Variability in the Study Quality Appraisals Reported in Systematic Reviews on the Acute:Chronic Workload Ratio and Injury Risk

A Narrative Review for a Machine Learning Application in Sports: An Example Based on Injury Forecasting in Soccer

In the last decade, the number of studies about machine learning algorithms applied to sports, e.g., injury forecasting and athlete performance prediction, have rapidly increased. Due to the number of works and experiments already present in the state-of-the-art regarding machine-learning techniques in sport science, the aim of this narrative review is to provide a guideline describing a correct approach for training, validating, and testing machine learning models to predict events in sports science. The main contribution of this narrative review is to highlight any possible strengths and limitations during all the stages of model development, i.e., training, validation, testing, and interpretation, in order to limit possible errors that could induce misleading results. In particular, this paper shows an example about injury forecaster that provides a description of all the features that could be used to predict injuries, all the possible pre-processing approaches for time series analysis, how to correctly split the dataset to train and test the predictive models, and the importance to explain the decision-making approach of the white and black box models.

Acute Effects of Warm-Up, Exercise and Recovery-Related Strategies on Assessments of Soccer Kicking Performance: A Critical and Systematic Review

BACKGROUND

A number of reviews have collated information on the impact of warming-up, physical exertion and recovery strategies on physical, subjective and physiological markers in soccer players yet none have solely analyzed their potential effects on components of kicking performance.

OBJECTIVE

To systematically analyse the influence of warm-up, exercise and/or recovery-related strategies on kicking performance in male soccer players and provide a critical appraisal on research paradigm related to kicking testing constraints and data acquisition methods.

METHODS

A systematic literature search was performed (until July 2020) in PubMed, Web of Science, SPORTDiscus, Scopus and ProQuest. Studies in male soccer populations, which included the effects of warm-up routines, physical exercise and/or recovery-related interventions, reported on comparisons pre-post or between experimental conditions and that computed at least one measure of kicking kinematics and/or performance were considered. Methodological quality and risk of bias were determined for the included studies. Constraints related to kicking testing and data acquisition methods were also summarized and discussed.

RESULTS

Altogether, 52 studies were included. Of these, 10 examined the respective effects of a warm-up, 34 physical exercise, and 21 recovery-related strategies. The results of eight studies showed that lower limb kinematics, kicking accuracy or ball velocity were improved following warm-ups involving dynamic but not static stretching. Declines in ball velocity occurred notably following intermittent endurance or graded until exhaustion exercise (three studies in both cases) without inclusion of any ball skills. In contrast, conflicting evidence in five studies was observed regarding ball velocity following intermittent endurance exercise interspersed with execution of ball skills. Kicking accuracy was less frequently affected by physical exercise (remained stable across 14 of 19 studies). One investigation indicated that consumption of a carbohydrate beverage pre- and mid-exercise demonstrated benefits in counteracting the potentially deleterious consequences of exercise on ball velocity, while four studies reported conflicting results regarding kicking accuracy. Most evidence synthesized for the interventions demonstrated moderate level (77%) and unclear-to-high risk of bias in at least one item evaluated (98%). The main limitations identified across studies were kicks generally performed over short distances (50%), in the absence of opposition (96%), and following experimental instructions which did not concomitantly consider velocity and accuracy (62%). Also, notational-based metrics were predominantly used to obtain accuracy outcomes (54%).

CONCLUSIONS

The results from this review can help inform future research and practical interventions in an attempt to measure and optimise soccer kicking performance. However, given the risk of bias and a relative lack of strong evidence, caution is required when applying some of the current findings in practice.

PROSPERO ID

CRD42018096942.