IDO MOVEMENT FOR CULTURE

Background. A year and a half before karate was scheduled to make its debut appearance in the Olympics in 2020, the World Karate Federation (WKF) elected to change its judging system for kata to a point-based system reminiscent of other Olympic sports.
Problem and aim. This paper analyzes data collected from WKF tournaments over the first year of the adoption of the new karate judging system to evaluate their implementation and factors that affected scores and inter-judge agreement.
Methods. The result books from the WKF’s 7 Karate1 Premier League and 4 Karate1 Series A League tournament were analyzed (3445 total kata performances). The distribution of scores and the standard deviation of judge scores per kata performance were calculated. Technical and athletic scores of each judge were plotted and Pearson’s correlation coefficient was found. Singular value decomposition (SVD) was used to cluster kata that were often run together by competitors and ANOVA and Tukey’s post hoc test were run on scores from the kata clusters.
Results. The frequency of scores follows a normal distribution with a slight negative skew. Judge agreement was affected by the quality of performance where better kata performances had a higher agreement. While judges give two independent scores for each technique and athleticism, there is a strong correlation between the two scores (r=0.932). Clustering and analysis of scores based on kata chosen to run showed that certain kata styles score higher.
Conclusions. This paper is the first attempt to evaluate the new kata scoring system and to identify extrinsic factors that affect the scoring.