The Impact of Aquatic Training on Athletic Performance and Physical Recovery in Competitive Athletes: A Meta-Analysis

Abstract

Zheng, S, Wang, S, Cao, Q, and Li, H. The impact of aquatic training on athletic performance and physical recovery in competitive athletes: A meta-analysis. J Strength Cond Res 40(6): 723–737, 2026—To evaluate the effects of aquatic training on performance and physical recovery in competitive athletes, particularly its impact on explosive power, speed, agility, strength, aerobic capacity, and balance. This study used a meta-analysis approach to search relevant literature in the Web of Science and Scopus databases since their inception. The selected studies were limited to English-language articles that met predefined inclusion criteria. All included studies were randomized controlled trials or controlled experiments, with subjects being competitive athletes with a systematic training background. Standardized mean differences (SMDs) were used for data analysis, and statistical analysis was conducted using R (version 4.3.0). A total of 13 studies were included in the analysis, with 392 subjects. Aquatic training significantly improved athletes' explosive power (SMD = 0.79, 95% confidence interval [CI] = 0.0.50 to 1.07, p < 0.001), speed (SMD = 1.14, 95% CI = 0.89 to 1.38, p < 0.001), and agility (SMD = 0.47, 95% CI = 0.0.15 to 0.78, p = 0.004). However, the improvement in strength (SMD = 0.32, 95% CI = −0.32 to 0.96, p = 0.321) and balance (SMD = 0.42, 95% CI = −0.35 to 1.19, p = 0.282) was not statistically significant, and there was high heterogeneity in the studies on aerobic capacity (SMD = 1.92, 95% CI = −1.47 to 5.32, p = 0.267). In addition, aquatic training showed a modest yet positive effect on physical recovery, including improvements related to injury rehabilitation and fatigue relief (SMD = 0.38, 95% CI = 0.12 to 0.63, p = 0.004). As an effective supplementary training method, aquatic training significantly enhances athletic performance, especially in explosive power, speed, and agility. It has beneficial effects on injury recovery and post-training fatigue reduction. Although the effects on strength and aerobic capacity require further validation, aquatic training has proven to be an important tool in sports training and rehabilitation with promising application potential.

Introduction

Aquatic training is a multidimensional exercise methodology that leverages the unique properties of water, including buoyancy and hydrostatic resistance, through systematically designed movements to enhance sports performance and facilitate rehabilitation. This approach has gained widespread application in sports rehabilitation, with its beneficial effects on physical conditioning and physiological recovery being well-documented. From a biomechanical perspective, water buoyancy significantly reduces joint loading and minimizes injury risks. (1); The hydrodynamic resistance provides dynamic loading to the musculoskeletal system, effectively enhancing strength, endurance, and joint stability—particularly in lower-extremity muscle groups. (1,11). There is research indicates that the buoyancy in aquatic plyometric training reduces landing impact forces and shortens the eccentric to concentric transition time, thereby enhancing muscle power during jumping (24). Consistent with these mechanisms, a review reported that deep-water running largely preserves land-based aerobic capacity and running performance, aquatic high-intensity interval training maintains or modestly improves aerobic and endurance outcomes, and aquatic plyometrics elicit explosive power and speed gains comparable with or exceeding those of land training (15). Similarly, a meta-analysis found that aquatic and land-based programs yield similar overall strength gains in healthy young and older adults; in older individuals, improvements in grip and lower-limb strength are generally modest and seem more pronounced with higher movement velocities and added external resistance, while isometric gains may be negligible in some subgroups (30).

Current research on the application mechanisms of aquatic training in rehabilitation is relatively comprehensive (19,39,43): In the field of orthopedics, it effectively increases joint range of motion and alleviates chronic pain, while using hydrostatic pressure to accelerate metabolic regulation and postexercise recovery (20). From a recovery standpoint, the thermoregulatory properties of the consistent water temperature facilitate muscle relaxation and accelerate tissue repair processes (11); The aquatic environment stimulates endogenous endorphin release, exerting analgesic effects that enhance athletes' psychological well-being. Its therapeutic benefits include helping ALS patients maintain mobility and slow muscle atrophy (26), and improving mobility and balance in stroke patients with less reliance on assistive devices (38).

Current research on aquatic training for improving physical function primarily focuses on rehabilitation patients (19), or on adolescents and general populations (21). Few studies have focused on the potential of aquatic training to enhance the athletic performance of land-based sport athletes. Owing to their distinct differences from nonathletes in skill acquisition and motor transfer, professional athletes exhibit highly developed executive functions and neurocognitive performance, demonstrating significantly superior rapid learning capabilities in complex dynamic environments compared with the general population (10). Their physiological adaptation rate to training is also accelerated, potentially resulting in different effect thresholds for aquatic training compared with nonathletes (4).

However, several questions urgently need to be addressed: Can aquatic exercise serve as an effective supplementary approach for land-based athletes to enhance their physical fitness? Which components of physical fitness can be developed more efficiently through water-based training than by land-based methods? And under the additional resistance and loads posed by water, are athletes more susceptible to injury? These issues require further investigation.

Therefore, this meta-analysis focuses on comprehensively synthesizing the effects of aquatic training on the athletic performance of land-based athletes, including its influence on power, speed, agility, strength, aerobic capacity, and balance, while also exploring the unique role of aquatic training for land-based athletes. Based on the aforementioned background, this study intends to conduct a literature search on aquatic training, restricting the intervention subjects to land-based athletes, and evaluate the intervention effects of aquatic training using meta-analysis. The research aimed to systematically assess the potential benefits of aquatic training as a supplementary training modality for land-based sports in performance transfer and adaptation. It further seeks to comprehensively analyze the effects of aquatic training on the athletic performance of land-based athletes and explore its role in fatigue recovery and injury rehabilitation for athletes, thereby providing theoretical foundations and practical paradigms for optimizing training periodization.

