Dataset Information

A holistic approach to performance prediction in collegiate athletics: player, team, and conference perspectives.

ABSTRACT: Predictive sports data analytics can be revolutionary for sports performance. Existing literature discusses players' or teams' performance, independently or in tandem. Using Machine Learning (ML), this paper aims to holistically evaluate player-, team-, and conference (season)-level performances in Division-1 Women's basketball. The players were monitored and tested through a full competitive year. The performance was quantified at the player level using the reactive strength index modified (RSImod), at the team level by the game score (GS) metric, and finally at the conference level through Player Efficiency Rating (PER). The data includes parameters from training, subjective stress, sleep, and recovery (WHOOP straps), in-game statistics (Polar monitors), and countermovement jumps. We used data balancing techniques and an Extreme Gradient Boosting (XGB) classifier to predict RSI and GS with greater than 90% accuracy and a 0.9 F1 score. The XGB regressor predicted PER with an MSE of 0.026 and an R² of 0.680. Ensemble of Random Forest, XGB, and correlation finds feature importance at all levels. We used Partial Dependence Plots to understand the impact of each feature on the target variable. Quantifying and predicting performance at all levels will allow coaches to monitor athlete readiness and help improve training.

SUBMITTER: Taber CB

PROVIDER: S-EPMC10786827 | biostudies-literature | 2024 Jan

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A holistic approach to performance prediction in collegiate athletics: player, team, and conference perspectives.

Taber Christopher B CB Sharma Srishti S Raval Mehul S MS Senbel Samah S Keefe Allison A Shah Jui J Patterson Emma E Nolan Julie J Sertac Artan N N Kaya Tolga T

Scientific reports 20240112 1

Predictive sports data analytics can be revolutionary for sports performance. Existing literature discusses players' or teams' performance, independently or in tandem. Using Machine Learning (ML), this paper aims to holistically evaluate player-, team-, and conference (season)-level performances in Division-1 Women's basketball. The players were monitored and tested through a full competitive year. The performance was quantified at the player level using the reactive strength index modified (RSI ...[more]

PMID: 38216641

Similar Datasets

Project description:This prospective cohort study examined the relationship between a panel of four serum proteomic biomarkers (glial fibrillary acidic protein [GFAP], ubiquitin C-terminal hydrolase-L1 [UCH-L1], total Tau, and neurofilament light chain polypeptide [NF-L]) in 52 players from two different cohorts of male collegiate student football athletes from two different competitive seasons of Division I National Collegiate Athletic Association Football Bowl Subdivision. This study evaluated changes in biomarker concentrations (as indicators of brain injury) over the course of the playing season (pre- and post-season) and also assessed biomarker concentrations by player position using two different published classification systems. Player positions were divided into: 1) speed (quarterbacks, running backs, halfbacks, fullbacks, wide receivers, tight ends, defensive backs, safety, and linebackers) versus non-speed (offensive and defensive linemen), and 2) "Profile 1" (low frequency/high strain magnitudes positions including quarterbacks, wide receivers, and defensive backs), "Profile 2" (mid-range impact frequency and strain positions including linebackers, running backs, and tight ends), and "Profile 3" (high frequency/low strains positions including defensive and offensive linemen). There were significant increases in GFAP 39.3 to 45.6 pg/mL and NF-L 3.5 to 5.4 pg/mL over the course of the season (p < 0.001) despite only five players being diagnosed with concussion. UCH-L1 decreased significantly, and Tau was not significantly different. In both the pre- and post-season blood samples Tau and NF-L concentrations were significantly higher in speed versus non-speed positions. Concentrations of GFAP, Tau, and NF-L increased incrementally from "Profile 3," to "Profile 2" to "Profile 1" in the post-season. UCH-L1 did not. GFAP increased (by Profiles 3, 2, 1) from 42.4 to 49.6 to 78.2, respectively (p = 0.051). Tau increased from 0.37 to 0.61 to 0.67, respectively (p = 0.024). NF-L increased from 3.5 to 4.9 to 8.2, respectively (p < 0.001). Although GFAP and Tau showed similar patterns of elevations by profile in the pre-season samples they were not statistically significant. Only NF-L showed significant differences between profiles 2.7 to 3.1 to 4.2 in the pre-season (p = 0.042). GFAP, Tau, and NF-L concentrations were significantly associated with different playing positions with the highest concentrations in speed and "Profile 1" positions and the lowest concentrations were in non-speed and "Profile 3" positions. Blood-based biomarkers (GFAP, Tau, NF-L) provide an additional layer of injury quantification that could contribute to a better understanding of the risks of playing different positions.

Project description:ObjectivesValue-based healthcare implies that healthcare issues are addressed most effectively with the 'physicians in the lead' (PIL) strategy. This study explores whether PIL also supports a holistic care approach that patients are increasingly demanding.DesignA qualitative research design was used.SettingThis study was conducted in a general hospital in the Netherlands with an integrated PIL strategy.ParticipantsSemistructured interviews were conducted with 14 hospital stakeholders: 13 stakeholders of an Obstetrics and Gynaecology department (the hospital's Patient Council (n=1), nurses (n=2), midwives (n=2), physicians (n=2), residents (n=2), the non-medical business managers of the Obstetrics and Gynaecology department (n=2) the Board of Directors (n=2)) and a member of the Dutch National Healthcare Institute's Innovative Healthcare Professions programme.ResultsAccording to diverse stakeholders, PIL does not support a holistic healthcare delivery approach, primarily because of the strong biomedical focus of the physicians. Although physicians can be educated to place more emphasis on the holistic outcome, holistic care delivery requires greater integration and teamwork in the care chain. As different healthcare professions are complementary to each other, a new strategy of a 'team in the lead' was suggested to meet the holistic healthcare demands. Besides this new strategy, there is a need for an extramural care management coordination centre where patients are able to receive support in managing their own care. This centre should also facilitate services similar to the core function of a church or community centre. These services should help patients to deal with different holistic dimensions that are important for their well-being.ConclusionsThe PIL strategy appears to be insufficient for holistic healthcare delivery. A 'team in the lead' approach should be considered to meet the holistic healthcare demands. Further research should focus on observing PIL in different cultures and exploring the effectiveness of the strategy 'team in the lead'.

Dataset Information

A holistic approach to performance prediction in collegiate athletics: player, team, and conference perspectives.

Publications

A holistic approach to performance prediction in collegiate athletics: player, team, and conference perspectives.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets