Differences In Player Load Of Professional Basketball Players As A Function Of Distance To The Game Day During A Competitive Season

Purpose: The aim of this study was to investigate the dynamics of external training load (eTL), internal training load (iTL), and well-being status, during a regular season week with one game, and to examine the differential workloads of players depending on their distance from game day during a competitive season. Method: Subjects were 10 full-time professional basketball players (24.6 ± 4.9 years old; 204.2 ± 16.8 cm; 97.9 ± 10.4 kg). Workload was recorded and classified as total duration training and duration of full game during a competitive season. A wearable tracking system collected eTL via Player Load (PL) and Player Load per minute (PL/min). Training sessions were classified based on days before a match (four days before the match day = MD-4, MD-3, MD-2, and MD-1), and MD. Session rate of perceived exertion (sRPE) and rate of perceived exertion (RPE) were used for iTL. In addition, the Hooper index (HI) was used for well-being. Results: A significant difference was found between MD-1 and MD workload, MD workload being the highest of all variables: RPE ( p < .001), PL/min ( p <.001), PL ( p <.001), and sRPE ( p <.001). Regarding Hooper’s categories, significant differences between training days and match were only found in soreness ( p <.001). Conclusion: The results show that MD provides a unique stimulus in terms of volume and intensity. Consequently, coaches must incorporate specific training exercises to adapt players to the demands of competition. Finally, special attention should be paid to MD-2 and MD-1 in terms of potential accumulated fatigue and thus to ensure appropriate recovery time for athletes to adapt before the match.


INTRODUCTION
Basketball is an intermittent, indoor court-based team sport where high-intensity movements, such as changes of direction, accelerations, decelerations, and jumps, alternate with low-to-moderate-intensity periods (Conte et  According to the German statutory accident insurance (VBG, 2022), 66.2% of the players competing in the Basketball Bundesliga (BBL, first German national league) were injured in 2021, with an average injury rate of 93 injuries per 1000 hours of competition. Compared to the National Collegiate Athletic Association, the injury rate for athletes who compete appears to be significantly lower at 4.3 per 1000 athletes (Dick et al., 2007). Given that many of these injuries are attributed to excessive training loads, they might be largely preventable if the appropriate training loads were prescribed . Load monitoring approaches and feedback on the effects and adjustments during rehabilitation have been an integral part of the return-to-play algorithm (Locus et al., 2021) and can also assist in reducing the likelihood of maladaptive responses in players (e.g., illness, injury, or nonfunctional over -reaching; Locus  It is important to know if a training intervention has been effective and whether the team as a whole has benefited. Quantifying the specific demands of a sport is important not only for developing team training plans, but also for analyzing individual athletic performance (Clemente et al., 2020;Taylor et al., 2017). When considering the effects of individual characteristics on an athlete's response to training, it may be more beneficial to use an individual approach to model this relationship. For example, the same load stimulus may trigger different effects and adaptive responses (Borresen & Lambert, 2009) in two athletes due to variations in factors such as genetics or level of fitness (Bouchard & Rankinen, 2001), injury history (Hulin et al., 2016), and age (de la Rubia et al., 2020). Monitoring workload, when correctly managed, may lead to a better understanding of athletes' responses to stimuli and may allow to obtain the desired training response (Impellizzeri et  To be effective, training programs should be tailored to the load imposed during matches (Scott et al., 2014) with appropriate periodization of a daily and weekly microcycle. A commonly used approach is tapering. Tapering is a reduction in workload for a period prior to a competition to minimize the psychobiological stress of chronic training and thereby improve performance (Svilar et  The aims of load management is to reduce risk factors for injury and to optimizing decision making by the coaching staff. As such, monitoring the external load and the internal load, during both training and competition, is recognized as key in informing the management of athletes (Fox et  There are several possibilities to quantify training load such as changes of direction, accelerations, decelerations, and jumps used in basketball (Fox et al., 2017). One of the most commonly adopted tools to assess external load in basketball are wearable inertial measurement units (IMUs) (Fox et al., 2017;Russell et al., 2020). These devices collect inertial data and combine the instantaneous rate of change of acceleration in all three planes of movement to obtain a single measure of accumulated load that reflects the external load imposed on the athlete, such as the player load (PL). Further parameters that can be calculated from the data are player load per minute played (PL/min) (©Catapult Innovation, Melbourne, Australia), which can provide information about the inertial movements that players execute on the court (Fox et al., 2017 Therefore, the aim of this longitudinal study was the quantification of workload difference in external load measured via IMUs, the perceived training load and well-being status over a competitive basketball season. Knowing these changes in workload and physical demands during in-season macrocycles could help coaches, athletic-performance staff and medical staff to optimize training and match performance.