Methods

Experimental Approach to the Problem

This systematic review and meta-analysis adhered to the guidelines outlined in the PRISMA 2020 statement (27). The study has been registered with the PROSPERO website under Registration Number: CRD420251022453.

Subjects

The study population comprised participants reported in the studies included in this systematic review and meta-analysis (13 intervention studies, 392 participants in total). Eligible studies involved land-based athletes. As this work is a systematic review and meta-analysis of previously published data, we did not recruit or enroll individual participants and did not collect or access any individual participant data;

Procedures

Literature search

The literature search was conducted in 2 major databases: Web of Science and Scopus. Only peer-reviewed journal articles written in English were included. The search spanned from each database's inception until February 10, 2025.

Search Strategy

To ensure the comprehensiveness of the literature search, the study classified keywords into 3 categories: water environment, athletes or competition performance, and training or exercise, which were then connected using Boolean operators for retrieval. Keywords related to water environment include: “aquatic,” “water-based,” “pool-based,” “hydrodynamic resistance,” “aqua fitness,” “underwater.” Keywords related to training include: “exercise*,” “train*,” “workout*,” “physical activity,” “conditioning.” Keywords pertaining to athlete categories include: “athlete*,” “sports performance,” “elite sports,” “athletic performance,” “training adaptation.” The search strategy uses “OR” operators to connect keywords within each of the 3 major categories, and “AND” operators to link keywords between these 3 major categories.

The initial search yielded 1,564 English-language articles. After removing duplicates using NoteExpress reference management software, a total of 1,520 unique articles remained for subsequent analysis.

Inclusion Criteria

(a) Study Design: Included literature must be randomized controlled trials (RCTs) or other controlled studies. The experimental group must follow a systematic aquatic-based training regimen (e.g., progressive aquatic resistance training, aquatic plyometric training, or aquatic aerobic endurance training). (b) Subject Characteristics: Subjects must be competitive athletes with a systematic training background; no sex restrictions apply. (c) Data Reporting Requirements: Studies must fully report descriptive statistics (mean ± SD), within-group variance parameters (e.g., standard error), or core data for effect size calculations (e.g., 95% confidence intervals [CIs] of means convertible to standardized mean differences (SMDs) or raw data).

Exclusion Criteria

(a) Literature Type Exclusion: Secondary literature, conference abstracts, brief communications, and incomplete studies excluded. (b) Content Deviation and Intervention Heterogeneity: Studies with major protocol-control inconsistencies (e.g., unstructured aquatic interventions) excluded. (c) Study Design Limitation: Animal model or observational studies excluded. (d) Training Duration Threshold: Interventions lasting <4 weeks (insufficient for physiological adaptation) excluded. (e) Intervention Definition Restriction: Nonexercise aquatic therapies (e.g., contrast water immersion) excluded. (f) Study Topic Relevance: Shoulder injury studies in competitive swimmers excluded to maintain focus and reduce heterogeneity.

Literature Quality Assessment

Quality Assessment of Included Literature

This meta-analysis used the Cochrane RoB 2.0 tool to assess the quality of included RCTs or other controlled study designs. The evaluation was conducted across 7 key dimensions: (a) Random sequence generation, (b) Allocation concealment, (c) Blinding of subjects and personnel, (d) Blinding of outcome assessors, (e) Incomplete outcome data, (f) Selective reporting, (g) Risk of other biases. If a study contained multiple eligible intervention groups, the one most closely aligned with the predefined inclusion criteria was selected for analysis.

Evaluation Process

The evaluation process was independently conducted by the first and second authors. Any discrepancies were resolved through arbitration by the third author. The scoring criteria were as follows: each of the 7 dimensions contributed 1 point, with studies scoring below 3 points classified as high risk of bias, 3–4 points as moderate risk of bias, and 5 points or higher as low risk of bias.

Indicator Extraction

In this study, similar indicators were consolidated into a single composite measure. Intergroup comparisons were performed by calculating relevant effect sizes. Specific indicators are summarized in Table 1.

Table 1.

Indicator classification table.

Categories	Specific indicators
Explosive power	CMJ, SLJ, SJ, MKPT
Speed	10 mS, 30 mS, 50 mS
Agility	T-T, 5–0–5 T, 4 × 10 m SR
Muscle strength	1RM-LP, LDC, IKT
Aerobic	B-T, V̇o2max
Balance	YBT, SEBT
Physical function and recovery	FAI, VAS, IKDC, CK

Selection and Data Collection Process

The first author and the second author independently conducted the literature screening process. After determining the articles to be included, they thoroughly reviewed the full texts. For studies lacking complete data on the corresponding indicators, the first author sent e-mails to the original authors requesting the data; those that received no response were excluded. In cases of disagreement, the 2 authors discussed and sought the opinion of the third author. On finalizing the included articles, both authors extracted the basic information and data using a standardized Excel template. On completion of the process, the 2 authors cross-checked each other's work.

Statistical Analysis

The statistical analysis was performed using the “meta” package in R software (version 4.3.0). For continuous variables, SMDs with 95% CIs were calculated. Hedges' g correction was applied to eliminate bias when substantial sample size disparities existed between studies. Between-study heterogeneity was assessed using the I² statistic and Q test. A random-effects model was used when I² > 50% or the Q test yielded p < 0.1; otherwise, a fixed-effects model was used. Sensitivity analysis was conducted by sequentially excluding individual studies to evaluate result robustness. Publication bias was assessed using funnel plots and Egger's test, with statistical significance set at p < 0.1. For interpreting the magnitude of SMD statistics throughout the article, Cohen's conventions were applied: SMD <0.2 was considered trivial, SMD 0.2–0.5 as small, SMD 0.5–0.8 as moderate, and SMD >0.8 as large effect sizes. These thresholds provided standardized interpretation of clinical and practical significance across all outcome measures.

Results

Literature Screening Results

In this study, 1,564 articles were retrieved from the Web of Science and Scopus databases. After removing duplicates, 1,520 articles remained. After keyword-based exclusion through title screening, 622 articles were preliminarily selected. After further review of titles and abstracts, 67 studies were identified for full-text assessment, with 13 articles ultimately included. The literature screening process is illustrated in Figure 1.

Figure 1.

Literature screening flowchart.

Quality Assessment of Included Studies

This study included 13 publications, among which 2 scored 5 points in the quality assessment (indicating low risk of bias), while the remaining studies were rated as moderate risk. Figure 2 presents the methodological quality assessment results, with “+” denoting fulfillment, “-” indicating nonfulfillment, and “?” representing unclear reporting of specific criteria. Figure 3 illustrates the proportion of studies meeting each assessment item.

Figure 2.

Visualization of methodological quality assessment.

Figure 3.

Proportion statistics of methodological assessment items.

All studies used randomized allocation. Two concealed allocation until intervention, maintaining blinding. Selective reporting and other biases showed 100% low risk. Random sequence generation and complete outcome data also neared 100% low risk.

Overall, the included studies exhibited satisfactory quality. Notably, blinding of subjects was inherently impractical in exercise intervention trials due to discernible differences in training environments, protocols, and pacing—a limitation exerting minimal impact on outcome validity.

Characteristics of Included Studies

Table 2 summarizes the 13 studies included in the meta-analysis, all of which were RCTs or other controlled studies with a total sample size of 392 subjects. According to the included literature, all subjects were systematically trained athletes, predominantly from team sports such as football and volleyball. The cohort comprised 315 men and 77 women. The experimental group included 198 subjects, while the control group had 194. All interventions in the experimental group involved different water-based training protocols.

Table 2.

Characteristics of included studies.^*

Study	N (exp. Group)	N (ctrl. Group)	Gender (M/F)	Age (y)	Pool depth	Training protocol	Training duration	Training frequency	Outcome measures
Yu (35)	10	10	M	24 ± 2	0.85 m (waist-deep)	Experimental group: Aquatic plyometric training conducted in 85 cm water depth, incorporating 40 cm box depth jumps, squat jumps, and tuck jumps Control group: Land-based plyometric training performed on a wooden floor, using identical exercises. (Healthy population)	6 W	2	Explosive power:CMJ Speed:10 mS Agility:T-T Muscle strength:1RM-LP
Naderi (19)	21 M11 F10	20 M10 F10	M F	31 ± 9	Unknown	Experimental group: Aquatic balance training, including single-leg stance, cross-leg sway, single-leg squat, and heel raises Control group: Land-based balance training with identical exercises performed while wearing standard athletic shoes. (Injured population, chronic ankle instability)	8 W	3	Physical function and recovery:FAI balance:YBT
Sadaak (26)	15 M7 F8	15 M6 F9	M F	24 ± 4	Unknown	Experimental group: Aquatic intervention comprising warm-up, stretching, resistance training, balance exercises, and functional training Control group: Land-based physical therapy involving structured therapeutic exercises, manual therapy, resistance training, postural re-education, balance training, and progressive functional training. (Injured population, acute ankle sprain)	6 W	Unknown	Agility:T-T Physical function and recovery: VAS Balance: YBT
Ghosh (33)	12	12	M	15 ± 2	Waist-deep	Experimental group: Aquatic plyometric training including single-leg hops and squat jumps (water depth at the waist level), with progressive intensity and volume Control group: Standard moderate-distance running program. (Healthy population)	14 W	3	Explosive power:SLJ speed:50 mS
Isnaini (14)	20 M11 F9	20 M8 F12	M F	20 ± 1	University swimming pool	Experimental group: Aquatic tabata training—a high-intensity interval training (HIIT) protocol (20 s exercise/10 s rest), leveraging water's buoyancy and resistance to enhance strength, flexibility, speed, and endurance Control group: Conventional aerobic and strength training, focusing on traditional physical conditioning. (Healthy population)	4 W	3	Explosive power:SJ Speed:10 mS Agility:4 × 10 m SR Muscle strength:LDC Aerobic:B-T
Cai (1)	10	10	M	20 ± 22	University swimming pool	Experimental group: 10-wk aquatic resistance plyometrics involving progressive squat jumps, vertical jumps, and lateral double-leg hops Control group: Volleyball team skill training. (Healthy population)	10 W	2	Physical function and recovery VAS, CK
Biswas (3)	12	12	M	15 ± 2	Waist-deep	Experimental group: Aquatic plyometric training (water depth at the waist level) including squat jumps, single-leg hops, double-leg bounding, split jumps, tuck jumps, box jumps, and depth jumps Control group: Identical plyometric exercises performed on grass surfaces. (Healthy population)	14 W	3	Explosive power:MKPT
Chen (6)	30	30	M	29 ± 4	Unknown	Experimental group: Aquatic aerobic exercises including continuous running, cross-country ski jumps, jumping jacks, and front kicks (8 exercises total) Control group: Stationary cycling at 60 RPM. (injured population, postmeniscal allograft transplantation)	12 W	3	Muscle strength:IKT Physical function and recovery:IKDC balance:YBT
Hajouj (12)	19	19	M	24 ± 4	Waist-to-sternum deep	Experimental group: Conventional rehabilitation (strength training, gait training, neuromuscular training) combined with aquatic proprioceptive training (single-leg stance, foam roll balance, ball throws) Control group: Conventional rehabilitation only (strength training, gait training, neuromuscular training) Injured population, Post-ACL reconstruction	6 W	3	Physical function and recovery:VAS, IKDC
Chomani (8)	20	20	M	16 ± 1	Unknown	Experimental group: Aquatic plyometric training (jumps, single-leg hops, depth jumps, lateral bounds, box jumps) with progressive intensity Control group: Identical training protocol performed on land. (Healthy population)	8 W	3	Explosive power:CMJ speed:30 mS
Sporri (29)	11	9	M	22 ± 3	1.2 m	Experimental group: Aquatic plyometric training (jumping, single/double-leg hops, etc.) Control group: Maintained regular ball sports training (no additional intervention) Healthy population	8 W	3	Explosive power:CMJ agility:5–0–5 T speed:10 mS
Martel (22)	10	9	F	15 ± 1	1.22 m	Experimental group: Aquatic plyometric training (water-based jumps, power leaps, squat jumps, depth jumps) Control group: Regular ball-sports training (no additional intervention) Healthy population	6 W	2	Explosive power:CMJ Muscle strength:IKT
Wilberrandalll (23)	8	8	M	28 ± 6	Deep water zone	Experimental group: Deep-water running (with buoyancy belt, alternating 30-min high-intensity sessions [90–100% V̇o2max] and 60-min moderate-intensity sessions [70–75% V̇o2max]) Control group: Land-based running (identical intensity and duration protocols) Healthy population	6 W	5	Aerobic:V̇o2max

