This systematic review and meta-analysis investigates the application of artificial intelligence (AI) in sports performance analysis. Sixteen peer-reviewed studies spanning 13 distinct sports disciplines were included, employing a variety of AI techniques—from classical machine learning algorithms to advanced deep learning and computer vision models. Methods applied encompassed Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM) networks, reinforcement learning, and predictive modeling architectures. The pooled average classification accuracy was 87.78% (95% CI: 82.66–92.90), although substantial heterogeneity was observed across studies (I² = 93.75%). Computer vision and deep learning-based approaches were associated with higher performance metrics in several studies, particularly in movement-intensive sports such as tennis and basketball. Nevertheless, several challenges were identified, including lack of standardization in model evaluation, limited algorithmic transparency, and difficulties in generalizing findings from controlled laboratory environments to real-world competitive settings. The results underscore the promising role of AI in optimizing training protocols, supporting tactical decisions, and enhancing injury prevention strategies. Further research is warranted to address the ethical, methodological, and practical considerations surrounding the deployment of AI in sports contexts.

Introduction: Ice hockey is a physically demanding sport that requires a combination of aerobic and anaerobic capacities, as well as specific on-ice skills. This study aimed to compare the effectiveness of individualized interval training on a slide board with training on a cyclo-ergometer, focusing on aerobic capacity and specific fitness in elite ice hockey players. Materials and Methods: Thirty players were randomly divided into an experimental (slide board) and a control (cyclo-ergometer) group. Both groups underwent a 6-week training program with the same intensity and volume. Results: The results showed that both forms of training led to significant improvements in aerobic capacity, as indicated by VO2max and VO2 lactate thresholds. The slide board training group demonstrated higher effect sizes (d = 1.49, 1.62) than the cyclo-ergometer one (0.76, 0.43). Specific on-ice performance tests showed that slide board training resulted in faster skating times (p = 0.0002), suggesting a greater transfer of training effects to on-ice skills. Conclusions: This study highlights the benefits of incorporating slide board interval training into the off-season training regimen of elite ice hockey players. Slide board training improves specific on-ice skills, which are crucial for high-level hockey performance. These findings provide valuable insights for coaches and strength and conditioning specialists working with elite ice hockey players.

Background/Objectives: COVID-19 poses significant physiological challenges for athletes, particularly those engaged in high-intensity intermittent sports such as ice hockey. This study aimed to evaluate the impact of SARS-CoV-2 infection—especially symptomatic cases—on aerobic and anaerobic performance in professional ice hockey players. Methods: Fifty athletes from the Polish Hockey League were assigned to three groups: control (CG, n = 13), asymptomatic COVID-19 (NSG, n = 28), and symptomatic COVID-19 with post-infection SpO₂ < 90% (WSG, n = 9). Each underwent assessments at three time points—pre-preparatory period 2020/2021, post-preparatory period 2020/2021, and pre-preparatory period 2021/2022. Aerobic capacity was measured via a progressive cycle ergometer test (VO₂max, VO₂ at lactate threshold [VO_2Lt], minute ventilation [V’E], breathing frequency [BF], and lactate clearance rate [ΔLa]), and anaerobic capacity via a 30 s Wingate test (relative mean power). Results: Compared with CG and NSG, symptomatic athletes exhibited significant post-infection declines in VO₂max (48.2 ± 2.9 vs. 56.2 ± 6.2 and 54.6 ± 3.9 mL/kg/min; p = 0.006, d = 1.56 vs. CG; p < 0.024, d = 1.79 vs. NSG) and VO_2Lt (p < 0.05). Relative mean power also decreased in WSG (p < 0.05). In contrast, CG and NSG improved or maintained these metrics over the same period. Symptomatic players showed elevated BF post infection (p = 0.022, d = 1.72) and reduced V’E (p = 0.035; d = 0.83), while ΔLa was markedly lower (p = 0.0004; d = 2.86). Conclusions: SARS-CoV-2 infection, particularly symptomatic cases, can significantly impair both aerobic and anaerobic capacity in elite hockey players. Targeted recovery protocols are essential for restoring performance in affected athletes.

