players’ pre-game recovery value. In addition, as the season progresses, the recovery values are lower.
- (c)
If a team plays with a high-level opponent, the pre-game recovery value of the next game will be lower.
- (d)
The players who obtain the highest PIR are those who spend the longest time on the court, the ones who accumulate the most load during the games, and those who have a medium or high experience.
- (e)
The players who obtain the highest PIR are the centers.
The knowledge that information related to weekly load, game load and pre-game recovery has an impact on the individual performance of female players in basketball should be taken into account by coaches and physical trainers during the planning process. Specifically, they should reduce training sessions as the season progresses, so that in the last weeks of a season they should train less than at the beginning of the season, since the accumulated load at that time will be high. In addition, by knowing which female players give the best performance, coaches can plan game systems so that these players can obtain advantages. In the study conducted, these female players were the pivots.
For future studies, in addition to the variables included in this research, it would be very interesting to be able to have the information related to the menstrual cycle of the female players, since it has already been shown that it can have an influence on sports performance and should be considered in the processes of training planning and recovery of the female players.