篮球中的球员原型:优化阵容组成以创建一支冠军球队。

IF 2.6 Q2 SPORT SCIENCES
Frontiers in Sports and Active Living Pub Date : 2025-08-29 eCollection Date: 2025-01-01 DOI:10.3389/fspor.2025.1639431
Luke S J Penner
{"title":"篮球中的球员原型:优化阵容组成以创建一支冠军球队。","authors":"Luke S J Penner","doi":"10.3389/fspor.2025.1639431","DOIUrl":null,"url":null,"abstract":"<p><p>The present study assessed the feasibility of grouping university and professional basketball players across different leagues based on their playing styles to optimize championship team construction, moving beyond traditional positional constraints towards a positionless approach. A comprehensive dataset of 22,500 elite professional and university athletes from 110 leagues, sourced from EuroBasket, RealGM, and USportsHoops, was analyzed. Player performance was quantified using 13 standardized box score statistics converted to per 48 min. Utilizing the k-means algorithm, players were clustered 1nto 9 distinct player archetypes. Multiple linear regression models were then developed for each archetype, predicting \"points per minute\" to facilitate player ranking, further refined by a data-driven league quality weighting system. Optimal player cluster proportions were derived from analyses of 2018/19 NBA lineups and 2014-2018 1nternational medal-winning teams to create the most effective team line-ups. The model's utility was demonstrated by selecting a hypothetical Team Canada roster for the 2019 FIBA World Cup, which showed effectiveness 1n 1dentifying a robust team composition and predicting the rise of future high-potential players. Additionally, the model's effectiveness was evaluated by comparing 1ts results to the composition of the 2000 and 2024 Canadian National Olympic teams. The findings revealed 9 unique player clusters, demonstrating the model's potential as a 5aluable tool for guiding coaching decisions 1n drafting and signing future players to fit their roster composition effectively. The model proved to be a novel method for talent identification due to 1ts evaluation of players across leagues and 1ts usage of readily accessible data for coaches and scouts alike. Despite limitations related to data sources and subjective weighting, this research provides a sophisticated analytical tool for players, coaches, scouts, and general managers, offering a comprehensive league strength metric and a nuanced player ranking system to enhance roster development alongside the expertise of coaches 1n the evolving global basketball landscape.</p>","PeriodicalId":12716,"journal":{"name":"Frontiers in Sports and Active Living","volume":"7 ","pages":"1639431"},"PeriodicalIF":2.6000,"publicationDate":"2025-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12426101/pdf/","citationCount":"0","resultStr":"{\"title\":\"Player archetypes within basketball: optimizing roster composition to create a championship team.\",\"authors\":\"Luke S J Penner\",\"doi\":\"10.3389/fspor.2025.1639431\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>The present study assessed the feasibility of grouping university and professional basketball players across different leagues based on their playing styles to optimize championship team construction, moving beyond traditional positional constraints towards a positionless approach. A comprehensive dataset of 22,500 elite professional and university athletes from 110 leagues, sourced from EuroBasket, RealGM, and USportsHoops, was analyzed. Player performance was quantified using 13 standardized box score statistics converted to per 48 min. Utilizing the k-means algorithm, players were clustered 1nto 9 distinct player archetypes. Multiple linear regression models were then developed for each archetype, predicting \\\"points per minute\\\" to facilitate player ranking, further refined by a data-driven league quality weighting system. Optimal player cluster proportions were derived from analyses of 2018/19 NBA lineups and 2014-2018 1nternational medal-winning teams to create the most effective team line-ups. The model's utility was demonstrated by selecting a hypothetical Team Canada roster for the 2019 FIBA World Cup, which showed effectiveness 1n 1dentifying a robust team composition and predicting the rise of future high-potential players. Additionally, the model's effectiveness was evaluated by comparing 1ts results to the composition of the 2000 and 2024 Canadian National Olympic teams. The findings revealed 9 unique player clusters, demonstrating the model's potential as a 5aluable tool for guiding coaching decisions 1n drafting and signing future players to fit their roster composition effectively. The model proved to be a novel method for talent identification due to 1ts evaluation of players across leagues and 1ts usage of readily accessible data for coaches and scouts alike. Despite limitations related to data sources and subjective weighting, this research provides a sophisticated analytical tool for players, coaches, scouts, and general managers, offering a comprehensive league strength metric and a nuanced player ranking system to enhance roster development alongside the expertise of coaches 1n the evolving global basketball landscape.</p>\",\"PeriodicalId\":12716,\"journal\":{\"name\":\"Frontiers in Sports and Active Living\",\"volume\":\"7 \",\"pages\":\"1639431\"},\"PeriodicalIF\":2.6000,\"publicationDate\":\"2025-08-29\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12426101/pdf/\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Frontiers in Sports and Active Living\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.3389/fspor.2025.1639431\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"2025/1/1 0:00:00\",\"PubModel\":\"eCollection\",\"JCR\":\"Q2\",\"JCRName\":\"SPORT SCIENCES\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Frontiers in Sports and Active Living","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.3389/fspor.2025.1639431","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2025/1/1 0:00:00","PubModel":"eCollection","JCR":"Q2","JCRName":"SPORT SCIENCES","Score":null,"Total":0}
引用次数: 0

