Changes in Biomechanical Profile of an Artistic Swimming Duet over a Training Macrocycle: A Case Study

Abstract

This study aimed to monitor the biomechanical development of an artistic swimming duet across a macrocycle through an individualised training approach. Two swimmers (17.5 ± 0.5 years), members of the Los Angeles 2028 National Olympic Project, were assessed in December 2023 (M1) and April 2024 (M2), corresponding to the beginning and the end of the macrocycle. Maximal (Fmax) and mean (Fmean) force in the prone sculling and kick pull action were measured using a 20 s tethered test. Split velocity (vSplit) was assessed in free format based on video recording. Dry-land strength included assessments of internal (IR) and external (ER) shoulder rotation strength of the dominant (D) and non-dominant (ND) limbs, and countermovement jump (CMJ) power. The standard duet choreography was analysed in competition at both time points. Percentage variation (∆%) between swimmers was calculated for M1 vs. M2. Results showed convergence (M1 vs. M2) in Fmean of the sculling (21.6% vs. 9.9%) and kick pull (45.1% vs. 29.1%), accompanied by greater similarity in vSplit (15.9% vs. 15.5%). Further convergence was observed in IR_ND (33.7% vs. 13.9%), ER_D (11.6% vs. 4.4%) and CMJ (7.4% vs. 3.6%). The duet’s competition score increased from 168.9943 to 190.7183 points. It can be concluded that individualised training was useful for the duet to become more homogeneous in in-water strength, in-water kinematics and dryland strength, resulting in improved competitive performance.

1. Introduction

The Artistic Swimming, formerly known as synchronised swimming, is an Olympic aquatic sport that integrates elements of swimming, dance and acrobatic movements performed in synchrony with musical accompaniment. Having individual and team-based disciplines, artistic swimming is among the most physically and artistically demanding sports, requiring the athletes to harmonise in physical strength and technical skill with expressive movement while performing various routines [1]. Routine composition and performance aim to demonstrate technical proficiency, synchronisation, expressiveness and artistic interpretation, all of which are evaluated by judges based on predetermined criteria [2].

Research about artistic swimming has increased in recent years, with investigations largely focused on physiological responses, performance determinants and injury risk [3]. It has been shown that physiological [4] and biomechanical characteristics [5] are central factors to optimise performance in a competition context. There was also growing interest in understanding the potential adaptations elicited by specific artistic swimming training. Training appears to influence several anthropometric and physical characteristics, including improvements in both upper [6] and lower limb’s strength [7]. Longitudinal adaptations also include somatotype changes towards a more mesomorphic body profile, which may directly enhance performance in this sport [8]. Furthermore, training-related improvements in pulmonary diffusion have been reported, though these appear to vary considerably between individuals [9].

Although having great literature to deal with, most of the previous studies focused on studying generic groups of artistic swimmers. The competitive routines can be presented in solo, duet (female and mixed), or team formats (four to eight swimmers), which may require detailed approaches in training. The duet performance plays a critical role in showcasing not only the individual athlete’s skill but also the seamless synchronisation between partners. The morphological similarity, visual harmony and movement uniformity in a duet are essential for achieving aesthetic cohesion, which directly impacts the judges’ evaluation. Additionally, duet routines require a high degree of biomechanical compatibility and mutual understanding between partners, making them a unique and challenging aspect of the sport, where the synergy between athletes can significantly elevate the overall performance quality.

The training load management in this sport must consider the specificity of each apparatus, as they have different demands and training load behaviours [10]. The training of duet partners should target similar body composition and limb proportions to exhibit comparable aquatic balance and motor control, contributing to more cohesive performances. Interestingly, four of the top five duets in the Paris 2024 Olympics were composed of twins, which sustains the importance of searching for similarities. However, the scientific literature on duet performance is limited and restricted to physiological assessment, e.g., [11]. This lack of empirical data presents an opportunity to explore novel strategies, such as individualised training interventions aiming to enhance duet homogeneity from a biomechanical standpoint and optimise performance.

Therefore, this study aimed to monitor the biomechanical development of an artistic swimming duet across a macrocycle through an individualised training approach.