Participants
Sixteen professional basketball players from a German first league club ( and matches in the course of the competitive season, so no ethics committee approval was needed.

Study Design
This study followed a longitudinal approach during the 2021/2022 basketball season (August 2021 -Mai 2022). During the pre-season phase, players were familiarized with the monitoring tools used. Following this period, weekly training load, game performance data and well-being questionnaires were collected during 37 weeks of the competitive season (including all regular season and cup games). The team weekly schedule featured five team-based basketball sessions of 90-120 min, which focused on skills development, game-based conditioning, two physical training sessions of 40-60 min including strength, power, and speed training. The researchers did not intervene in the training plans or the tasks of the trainers. Therefore, the data for the analysis was collected four days before the match day (MD-4), three days before the match day (MD-3), two days before the match day (MD-2), one day before the match day (MD-1), and on MD.

Match Analysis
IMA analysis was made for the four quarters in every competitive game including the 30 min standardized warm-up, excluding the rest intervals between quarters (Torres-Ronda et al., 2016). Game quarters lasted a total of 19 to 26 min. Only the players on the court were analyzed. According to the FIBA rules, games consisted of four 10-min quarters, with 24-s shot clock, 2-min inter-quarter breaks and a 15-min half-time break (FIBA, 2018).

Practice Analysis
IMA analysis was made for all training sessions. All training sessions started with a standardized team warm-up and were performed on the practice or game court under similarly controlled environmental conditions. Players were allowed to consume water during recovery periods. During these practice sessions, groups of teammates and opponents were varied randomly.

Data Collection and Processing
Openfield TM was used to process inertial movement data. As described above, PL was calculated using the manufacturer's algorithm (t = time, fwd = forward acceleration, side = sideways acceleration, up = vertical acceleration), using the formula presented above. PL describes the sum of movements and their intensity in different axes during the entire activity or during one minute of the activity (PL/ min). Established literature refers PL as a reliable and reproducible metric in the quantification of cumulative motion for indoor sports (Peterson & Quiggle, 2017). The manufacturer's inertial movement analysis (IMA) can be used to analyze micro-movements, regardless of unit orientation and positional data. The algorithm considers tri-axial accelerometer, and gyroscope data (100 Hz

Monitoring of Well Being
The Hooper Questionnaire (Hooper & Mackinnon, 1995) with four categories (delayed onset muscle soreness -DOMS; stress, fatigue, and sleep) was completed approximately 30 min after awakening via an application (Catapult Forms). Each category can be rated from 1 to 7. For DOMS, stress, and fatigue, 1 represents "very, very low" and 7 means "very, very high". For sleep quality, 1 represents "very, very good" and 7 represents "very, very poor" (Clemente et al., 2017). The sum of the four categories is the Hooper index (Haddad et al., 2013). Lower indices mean better well-being.

Monitoring of External Training Load
The external training load (eTL) data were processed using the manufacturer's software

Statistical Analysis
The results were expressed as means (M) ± standard deviation (SD). The differences in player load, RPE, sRPE and Hooper categories for days with different distance to the match (MD-4, MD-3, MD-2, MD-1 and MD) were analyzed using one-way analysis of variance (ANOVA). Partial eta squared (η 2 p) effect size (ES) was used for ANOVA and classified as no effect (ES < 0.04); minimum effect (0.04 ≤ ES < 0.25); moderate effect (0.25 ≤ ES < 0.64); and strong effect (ES≥ 0.64) (Ferguson, 2009). The least significant difference (LSD) test was used in ANOVA as a post-hoc approach. Independent t-tests were used for pairwise comparison between training day and match day. All statistical analyses were carried out using the SPSS statistical analysis software (SPSS version 28.0, Chicago, USA). The level of statistical significance was set at p ≤ 0.05. All figures were produced using R Studio (Version 4.0.0).