^*CMJ = countermovement jump; SLJ = standing long jump; SJ = squat jump; MKPT = Margaria-Kalamen power test; 10 mS = 10-meter sprint; 30 mS = 30-meter sprint; 50 mS = 50-meter sprint; T-T = T-test agility; 5–0–5 T = 5–0–5 agility test; 4 × 10 m SR = 4 × 10-meter shuttle run; 1RM-LP = 1 maximum leg press; LDC = leg dynamometer; IKT = isokinetic strength test; B-T = beep test; V̇o₂max = maximal oxygen uptake; YBT = Y-balance test; SEBT = star excursion balance test; FAI = functional ankle instability test; VAS = visual analog scale; IKDC = International Knee Documentation Committee subjective knee evaluation form; CK = blood creatine kinase concentration.

Pooled Analysis Results

Although the included studies all focused on the same 7 evaluation dimensions, they used numerous divergent metrics with varying units of measurement. To facilitate cross-study comparisons, this study used SMDs and 95% CIs. For each unified dimension, the most widely adopted indicator was selected—for example, the countermovement jump (CMJ) was prioritized in explosive power assessments when available. If most indicators within a given dimension relied on subjective scales, multiple metrics were selectively incorporated to enhance accuracy, and a combined SMD was computed.

Meta-Analysis of Effect Sizes for Explosive Power and Publication Bias Assessment

The study included 7 articles for meta-analysis. Heterogeneity assessment revealed an I² of 0.0% (p = 0.730), indicating high consistency across studies with no significant bias or interference in effect sizes.

A fixed-effects model yielded an overall SMD of 0.79 (95% CI: [0.50–1.07], p < 0.001), demonstrated a statistically significant moderate-to-large effect size. These results indicate that the interventions consistently produced moderate effects across studies. Figure 4 summarizes the effect sizes and heterogeneity analysis, while Figure 5 shows no evident publication bias, despite 2 studies with nonsignificant outcomes—the remaining studies exhibited robust effect magnitudes.

Figure 4.

Forest plot of the meta-analysis on aquatic training effects for explosive power.

Figure 5.

Funnel plot of the meta-analysis on aquatic training effects for explosive power.

Meta-Analysis of Speed Effect Size and Publication Bias Assessment

The meta-analysis included 5 studies on speed performance. Figure 6 presents the effect sizes and heterogeneity analysis of the included studies. Figure 7 provides the publication bias assessment, where the funnel plot exhibited near-symmetrical distribution, indicating no significant publication bias. Heterogeneity analysis revealed an I² value of 19.7%, along with a Q value of 4.98 (df = 4, p = 0.289), suggesting low between-study variance. Egger's regression test for potential bias yielded t = −2.13 (p = 0.12), confirming no significant publication bias (p > 0.05).

Figure 6.

Forest plot of the meta-analysis on aquatic training effects on speed performance.

Figure 7.

Funnel plot of the meta-analysis on aquatic training effects on speed performance.

Given the low heterogeneity, a fixed-effects model was used. The pooled effect size was SMD = 1.14 (95% CI: 0.89–1.38, Z = 9.06, p < 0.0001). The pooled SMD reached statistical significance, indicating a small-to-moderate positive effect of the intervention on athletes' agility.

Meta-Analysis of Agility Effect Size and Publication Bias Assessment

Four studies were included in the meta-analysis. Figure 8 presents the effect size estimates. Heterogeneity testing indicated explainable variation across studies (I² = 35.0%). The nonsignificant Q-test (Q = 4.61, p = 0.202) suggested low between-study heterogeneity, justifying the use of a fixed-effects model. Regarding publication bias risk—a common concern in small meta-analyses—Egger's regression test showed no evidence of systematic bias (bias coefficient = 2.56, 95% CI crossing zero; t = 1.25, SE = 2.04, p = 0.3368). Symmetric funnel plot morphology (Figure 9) further supported this conclusion. The pooled SMD (SMD = 0.47, 95% CI: 0.15–0.78, Z = 2.904, p = 0.004) reached statistical significance, indicating a small-to-moderate positive effect of the intervention on athletes' agility.

Figure 8.

Forest plot of the meta-analysis on aquatic training effects for agility.

Figure 9.

Funnel plot of the meta-analysis on aquatic training effects for agility.

Meta-Analysis of Effect Size on Strength Level and Publication Bias Assessment

The meta-analysis on strength level included 4 clinical trials for effect size synthesis. Heterogeneity testing revealed moderate-to-high between-study heterogeneity (I² = 79.1%), suggesting substantial variability likely due to diverse outcome measures or potential confounding factors. The significant Q-test result (Q = 14.38, p = 0.0024) further confirmed this heterogeneity, justifying the use of a random-effects model for pooling.

