Research paper (original)
EstimatingNumber of goals and shooting efficiency from soccer highlights and verifyingobjective evaluation of teams and players using the K means method.
Ryuji Sasaki
Tokai University, Faculty of Physical Education, Department of Athletic Sports
Keywords: J1 League, J2 League, performance analysis, soccer, predictive modeling
Abstract
The purpose of this study is to infer goal expectancy and shooting efficiency of players andteams in the J1 and J2 leagues in Japan from soccer highlight videos, and to validate theobjective evaluation of teams and players using the K means meth od and principal componentanalysis.
This study helps to evaluate teams and players in terms of goals scored per season.
Therefore,this study examines all shots in the J1 League (0~22 sections, 429 J1 players) and J2 League(0~23 sections, 475 J2 matches (218 and 253 games) of the 2020 2021 season, asample size of 95% confidence interval.
All shots were classified into sections on field play, and theoretical goal values were applied toeach area.
The factors analyzed were the shot distance from the center of the goal and the shot angle.
It has been suggested that shot distance and shot angle are important when calculating goalexpectation values. By combining these two factors, the expected goal value can be calculatedmore accurately than using e ach variable alone.
Furthermore, this examination of goal expectancy provides relatively accurate evidence of thecharacteristics of the top and bottom teams in the league.
Managers and coaches may find this study useful in evaluating teams and players thatconsistently score near or above their expectations.
Research paper (original)
Expected Goal (article)
https://www.guillembaches.com/en/how-to-predict-expected-goals/
この記事が気に入ったらサポートをしてみませんか？