2. Materials and Methods

2.1. Participants

An international-level women’s duet (17.5 ± 0.5 years of age) participated in this study as members of the Portuguese Los Angeles 2028 National Olympic Project. Both athletes have been training together since 2021 and represented the national team at the Junior European Championships in 2023, achieving one of the best results ever for their country at the junior level. Their weekly training routine comprised a total of 25 h including two to three pool swimming sessions with the club’s senior swimming team, two to four strength and conditioning sessions at the gym with the club’s strength and conditioning coach, one online ballet session with the ballet teacher of the National artistic swimming team and five sessions focusing on choreography technique, flexibility and choreography endurance with the club’s artistic swimming coach. The swimmers and their legal guardians (if necessary) were informed of all data collection procedures and provided written informed consent. The study complied with the principles outlined in the Declaration of Helsinki concerning research involving human participants.

2.2. Study Design and Experimental Protocol

This study followed a longitudinal research design during the traditional periodisation of a macrocycle with two time points of data collection: (i) the end of the first macrocycle, corresponding to the end of the first national competition (M1, December); (ii) the end of the second macrocycle, corresponding to the second national competition (M2, April). Each moment of data collection comprised two sessions at the same time of day with 48 h apart, to ensure adequate recovery and minimise circadian variation. Tests were distributed across both sessions with the first session dedicated to anthropometric and in-water strength, and the second session for in-water kinematics and dryland strength testing. All assessments were conducted in a 25 m swimming pool and adjacent dryland areas with a water temperature of 26 °C and humidity of 60%. The swimmers were instructed to avoid intense physical activity and training 24–48 h before the testing sessions. All details regarding sleep, nutrition/hydration, menstrual cycle were also monitored to avoid any bias between pre and post test measurements.

2.3. Anthropometrics

The body mass, height and arm span were measured by the same assessor following the standardised protocols [12] using a portable scale (TANITA, BC-730, TANITA, Kowloon, Hong Kong), a stadiometer (SECA, 242, SECA, Hamburg, Germany) and a measuring tape (Rosscraft, Minneapolis, MN, USA), respectively. All measurements were taken with the swimmer’s barefoot, wearing a swimsuit and cap.

2.4. In-Water Strength

The assessment of in-water strength included the execution of the prone standard sculling and kick-pull actions under tethered swimming conditions. The swimmers remained tethered at 5 m from the wall, connected via a cable to an integrated force assessment system (Swim-Spektro, Talamonti, Italy) with a belt secured around their waist. The system was calibrated as previously described by Cortesi et al. [13]. The load cell was aligned with the swimming plane, forming a 25° angle with the water surface. To mitigate inertial effects, swimmers began each test with 5 s of low-intensity actions, followed by 20 s at maximum effort. One of the assessors used a sound signal to indicate the start and end of the test. The first 2 s of the signal were discarded due to cable tension issues. The prone standard sculling is characterised by horizontal sculling movements performed by the upper limbs without the involvement of the lower limbs, maintaining a position opposite to the direction of propulsion. The kick-pull involves simultaneous movements of the lower limbs and separated movements of the upper limbs, both derived from the breaststroke swimming technique.

The data from both tests were collected using the system’s dedicated software (DB:4, Talamonti Spa, Ascoli Piceno, Italy). Subsequently, a custom routine (v. 0.5) developed in MATLAB (R2023b, MathWorks, Inc., Natick Apple Hill, MA, USA) was used for signal processing using a smooth filter (10 Hz) and an angular correction of 25°, taking into account the angle formed between the system and the water surface. This correction allowed for the accurate determination of the horizontal component of both the peak force (Fmax, N) and the mean force (Fmean, N). Instantaneous force values from the initial 2 s were excluded due to the initial cable tension.

2.5. In-Water Kinematics

The in-water kinematics included the execution of five split movements with just one second of pause. The swimmers positioned themselves four metres from the wall and performed five repetitions at maximal intensity and with the highest possible technical quality. The split action involves initiating movement from a carp (inverted) position with the lower limbs flexed, executing a rapid and explosive movement into the split position and returning to the carp position. All executions were recorded in the sagittal plane using two cameras (GoPro 6, GoPro, San Mateo, CA, USA; 120 Hz) in wide mode, positioned 0.50 m above and 0.26 m below the surface, providing aerial and underwater views, respectively. The clips were then trimmed for kinematic analysis using Kinovea software (v. 0.9.5) through the “stopwatch” function to determine the duration (s) of each split (tSplit in s). Given that one correct execution of the split involves an angular displacement of 360°, the velocity (vSplit, in rad/s) was then calculated based on the ratio between the displacement and the time to perform the split.

2.6. Dryland Strength

The dryland strength was determined using two distinct tests: (i) isometric force test for shoulder internal and external rotation; and (ii) vertical jump test using countermovement action. In the isometric strength test, the swimmers were randomly instructed to sit with their backs against a rigid surface and their lower limbs fully extended. The upper limb (arm and shoulder) was manually stabilised by the evaluator, while the contralateral arm remained steady, aligned with the trunk. The isometric force values were collected using a reliable portable digital dynamometer (microFET^®2, Hoggan Scientific, Salt Lake City, UT, USA) [14]. The curved pad of the dynamometer was positioned against the ulnar and radial styloid processes (medial and lateral surfaces), with a non-elastic strap fixed to a treatment table. Before testing, swimmers performed a warm-up consisting of two submaximal isometric repetitions and one maximal repetition (with 30 s of rest) for both internal rotation (IR) and external rotation (ER), following the protocol of Riemann et al. [15]. The peak isometric force (in N) was assessed via two five-second repetitions for both the dominant (D) and non-dominant (ND) upper limbs, in both IR and ER actions. The mean of the two repetitions was used for further analysis. The limb’s dominance (dominant and non-dominant) was self-reported by the swimmers.

In the countermovement vertical jump (CMJ) test, swimmers first completed a warm-up consisting of five repetitions with 20 s of rest, followed by three maximal jumps, with a two-minute rest interval between attempts. To avoid any contribution from the upper limbs, swimmers were instructed to keep their hands placed on their hips throughout the entire jump. Maximum jump height (in m) during the CMJ was measured using a contact mat (Ergojump Digitime 1000, Digest, Finland), and the mean value of the three jumps was used for further analysis.

2.7. Macrocycle Planning (Individualised Approach)

Between the two assessment time points, the swimmers underwent an 18-week intervention period requiring an individualised approach based on the results of M1. The first stage of this individualised training involved introducing an additional strength training session with fewer repetitions but higher intensity for swimmer 2.

Both athletes followed a macrocycle training programme that included five weekly water-based sessions, each approximately three hours in duration. These sessions comprised sprint swimming, specific artistic swimming skills and choreographic training. Gym-based strength training involved two to four sessions per week lasting 30 to 40 min each. These sessions followed the “Every Two Minutes On the Minute” (ETMOM) method; during even-numbered minutes, a fixed number of repetitions of one exercise were performed, while during odd-numbered minutes, a different exercise was executed, with only 10 s of rest between them. This method aimed to enhance explosive strength, which is a critical component for choreographic performance.

Each strength training session included two 10 min ETMOM blocks, totalling 20 min, followed by a four min Tabata protocol (20 s of work followed by 10 s of rest). The ETMOMs focused primarily on upper body and core work, while the Tabata exercises targeted the lower limbs. This approach was designed based on the strength requirements for the sport and the need for core control to manage body positioning and flotation. So, the Tabata exercises were selected to tone and strengthen the lower limbs without inducing excessive hypertrophy, helping maintain both functional efficiency and aesthetic requirements for Artistic Swimming.

The training programme also included two sessions per week of continuous swimming (two hours each) with the senior squad. The remaining hours were allocated to choreographic work, both in and out of the water. Within this choreography work, further individual adaptations were introduced for swimmer 2, who exhibited greater difficulty producing explosive force during aquatic movements. As such, there was an attempt to complete fewer repetitions of each drill but receive detailed feedback between attempts. This approach aimed to help swimmer 2 understand corrections, improve technique and maximise the effectiveness of each training session rather than just executing mechanically.

Over the course of the macrocycle, each swimmer also followed a tailored set of exercises designed according to their individual needs. Swimmer 1 was found to have difficulty executing a specific rotational movement critical to the pre-defined choreography, not applying equal force through her left adductor, which could be derived from a previous injury. To overcome that issue, a few exercises like lying supine with legs extended and arms along the body, holding a medium-sized medicine ball between the feet were prescribed. Likewise, swimmer 2 demonstrated issues with coordination and the activation of muscular chains necessary for generating explosive force. A separate routine of activation exercises, like gripping a pen with the toes, while simulating lower limb movement on land, was therefore developed to promote more efficient muscle chain recruitment.