Three-quarters of included studies showed nonpositive training effects. The SMD was 0.32 (95% CI: −0.32 to 0.96), which representing a small effect size that was not statistically significant (Z = 0.992, p = 0.321), indicating inconclusive evidence for aquatic training benefits on strength. The wide prediction interval (−0.96 to 1.60) further reflected uncertainty and heterogeneity, underscoring limited result stability (Figure 10).

Figure 10.

Forest plot of the meta-analysis on aquatic training effects for muscle strength.

For publication bias assessment, Figure 11 presented funnel plot symmetry, although 2 studies fell outside the funnel margins. Egger's regression test showed a bias coefficient of 3.0189 (SE = 2.1260), with nonsignificant results (t = 1.42, p = 0.2915).

Figure 11.

Funnel plot of the meta-analysis on aquatic training effects for muscle strength.

While Egger's test did not indicate significant bias, the limited number of included studies (n = 4) reduced statistical power, weakening conclusions on bias. Key limitations included (25): insufficient study counts reducing model stability and statistical power (1); high heterogeneity suggesting unaccounted confounders.

Meta-Analysis of Aerobic Capacity Effect Sizes and Publication Bias Assessment

Two studies were included to evaluate consistent training effects of aquatic exercise on aerobic capacity. Significant heterogeneity was observed (I² = 75.4%, Q = 4.07, p = 0.0436) (Figure 12), indicating substantial variability in study designs, subjects characteristics, or methodologies that precluded direct pooling of results. A random-effects model was applied due to heterogeneity. The analysis demonstrated a statistically significant improvement in aerobic capacity (SMD = 1.92, 95% CI = [−1.47 to 5.32], Z = 5.96, p < 0.001), supporting the beneficial effect of aquatic training. Publication bias was assessed through funnel plot symmetry and Egger's test. Despite apparent symmetry between the 2 studies, the small number of included studies (n = 2) precluded meaningful bias evaluation (Figure 13).

Figure 12.

Forest plot of the meta-analysis on aquatic training effects for aerobic capacity.

Figure 13.

Funnel plot of the meta-analysis on aquatic training effects for aerobic capacity.

Meta-Analysis of Effect Sizes for Balance Capacity and Publication Bias Assessment

The meta-analysis on balance ability showed substantial heterogeneity (I² = 79.5%, Q = 9.76, p = 0.0076), warranting a random-effects model. The results of the pooled analysis indicated no significant differences(SMD = 0.42 95% CI [−0.35 to 1.19]; z = 1.076, p = 0.282). Besides, Naderi's 2025 study reported a negative effect (SMD = −0.360), substantially influencing the overall estimate and appearing as an outlier in the funnel plot (Figure 14 and 15).

Figure 14.

Forest plot of the meta-analysis on aquatic-based training effects for balance ability.

Figure 15.

Funnel plot of the meta-analysis on aquatic training effects on balance ability.

Owing to the small sample size and high heterogeneity, the meta-analysis results should be considered as a reference only. Retaining this analysis helps explore the sources of the starkly contrasting results and understand the underlying mechanisms.

Effect Sizes and Publication Bias in Meta-Analysis of Aquatic Training for Physical Function and Recovery

Five studies were included in the meta-analysis. As shown in Figure 16, heterogeneity was low (I² = 26.2%, Q = 5.42, p = 0.2469), indicating consistency across studies despite minor variations. A fixed-effects model was therefore applied. Publication bias was assessed using a funnel plot (Figure 17), showing slight asymmetry. Egger's test further confirmed no significant bias (p = 0.0646). The overall meta-analysis revealed statistically significant positive effects of aquatic training on injury recovery (SMD = 0.38, 95% CI = [0.12–0.63], p = 0.004).

Figure 16.

Forest plot of the meta-analysis on aquatic training effects for physical function and recovery.

Figure 17.

Funnel plot of the meta-analysis on aquatic training effects for physical function and recovery.

In summary, aquatic training demonstrates a robust positive effect on injury recovery, supported by low heterogeneity and marginal publication bias. While effect sizes may vary, the findings consistently favor aquatic training for functional rehabilitation in athletes.

Discussion

Aquatic training demonstrated moderate to high-moderate improvements in athletes' explosive power, speed, and agility although strength enhancement did not reach statistical significance, potentially due to insufficient resistance elements in the training protocols. Initial findings suggested a large effect size for aerobic capacity, but due to the limited sample size and high heterogeneity (I² = 75.4%), the conclusions are for reference only.

Notably, aquatic training significantly supported injury rehabilitation and fatigue reduction. Buoyancy and hydraulic resistance created a low-impact environment, enabling recovery exercises with minimized joint and muscular stress. According to the study by Kutzner, Richter, and Gordt et al. (18), aquatic exercise significantly reduces joint loading compared with land-based exercise. Specifically, when standing on one leg in chest-deep water, hip joint load is reduced by 58% and knee joint load by 62%. The water environment facilitated muscle relaxation, reduced soreness, and enhanced circulation. Existing research has demonstrated that the resistance of water compels simultaneous activation of agonist and antagonist muscles, which contributes to improved proprioceptive feedback and enhanced dynamic stability of the knee and ankle joints (32).

Although aquatic training has demonstrated significant benefits for athletic performance and rehabilitation, certain limitations remain in its clinical application. In track and field, muscle strains and overuse injuries are the most common types of damage. Once athletes transition from the acute phase to the rehabilitation stage, aquatic training can effectively accelerate muscle recovery. However, in cases involving open wounds, exposure to water should be avoided to prevent infection or aggravation of the injury. Moreover, factors such as water quality, pool accessibility, and individual adaptability to aquatic environments may also restrict its use. Therefore, the applicability of aquatic training should be carefully evaluated based on the athlete's individual characteristics and injury profile.