摘要

本研究评估了将不同联赛的大学和职业篮球运动员根据他们的比赛风格分组以优化冠军球队建设的可行性,超越传统的位置限制,转向无位置方法。该研究分析了来自110个联赛的22500名精英职业和大学运动员的综合数据集,这些数据来自EuroBasket、RealGM和USportsHoops。球员的表现是用13个标准化的统计数据来量化的,换算成每48分钟。利用k-means算法,玩家被分成9个不同的玩家原型。然后为每个原型开发了多个线性回归模型,预测“每分钟得分”以促进球员排名,并通过数据驱动的联赛质量加权系统进一步完善。通过对2018/19赛季NBA阵容和2014-2018赛季国际获奖球队的分析,得出最佳球员集群比例,以创建最有效的球队阵容。通过选择2019年国际篮联世界杯的假设加拿大队名单,证明了该模型的实用性,该模型在确定强大的球队组成和预测未来高潜力球员的崛起方面显示出有效性。此外,通过将模型的结果与2000年和2024年加拿大国家奥运代表队的组成进行比较,对模型的有效性进行了评估。调查结果显示了9个独特的球员集群,证明了该模型作为指导教练决策的有价值的工具的潜力,可以在选秀和签下未来的球员,以有效地适应他们的阵容构成。该模型被证明是一种新的人才识别方法,因为它对不同联赛的球员进行了评估,并且为教练和球探等人提供了易于获取的数据。尽管数据来源和主观权重存在局限性,但本研究为球员、教练、球探和总经理提供了一个复杂的分析工具,提供了一个全面的联赛实力指标和一个细致入微的球员排名系统,以促进球员名单的发展,以及教练在不断发展的全球篮球格局中的专业知识。
本文章由计算机程序翻译,如有差异,请以英文原文为准。

Player archetypes within basketball: optimizing roster composition to create a championship team.

Player archetypes within basketball: optimizing roster composition to create a championship team.

The present study assessed the feasibility of grouping university and professional basketball players across different leagues based on their playing styles to optimize championship team construction, moving beyond traditional positional constraints towards a positionless approach. A comprehensive dataset of 22,500 elite professional and university athletes from 110 leagues, sourced from EuroBasket, RealGM, and USportsHoops, was analyzed. Player performance was quantified using 13 standardized box score statistics converted to per 48 min. Utilizing the k-means algorithm, players were clustered 1nto 9 distinct player archetypes. Multiple linear regression models were then developed for each archetype, predicting "points per minute" to facilitate player ranking, further refined by a data-driven league quality weighting system. Optimal player cluster proportions were derived from analyses of 2018/19 NBA lineups and 2014-2018 1nternational medal-winning teams to create the most effective team line-ups. The model's utility was demonstrated by selecting a hypothetical Team Canada roster for the 2019 FIBA World Cup, which showed effectiveness 1n 1dentifying a robust team composition and predicting the rise of future high-potential players. Additionally, the model's effectiveness was evaluated by comparing 1ts results to the composition of the 2000 and 2024 Canadian National Olympic teams. The findings revealed 9 unique player clusters, demonstrating the model's potential as a 5aluable tool for guiding coaching decisions 1n drafting and signing future players to fit their roster composition effectively. The model proved to be a novel method for talent identification due to 1ts evaluation of players across leagues and 1ts usage of readily accessible data for coaches and scouts alike. Despite limitations related to data sources and subjective weighting, this research provides a sophisticated analytical tool for players, coaches, scouts, and general managers, offering a comprehensive league strength metric and a nuanced player ranking system to enhance roster development alongside the expertise of coaches 1n the evolving global basketball landscape.

求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
CiteScore
2.60
自引率
7.40%
发文量
459
审稿时长
15 weeks
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信