This study investigated the influence of aerobic capacity on lactate clearance rate and heart rate recovery during ice hockey matches. Considering the growing intensity and anaerobic demands of modern ice hockey, the ability to recover quickly between high-intensity shifts is essential for optimal performance. Thirty-eight amateur ice hockey players (age: 35 ± 5 years; VO₂max: 48.93 ± 3.88 mL·min⁻¹·kg⁻¹) from the Silesian Amateur Hockey League underwent laboratory ramp tests to determine VO₂max, followed by on-ice repeated sprint tests and heart rate monitoring during matches. The results demonstrated significant positive correlations between VO₂max and lactate clearance (ΔLa4–8min [mmol/L]= 2.55 ± 0.58 mmol·L⁻¹; rho = 0.545; p < 0.001), as well as heart rate recovery (Δ%HRmax = 25.88 ± 2.09%; rho = 0.682; p < 0.001). Players with higher VO₂max exhibited a faster reduction in heart rate during recovery periods between shifts and maintained better sprint performance (rho = –0.877; p < 0.001). These findings confirm that higher aerobic capacity enhances both metabolic and autonomic recovery processes, enabling players to sustain high-intensity efforts more effectively during the game. The study highlights the importance of developing aerobic fitness in hockey training to improve recovery efficiency and match performance.

Background: Shooting velocity is a critical determinant of competitive success in ice hockey, yet evidence for the use of weighted-implement training in high-level junior players is limited and the long-term retention of such adaptations has not been documented. The aim of this study was to compare the effects of off-ice shooting training performed with a heavy (260 g) versus a regular (170 g) puck on on-ice shooting velocity, accuracy and handgrip strength in junior players, and to examine the retention of these changes. Methods: Twenty male junior ice hockey players (18–19 years) were randomly allocated to a Heavy-puck group (n = 10) or a Regular-puck group (n = 10) and completed an identical six-week off-ice shooting programme (18 sessions, 100 shots per session) with their respective pucks. On-ice wrist-shot and snap-shot speed (radar; standard 170 g puck for both groups), on-ice shooting accuracy and bilateral handgrip strength were assessed before the intervention (pre-test), immediately after six weeks (post 6 weeks) and after a six-week retention period of normal on-ice training (post 12 weeks). Data were analysed with 2 × 3 mixed-model ANOVA with Bonferroni-corrected post hoc comparisons. Results: A significant Group × Time interaction was found for wrist-shot speed (ηp² = 0.61), snap-shot speed (ηp² = 0.78), left-hand handgrip strength (ηp² = 0.30) and shooting accuracy (ηp² = 0.24). The Heavy-puck group displayed substantially larger velocity gains at both post 6 weeks (wrist shot d = 2.97; snap shot d = 4.73) and post 12 weeks (d = 2.56 and d = 3.21, respectively). Left-hand handgrip strength gain was also greater in the Heavy-puck group at post 12 weeks (d = 1.40). A short-term cost on accuracy was observed in the Heavy-puck group at post 6 weeks (d = −1.21), which was fully recovered at post 12 weeks. Conclusions: Heavy-puck off-ice training produced large and durable improvements in on-ice puck velocity, with a transient and recoverable cost on accuracy, supporting its inclusion in the off-ice preparation of junior ice hockey players.

Post-industrial cities increasingly present science, health and physical activity as engines of sustainable transformation, yet residents’ own expectations are rarely measured before programmes are designed. This study examined how the inhabitants of Katowice (Poland) and people functionally linked with the city view the sciences of physical culture as a resource for the sustainable transformation of a post-industrial city, and modelled their expectations in the fields of physical culture, sport and recreation in the run-up to the city’s designation as European City of Science 2024. A convergent mixed-methods design (November 2022) combined a non-probability, mixed-mode survey of 500 respondents (a broad and diverse, but not representative, sample) with focus groups involving 130 residents, sector representatives and elite athletes. Exploratory principal component analysis identified four dimensions explaining 58.2% of the variance (α = 0.85–0.90): perceived quality of the current offer, demand for development, valuation of physical-culture science and health, and institutional and ambassadorial visibility; a sensitivity analysis using polychoric correlations and oblique rotation reproduced the allocation of 29 of the 30 items, although the structure still requires confirmatory validation. Valuation of science and health was the highest-rated dimension (M = 4.20) and rose with education and age (ε² = 0.13 for both). Self-rated physical activity was the only predictor significant across three of four dimensions in adjusted models, although the explained variance was modest (R² = 0.06–0.17). Only 35.2% of respondents were satisfied with health-promoting classes, whereas 80.6–90.0% rated health-related knowledge and healthy ageing as important, suggesting a possible mismatch between priorities and perceived provision. The findings are translated into five context-specific pillars for health-oriented policy in Katowice and into conditional lessons for other post-industrial cities.