Furthermore, it became evident that the two swimmers required different pre-competition activation strategies. Swimmer 1 benefited from a longer warm-up focused on preparing the body for apnea tolerance, whereas swimmer 2 required a more intense warm-up that targeted core, upper body and lower limb activation.

2.8. Performance

To assess the competitive progression, the duet’s standard choreography score was used as a performance indicator at M1 and M2. The routine included two hybrids comprising elements from various difficulty families, five required technical elements, one acrobatic move and a sequence of transitions between elements.

2.9. Statistical Procedures

As this is a case study, inferential statistics were not required. Values for all variables within the different domains were described individually. The percentage variation between M1 and M2 (∆%), as well as between the two swimmers, was calculated for all variables using the formula (Swimmer 1 − Swimmer 2)/Swimmer 1 × 100, in order to express both inter- and intra-individual differences within the duet. The symmetric absolute percentage difference (|Δ%|) was also computed to better reflect “homogeneity”.

3. Results

Figure 1 presents the signal acquisition during the tethered test, illustrating the force-time curves for the kick pull action. At the M1, swimmer 1 exhibited considerably higher values than swimmer 2. The kick pull action of swimmer 2 showed a tendency towards asymmetries between successive peaks, which may indicate a muscular imbalance in one of the upper limbs. At M2, both swimmers get closer in the peak-curves profile, with swimmer 2 showing a more homogeneous profile in force application.

Figure 2 presents an example of signal acquisition during the tethered test, illustrating the force-time curves for the prone standard sculling. Both swimmers were able to improve in-water strength during prone standard sculling from M1 to M2. Specifically, swimmer 2 was able to reach a higher frequency during the sculling in M2 without losing any in-water strength capacity.

Table 1. Intra-individual changes at the different time points of the training macrocycle.

Variables (Units)	Swimmer 1			Swimmer 2
	M1	M2	∆ (%)	M1	M2	∆ (%)
Anthropometrics
BM (kg)	57.80	59.40	2.77	54.80	56.80	3.65
Height (m)	1.66	1.66	0.00	1.66	1.66	0.00
Arm span (m)	1.66	1.68	1.20	1.77	1.77	0.00
In-water strength
Fmax kick pull (N)	656.54	704.67	7.34	396.51	533.64	34.58
Fmean kick pull (N)	187.04	193.88	3.66	102.72	137.54	33.90
Fmax sculling (N)	196.07	205.60	4.86	163.94	153.35	−6.46
Fmean sculling (N)	84.18	83.71	−0.56	65.95	75.39	14.31
In-water kinematics
tSplit (s)	1.49	1.22	−18.12	1.77	1.45	−18.08
vSplit (rad/s)	4.22	5.15	22.14	3.55	4.35	22.69
Dryland strength
IRD (N)	97.85	118.7	21.31	74.70	90.70	21.42
IRND (N)	109.40	98.95	−9.55	72.50	85.15	17.45
ERD (N)	85.50	81.85	−4.27	75.60	78.25	3.51
ERND (N)	61.60	62.45	1.38	62.90	72.00	14.47
CMJ (m)	0.27	0.28	3.70	0.25	0.27	8.00

M1, testing moment 1 (December 2023); M2, testing moment 2 (April 2024); BM, body mass; Fmax, maximum force; Fmean, mean force; tSplit, time of the split action; vSplit, velocity of the split action; IR, internal rotation; ER, external rotation; D, dominant; ND, nondominant; CMJ, countermovement jump.

presents the individual values by domain and the corresponding intra-individual variations (i.e., each swimmer change) over the training macrocycle. Both swimmers increased body mass, which was more pronounced in swimmer 2. The remaining anthropometric characteristics remained relatively stable for both swimmers. In the biomechanical domain, there were mixed changes in Fmax and Fmean during the prone standard sculling and kick pull actions. The sculling Fmax decreased for swimmer 2 and Fmean decreased slightly for swimmer 1, while the remaining in-water strength variables increased. Additionally, an increase in vSplit was also observed. In the dryland strength, improvements were also observed in isometric contraction and jump power, with these gains being more marked in swimmer 2.

Table 2. Inter-individual changes at the different time points of the training macrocycle.