The meta-analysis results on explosive ability show that aquatic training moderately improves explosive ability in athletes, with high consistency and no significant bias. Among the included studies, 2 provided comparatively weaker evidence for an explosive-power benefit. The study used standing long jump (SLJ) as an explosive power assessment metric, differing from the mainstream CMJ indicator, which may partly account for the reduced significance. Tishukaj et al. (37) reported in their research that SLJ can be used to evaluate lower-limb explosive power in adolescents. The test design emphasizes its capability to reflect the rapid force output of lower-limb muscle groups through single maximal-effort jump distance. Melo Buitrago et al. (23) demonstrated partial overlap between CMJ and SLJ results in explosive strength assessments, but CMJ relies more on instantaneous power from lower-limb extensor muscles, while SLJ demands whole-body coordination (e.g., arm swing, take-off angle). Another included study used intervention measures (aquatic plyometric training) and evaluation metrics (CMJ) consistent with most studies in the meta-analysis, with no significant differences in subject sex or age characteristics. However, the athletes included in the study were all ball sports players (rugby, basketball, soccer), who may have more exposure to explosive power training during their daily routines compared with athletes in other sports. This could diminish the advantage of resistance provided by the aquatic environment. In addition, the small sample size (N = 20) increased the statistical requirements for detecting differences, which may explain why the impact of the water environment on explosive power improvement was similar to that of the land environment in this study. Overall, other studies demonstrated larger effect sizes, indicating broad effectiveness of aquatic training.

In addition, the studies included in this review primarily adopted water depths ranging from the waist to the chest. None of the studies, however, provided a rationale for the choice of water depth. Heywood et al. (14) reported that variations in water depth between waist and chest levels result in only subtle differences in joint loading and power output patterns, particularly during maximum speed or explosive movements. By contrast, deeper water depths have a more significant impact on joint loading during slow, closed-chain movements, as they better attenuate impact forces, making them more suitable for rehabilitation populations. Because the interventions in this study predominantly involved water depths ranging from waist to chest level, the differences in water depth have minimal impact on the variation in training outcomes.

$$F=\frac{1}{2}\rho C_{d}A{v}^2.$$ (1)

The effectiveness of aquatic explosive training primarily hinges on its dynamic resistance characteristics. The resistance in water follows formula 1, where resistance is proportional to the square of the movement velocity (1). Owing to the density and viscosity of water, during the execution of explosive movements in an aquatic environment, an increase in movement velocity leads to an exponential growth in water resistance. At the peak of the movement, when athletes attempt to maintain maximum acceleration, the water resistance reaches its peak, making it easier to reach the high depolarization threshold required for the activation of type II muscle fibers. To generate sufficient force to overcome the rapidly increasing resistance, the nervous system must recruit high-threshold type II muscle fibers. This resistance pattern is more effective in activating type II fast-twitch muscle fibers, promoting the phosphorylation of myosin regulatory light chains within muscle fibers, which enhances the sensitivity of the actin-myosin complex to calcium ions. This, in turn, accelerates the rate of cross-bridge formation, increasing the rate of muscle strength development. Alternatively, it may enhance spinal excitability, reduce presynaptic inhibition, and increase motor neuron output, ultimately contributing to improved explosive power in athletes (1). However, current experimental research on aquatic training and explosive power remains limited.

Across all studies reviewed, although not all used identical training protocols, the primary objective was to investigate the unique effects induced by the aquatic environment. Consequently, certain experimental designs were adapted according to environmental conditions, which precludes complete consistency of interventions across studies. Notably, the 2 studies reporting the largest effect sizes implemented fully identical intervention protocols. Taken together, these findings support the conclusion that aquatic training confers distinctive advantages over land-based training in enhancing explosive power.

For speed, the meta-analysis results demonstrate a moderate to substantial positive effect (SMD = 1.14) of aquatic training on speed enhancement in athlete populations. Both heterogeneity tests and funnel plot analysis indicated no significant interference or bias in the findings.

All included studies adopted short-distance sprint tests as the primary speed measure and consistently reported that aquatic training effectively improves speed by enhancing muscle stretch-shortening cycle capacity. This training modality facilitates rapid maximal force production, thereby enhancing sprint performance. These findings align with power-related outcomes. As highlighted in Sporri et al.'s (35) study, the aquatic environment may promote neuromuscular recruitment and myogenic adaptations during the concentric phase by reducing rebound-phase duration (the transition from muscle stretch to contraction) and leveraging fluid resistance-induced loading augmentation. This mechanism could augment force generation.

As for agility, the analysis demonstrated a significant effect of aquatic training in improving agility (SMD = 0.46), indicating a moderate positive impact on athletes' agility performance. The low heterogeneity and absence of significant publication bias support the reliability of the findings. Although aquatic training showed notable short-term benefits, the prediction interval crossed zero, suggesting that the current evidence cannot guarantee these favorable outcomes will persist over time—they may diminish with prolonged observation or fail to materialize across different populations and training contexts. While the included studies used 3 distinct assessment metrics (T-T, 5–0–5 T, and 4 × 10 m shuttle run [SR]), but Jose, Irineu, and Lucas et al. highlighted a significant correlation between 5–0–5 T and T-T (r = 0.42), indicating their ability to independently reflect specific dimensions of agility. Moreover, linear sprint speed showed a significant correlation with T-T (r = 0.53–0.73), suggesting that 4 × 10 m SR indirectly assesses agility by evaluating acceleration and deceleration capacity (13). Agility assessments evaluate multiple dimensions, including direction change, acceleration, and reactive speed. The T-T, 5–0–5 T, and 4 × 10 m SR tests all require rapid deceleration and reacceleration, aligning with the biomechanical optimization demands of agility. Combined with the nonsignificant Egger's test results, this suggests low heterogeneity among the 3 metrics, supporting their concurrent use as valid assessments of agility performance.