DISCUSSION
This study aimed to describe the volume and intensity of in-season workload of professional basketball players and to compare the workload between training sessions (MD-4>MD-3>MD-2>MD-1) match (MD). Workload quantification of eTL was achieved via IMUs, iTL and wellbeing status were queried with an application before and after practice. The results of the study showed differences in all workload variables (volume and intensity) between the sessions analyzed (MD-4>MD-3>MD-2>MD-1 and MD). MD workload was the most demanding not only in volume but also in intensity. Significant workload differences in eTL (PL and PL/min) and iTL (RPE and sRPE) variables were only found between MD-1 and MD. In addition, Hooper index results show higher DOMS from MD-4 up to MD.
In team sports, tapering strategies have been implemented, as an attempt to decrease the stress of training and prepare players better for the official match (Moraes et al., 2017;Nunes et al., 2014). Coaches tend to reduce physical load parameters the days before a competition as part of a tapering strategy to achieve maximum performance through . This underestimation of the workload over a longer period can lead to maladaptive training, insufficient recovery, increased risk of injury, overtraining, and negative changes in psychophysiological state (Heidari et al., 2018;Kenttä & Hassmén, 1998). It seems that coaches misjudge the accumulating effects of volume and intensity over an entire training's session. Also with respect to physical variability of individuals, objective monitoring of the training sessions and matches is more accurate than subjective appraisals.
Another iTL variable, the session -RPE (sRPE), showed the exact same pattern and a strong interday relationship -similar to PL, which confirms previous studies (Manzi et al., 2010;. MD was the most demanding with the highest sRPE. During the training days, MD-3 had the highest value. Only slight differences were found at MD-4 and MD-2. A significant drop in load was observed on MD-1, which supports the tapering concept of training volume decrease. Our study covered four days leading up to a league game and the game itself. sRPE has shown some associations with changes in training outcomes such as fitness . These associations appear stronger than those with eTL (Impellizzeri et al., 2019), which highlights the importance of internal load quantification. A second finding using sRPE for monitoring iTL is that it is influenced by training volume. These findings correspond to previous studies in basketball (Aoki et al., 2016;Nunes et al., 2014). A possible explanation of higher sRPE in games could be the variability of game intensity, mainly due to the increase in actions requiring changes in direction, accelerations and decelerations, high-speed sprints, and other related specific basketball actions that might lead to a higher mechanical load, which can also be associated with a higher playing time ( . A possible explanation could be that wellness status can be influenced by training factors, like intensification, which can cause psychological disturbances such as fatigue, more muscle soreness, and a worse recovery state (Haddad et al., 2013;Hooper & Mackinnon, 1995). However, basketball players in this study showed very good overall wellness status, with very low DOMS, fatigue, stress, and very good sleep quality (i.e., mean categories' scores around 2). It could be that, in general, this team-training process and players' routines did not represent highly stressful factors. The second suggestion could be a lack of experience with ASRM. That means athletes could try to make a good impression or athletes felt that their coach had some resistance to ASRM so no matter what the ASRM is showing, athletes must perform on the field (Saw et al., 2015). To be more specific, if daily loads are not adjusted according to athletes' ASRM and demands do not follow a planned schedule, athlete compliance may suffer. The present study had some limitations. One of them was the sample size, as only one team was analyzed and there could be possible dependencies on player position. Another limitation is the external load quantification through IMA. There is a lack of information regarding isometric muscle contractions or the physical effort during static position fights and collisions between players, for instance, to map the additional mechanical load in the entire workload. Additionally, our results did not consider contextual factors such as game locations and playing positions, neither in-game technical and tactical performance. Therefore, future studies should assess the fluctuations of weekly training and game load considering these contextual factors.

PRACTICAL APPLICATIONS
Despite these limitations, our study offers some practical applications. Practitioners should consider implementing workload monitoring strategies, taking into account game scheduling during the season. Since several games are played during the season, adequate recovery from the high intensities that games require should be considered. Monitoring can help with good periodization strategies to avoid excessive workload during regular weeks and to prevent non-functional overload and increased risk of injury. It can help to plan for peak workloads and adjust training accordingly. In addition, basketball coaches should monitor player workload in relation to minutes played to receive adequate information on player strain. Training planning should be individualized to avoid exacerbating match loads of players with many minutes played by adding an additional high workload. Successful training load monitoring should occur for two primary reasons: to reduce the risk of injury and to ensure optimal levels of loading and adaptation that result in improved physical and athletic performance.

CONCLUSIONS
Our results suggest, that in a normal week with only one game, it is harder to find the right dosage to prepare for a game. Managing TL in basketball is a complex issue (Capranica & Millard-Stafford, 2011). In game-based sport, it is difficult to design individualized training plans since collective drills are widely used to enhance game-based technical and tactical skills concurrently with fitness components (Dragonea et al., 2018). It may be related to the fact that the requirements of the game are difficult to imitate in training. Basketball players are a specific population characterized by very different anthropometric and physical characteristics. In practice planning, it is important to consider the individual variability between the players. This variability can include psychological factors, individual difference in performance and workload. Furthermore, players differ from each other and show individual game performance profiles (guards, big men, shooters). If this variability is not considered at the individual level, it could have an accumulative effect of workload on match load during a regular week. So, it would be naïve to assume that every player has the same baseline. Therefore, a holistic athlete monitoring strategy can help to provide an appropriate training stimulus in training for players with such diverse characteristics, and it seems reasonable to investigate possible factors that influence player workload. Athlete monitoring should not be seen as limited to either subjective or objective measures. They can both be used to complement each other and help coaches with practice calibration.