Variables (Units)	M1			M2
	Swimmer 1	Swimmer 2	∆ (%)	Swimmer 1	Swimmer 2	∆ (%)	|Δ%| M1 vs. M2
Anthropometrics
BM (kg)	57.80	54.80	5.19	59.40	56.80	4.38	0.81
Height (m)	1.66	1.66	0.00	1.66	1.66	0.00	0.00
Arm Span (m)	1.66	1.77	−6.63	1.68	1.77	−5.36	1.27
In-water strength
Fmax kick pull (N)	656.54	396.51	39.60	704.67	533.64	24.27	15.33
Fmean kick pull (N)	187.04	102.72	45.08	193.88	137.54	29.06	16.02
Fmax sculling (N)	196.07	163.94	16.39	205.60	153.35	25.41	−9.02
Fmean sculling (N)	84.18	65.95	21.66	83.71	75.39	9.94	11.72
In-water kinematics
tSplit (s)	1.49	1.77	−18.79	1.22	1.45	−18.85	−0.06
vSplit (rad/s)	4.22	3.55	15.91	5.15	4.35	15.53	0.38
Dryland strength
IRD (N)	97.85	74.70	23.66	118.70	90.70	23.59	0.07
IRND (N)	109.4	72.50	33.73	98.95	85.15	13.95	19.78
ERD (N)	85.50	75.60	11.58	81.85	78.25	4.40	7.18
ERND (N)	61.60	62.90	−2.11	62.45	72.00	−15.29	−13.18
CMJ (m)	0.27	0.25	7.41	0.28	0.27	3.57	3.84

M1, testing moment 1 (December 2023); M2, testing moment 2 (April 2024); BM, body mass; Fmax, maximum force; Fmean, mean force; tSplit, time of the split action; vSplit, velocity of the split action; IR, internal rotation; ER, external rotation; D, dominant; ND, nondominant; CMJ, countermovement jump.

presents the individual values by domain and the respective inter-individual variations (i.e., if the pair converge) at each testing moment. In the anthropometric domain, a closer similarity in BM between swimmers was observed, primarily due to the increase in swimmer 2. The remaining anthropometric characteristics remained relatively stable throughout the macrocycle. In the biomechanical data the convergence was not uniform across variables. While some differences shrink (e.g., kick-pull Fmax; sculling Fmean), others increased (e.g., sculling Fmax; ER_ND) and some remain nearly unchanged (e.g., vSplit). Regarding dryland strength, the results indicated convergence between both swimmers in IR and ER (except ER_ND) and CMJ.

The choreography score in M1 and M2 was 168.9943 and 190.7183 points, respectively.

4. Discussion

This study aimed to characterise and monitor an artistic swimming duet from a biomechanical perspective throughout a full training macrocycle. The swimmers were able to improve anthropometrics, in-water strength, in-water kinematics and dryland strength over the 18 weeks of training. The implementation of an individualised training allowed the duet to become more homogeneous, with a positive impact on choreography score in competition.

In the anthropometric domain, the increases in BM were more pronounced in swimmer 2, while the other anthropometric characteristics remained relatively stable for both. This increase in BM follows a literature trend, where the additional strength training sessions may lead to notable changes in body composition, mostly by lean mass gain, e.g., [16]. That anthropometric transformation makes the swimmers more similar in height, wingspan and body mass, which in some way facilitated the synchronisation of movements and consequently promoted an improvement in their performance. A more harmonious anthropometric measurements contribute not only to aesthetics, but also to the efficiency and effectiveness of routines in Artistic Swimming [17]. This bodily similarity is desirable and guarantees visual symmetry, an aspect highly valued by judges in competition.

The biomechanical data showed mixed changes in Fmax and Fmean during the prone standard sculling and kick pull actions. The sculling Fmax decreased for swimmer 2 and Fmean decreased slightly for swimmer 1, while the remaining in-water strength variables increased. Despite that, the absolute percentage values showed that swimmers converged in most of the in-water strength variables, except for the Fmean in sculling. This was not alarming as the analysis of the force-time curves showed that the two swimmers exhibited closer force scores at M2, indicating that both finished the macrocycle sculling more precisely and applied similar force levels compared to M1. This aspect translated into a more coordinated and efficient action with potential impact on how propulsive actions were performed.

The prone standard sculling action has a propulsion phase where the swimmer applies maximum force to generate forward movement, followed by a recovery phase where the swimmer repositions the upper limbs and prepares for the next movement by applying less force. The force amplitude became more consistent at M2 with the values near those reported in the literature [18], possibly indicating technical refinement and a better use of force application for propulsion. This increase in uniformity while applying in-water forces suggests the effectiveness of the individualised training, allowing the swimmers to achieve a higher level of synchronisation between limbs and with greater efficiency. The level of coherence between propulsion and symmetry in sculling should be of utmost importance due to its effects on stabilising body position [19].

A similar trend was observed for the kick pull actions, with greater force values being applied at the end of the macrocycle by both swimmers. Knowing that the kick pull action is heavily centred on the explosive power of the lower limbs, critical attention was needed on this throughout the training weeks. Indeed, the way the lower limbs act and translate into the water can have repercussions on performance indicators [20]. From a duet perspective, both swimmers increased lower limbs’ power, which resulted in gains in height and velocity during choreography execution, positively impacting the choreography core in competition.

An increase in vSplit was notable after the macrocycle, but the swimmers’ convergence in this variable was less marked than in others previously mentioned. In Artistic Swimming, the relationship between applied force and movement velocity is fundamental, particularly in movements requiring rapid changes in direction and fine coordination [7]. The greater in-water force reached by the swimmers in M2, even in elements like sculling or kick pull, enabled the swimmers to apply higher in-water muscle tension, which was crucial for performing the split more powerfully and explosively and stabilising the body in the water. Despite the increased velocity, the split action was not compromised in precision, due to the rigorous technical work that accompanied velocity-based work.

Improvements were also noted in isometric contraction and jump power, with a more pronounced change in swimmer 2. Strength plays a crucial role in the swimmers’ ability to perform figures and routines [21]. A large proportion of technical and free duet routines are performed in the inverted vertical position (41.6% and 45.4%, respectively), followed by the vertical (29.2% and 28.4%) and horizontal (29.2% and 26.2%) positions [22]. So, high strength levels are essential for support, propulsion and body thrust in these positions [6], as well as for transitions and rotational movements [23].

The individualised training of the duet also appears to lead to muscular imbalances between IR and ER. Nevertheless, performance was better at M2 with a high choreography score reached in competition. A recent study [6] found a positive correlation between strength (right and left) and the Barracuda element, while only the right-hand strength correlated positively with the vertical position element. This may suggest a lack of interdependence between imbalances or unilateral strength and competitive performance. This means that there could be elements in artistic swimming requiring consolidated bilateral symmetry, while other elements are not dependent on bilateral force. However, this cannot be concluded with the present data since no IR/ER ratios or element-specific analyses were conducted, and any element-level relationships require targeted testing, which may be performed in future studies.

Jumping power has been reported as a determinant skill for performance in various sports, such as gymnastics [24]. Literature in Artistic Swimming is limited, and previous studies have primarily focused on the kinematics of the eggbeater kick, e.g., [25]. It can be argued that the lower limbs’ power could be indispensable for performing elements such as kick pull, or boosts. Although the improvements in lower limb power for swimmers 1 and 2 were approximately 4% and 8%, at M2, the duet narrowed their performance gap (from 7% to 4%). This seems to demonstrate that the individualised training focused on strengthening weaknesses and maximising the lower limb’s power in each swimmer was important to achieve a more homogeneous duet in this capacity.

Few limitations can be addressed in this study: (i) control group or comparison with other duets was not made to clearly attribute the observed improvements to the individualised training itself; (ii) body mass measurements were not dissected in lean or fat mass to better impairing to establish more detailed links with other domains.

5. Conclusions

The biomechanical characteristics of an artistic swimming duet improved throughout a full training macrocycle. The implementation of an individualised training regime allowed the duet to converge in most of the anthropometrics, in-water strength, in-water kinematics and dryland strength variables, being accompanied by an improvement in the choreography score in a competition context. The findings of this study offer several practical recommendations for coaches working with artistic swimming duets. Beyond confirming the importance of systematic monitoring, they highlight concrete strategies that can be incorporated into training routines. Coaches may include dryland and in-water exercises focused on trunk position and limb symmetry, ensuring both athletes develop a shared body orientation. Simple mirror-based dryland work or partner-assisted corrections in the water can enhance awareness of relative positioning. The feedback through video analysis allows athletes to visualise small differences in timing and alignment. Reviewing short training segments with immediate feedback can accelerate the correction process. By making individualised technical reports, coaches can also identify biomechanical weaknesses of each athlete (e.g., differences in leg extension or trunk control) and design tailored corrective exercises.