From a biomechanical perspective, the aquatic environment simultaneously provides 2 opposite effects: load reduction and load enhancement. However, these effects act on different dimensions of movement and bodily systems. The buoyancy of water significantly reduces ground reaction forces on joints and muscles (18), creating a low-load training environment that allows the neuromuscular system to adapt efficiently under reduced physical stress. In the vertical direction, buoyancy significantly reduces joint compressive load and impact stress. When movement involves vertical motion, the weight reduction caused by buoyancy leads to a decrease in metabolic cost and cardiovascular response compared with the same exercise on land (9). This “joint load reduction effect” provides athletes with a safe training environment, especially beneficial for those with joint issues or in rehabilitation, facilitating agility training. Simultaneously, water resistance provides a unique training stimulus, requiring athletes to overcome drag during movement. The resistance, which increases exponentially with speed, amplifies the load on muscles during exertion, thereby enhancing explosive power and endurance—both critical for agility performance (35). Furthermore, during explosive movements, water resistance amplifies the training effect on muscles. The research by Yu et al. (42) confirmed this point. They suggested that during resistance training in an aquatic environment, the density and high viscosity of water significantly increased the resistance to muscle movement, thereby enhancing neuromuscular stimulation to the muscles. This resistance forces muscle fibers, particularly fast-twitch fibers, to be activated to a greater extent and extends the duration of tension. The properties of water may also further optimize the training effects of explosive movements by increasing the recruitment frequency of motor units and improving the coordination of synergistic muscles (42). Agility requires neuromuscular coordination. Because aquatic training simultaneously enhances explosive power and speed, these combined effects optimize neuromuscular control, improving rapid reaction and directional change abilities to promote agility performance.

Based on the results of the muscle strength meta-analysis, aquatic training does not significantly enhance muscle strength. Although aquatic training may offer some benefits, its effects are similar to those of the control group. The included studies did not incorporate resistance training, despite numerous studies indicating that resistance training at approximately 80% of 1 repetition maximum is the most effective method for increasing muscle strength (17,29). This suggests that higher intensity training is required to achieve more substantial gains in muscle strength.

The meta-analysis suggested improved aerobic capacity after aquatic training, but the small sample size, high heterogeneity, and wide effect size range limit interpretation. The funnel plot could not assess publication bias. These findings are for reference only. Analysis of aquatic training's efficacy must account for its unique characteristics. The buoyancy and reduced joint/soft-tissue loading make it particularly suitable for athletes with injuries or low physical conditioning. Wilber, Moffatt, and Scott et al. (40) found that deep-water running effectively maintains aerobic capacity with minimal joint stress, highlighting its utility in injury rehabilitation for sustaining physiological adaptations. Deep-water running serves as an effective alternative training modality for maintaining aerobic capacity. While no significant changes were observed in physiological markers such as V̇o₂max or ventilatory threshold, it provides training effects comparable with land-based running, particularly benefiting injured athletes by reducing joint stress—a key consideration in this study.

In addition, another included study demonstrated that aquatic Tabata training significantly improved endurance performance in experimental groups compared with controls (16) Athletes generally perceived this training as highly efficient and easy to learn, although some noted its limited suitability for certain age groups or individuals with poor water adaptability. From another perspective, aquatic Tabata also offers a viable approach for enhancing endurance in athletes with physical impairments.

This balance meta-analysis examines the effect of aquatic training interventions on balance ability, focusing on aerobic and stretching exercises in water. The intervention methods exhibited low heterogeneity based on study inclusion criteria. For outcome measures, the included studies used the Y-Balance Test (YBT) and Star Excursion Balance Test (SEBT), both demonstrating high reliability and validity in dynamic balance assessment (ICC: 0.85–0.99 for YBT, comparable with modified SEBT reliability) (28,34) for.

Heterogeneity among studies is noteworthy. The 3 included studies had different intervention plans. Aynollah and Khadije (2025) used the strictest experimental design, implementing exactly the same training program in both aquatic and terrestrial environments. This study found that water-based training and land-based training had similar effects on improving dynamic balance (p > 0.05), challenging the hypothesis that aquatic environments are inherently superior to land environments. By contrast (25), the studies by Sadaak et al. (31) and Chen et al. (7) which used different training programs, particularly Chen et al.'s study where the intervention plans for different groups were completely different, observed that the environmental differences might more reflect the differences in the training programs themselves rather than the influence of environmental factors.

It is worth noting that all 3 studies focused on injured populations, and the test indicators used reflected dynamic balance ability and joint extension capacity, which are related to the rehabilitation progress and injury severity of the individuals. Naderi suggested that the mechanism by which water training improves balance differs from land training. It alleviates joint pressure through buoyancy, promotes recovery at the injury site, increases comfort, and enhances muscle strength and proprioception, thereby improving balance. However, it is more specific and aligned with daily movement patterns, helping to improve functional skills such as jumping and lateral movements, making it more effective in enhancing balance (25). Chen, on the other hand, believed that water training is more effective than land training in increasing the range of motion and alleviating joint pressure, and because balance ability is somewhat influenced by joint recovery and range of motion, the water training group in their study showed better results (7).

Combining the findings from these 3 studies, we find that for populations with chronic joint issues, both water-based and land-based balance training can effectively improve balance ability. However, for individuals with acute joint injuries, water training, with its unique environment, provides effective joint protection during exercise, thus enhancing recovery efficiency and more effectively promoting balance recovery. Therefore, different training programs should be chosen according to the actual goals. However, current research on the differences in balance development between different environments for healthy athletes is almost nonexistent. This may be due to the fact that the environmental settings in current studies may not have sufficiently simulated the actual sports environment for athletes, leading to inaccurate assessments of behavioral differences (36). On the other hand, for healthy athletes, the development of balance ability may already meet the requirements for competitive sports, and other abilities such as physical fitness, strength, and speed are more in need of improvement.

Physical function and recovery meta-analysis results and relevant literature demonstrate significant positive effects of aquatic training on athletes' functional recovery. The pooled effect size indicates a statistically meaningful benefit, confirming its role in injury rehabilitation.

To further examine the superiority of aquatic training, we analyzed included studies. Aynollah and Khadije (2025) highlighted buoyancy as a key advantage—reducing joint and muscular loading while facilitating safe movement. The viscosity of water additionally enhances neuromuscular activation and proprioceptive retraining, crucial for functional restoration. These mechanisms collectively contribute to more efficient recovery compared with land-based interventions (25). For athletes, this load-reducing effect is particularly crucial as it effectively minimizes impact forces and joint stress during movement, thereby reducing articular wear—especially beneficial for those in postinjury recovery. The buoyancy of water provides protective support for high-impact joints such as knees and ankles, lowering reinjury risks.

Moreover, water's higher resistance compared with air generates multiplanar loading, engaging diverse muscle groups more evenly and avoiding compensatory recruitment patterns common in land-based training. This enhances overall muscular strength and balance while providing a low-impact yet high-efficiency training environment.

Water temperature also significantly influences soft tissues. Thermoneutral immersion improves peripheral circulation, alleviates muscle tension, and reduces exercise-induced fatigue. The combined effects of hydrostatic pressure and warmth enhance metabolic clearance, accelerate recovery, and provide analgesic benefits—particularly valuable for managing post-traumatic pain and stiffness during early rehabilitation (25).

For athletes recovering from sports injuries, particularly those with joint or ligament damage, fear of reinjury and a lack of confidence are common barriers to rehabilitation. As noted in Sadaak, AbdElMageed, and Ibrahim's study, aquatic training provides a safer and more comfortable environment, allowing athletes to gradually rebuild training intensity without excessive anxiety, thanks to water's supportive properties and low-impact nature (31). This psychological ease and restored confidence constitute a key rehabilitation advantage of aquatic training, aiding athletes in returning to competition sooner.

In addition, hydrostatic pressure increases with depth, enhancing venous return and lymph flow to reduce peripheral resistance and improve circulatory efficiency. The cold stimulus of water induces vasoconstriction-dilation cycles that enhance vascular elasticity, while hydrostatic pressure-driven fluid shifts optimize oxygen delivery and metabolic waste clearance (e.g., lactate). For injured athletes, aquatic training not only rebuilds muscular function but also prevents detraining effects by maintaining overall fitness during recovery—counteracting metabolic decline and physical deconditioning from prolonged inactivity (41,44).

In summary, aquatic training demonstrates significant benefits in functional recovery, leveraging water's physical properties, thermoregulatory effects, and psychological soothing mechanisms to facilitate postinjury rehabilitation in athletes. Its low-impact yet highly efficient nature makes it an ideal rehabilitation method. Although prediction intervals indicate potential variability in effect sizes across studies, the consistent adaptability of aquatic training as a low-impact rehabilitation tool remains evident. This adaptability proves particularly valuable for promoting recovery in injured track and field athletes. Moreover, during periods of high-intensity training fatigue, aquatic training can serve as a supplementary training modality to enhance recovery and mitigate injury risks.

Overall, aquatic training demonstrates significant positive effects on explosive power, speed, and agility, making it particularly suitable for balance, postexercise recovery, and injury rehabilitation. However, its efficacy in improving aerobic capacity and muscle strength requires further investigation, and the study still has several limitations. The included studies do not exclusively focus on training in aquatic environments but rather compare different types of training protocols. This variability makes it difficult to attribute the study results directly to the unique characteristics of the aquatic environment, such as buoyancy or resistance, although these factors may have had some influence on the outcomes. Existing evidence does not confirm that the aquatic environment is the direct cause of the observed improvements. However, the studies included in this research suggest that the aquatic environment may have contributed to certain performance enhancements, which could be the result of the interaction between the training protocols and the water environment. Future research or training programs might consider integrating training protocols with aquatic environments to further explore their combined effects. Aquatic training has established itself as an innovative modality that effectively enhances athletic performance and functional recovery. Current research is constrained by high heterogeneity, small sample sizes. Future studies should standardize protocols and expand longitudinal assessments to enhance generalizability. With advancing research, aquatic training holds promising potential for broader applications in sports training and rehabilitation. Future research should focus on these outcomes and standardize aquatic training protocols to validate long-term efficacy.

Practical Applications

Aquatic training provides a novel approach for competitive land sports athletes, demonstrating advantages over land-based training in enhancing explosive power (SMD = 0.79, moderate effect), speed (SMD = 1.14, large effect), agility (SMD = 0.47, small-to-moderate effect), and physical recovery (SMD = 0.38, small effect). To improve explosive power and speed, strength and conditioning coaches can implement aquatic training protocols, such as aquatic plyometrics or resistance exercises, conducted 2–3 times per week for 30–45 minutes in waist-to chest-deep water. These protocols should progress in intensity or volume over 6–14 weeks to optimize neuromuscular adaptations while reducing joint stress. For athletes recovering from injuries, sports medicine clinicians can incorporate aquatic exercises, such as balance training or deep-water running, into rehabilitation programs to promote recovery and alleviate fatigue, particularly for those with knee or ankle injuries. Coaches and athletes should integrate aquatic training with land-based programs to balance performance enhancement and recovery, tailoring protocols to individual needs and training phases. However, due to high heterogeneity and limited studies, the effects on muscle strength balance, and aerobic capacity remain inconclusive. Future research could focus on controlled experiments comparing aquatic running with land running to assess the differences in their effects on cardiovascular fitness. While balance ability has primarily been studied in injured athletes, researchers may shift their attention to athletes in high balance-demand sports. In addition, due to the limitations of the aquatic environment in providing sufficient load for large muscle groups, researchers should further develop aquatic resistance training protocols.