Scoring Bias of Judges in International Figure Skating
The purpose of our project is to shed light on the scoring bias of the judges in international figure skating competitions, mainly based on nationality. We will analyze the data on points given to skaters by the judges collected from such competitions. Initial analysis will be on judges having biases for skaters from their own nationalities. Secondary analysis will be on the positive or negative bias of certain countries for other countries. We will make a website where the users can view and filter the data and analysis on tables and graphs.
While we are not figure skating officials, we hope to bring awareness to bias in figure skating judging through a statistics-based approach. Biases can be seen through judges consistently giving skaters lower or higher scores relative to other judges on the judging panel. We will compare the points given to similar moves made by the skaters by different judges, scores that increase the standard deviation and scores that stand out as vastly different from other judges’ scores to the same skaters. Using this analysis, we will design a website and showcase our results using various graphs and charts. We plan the website to be view-only, where people can filter the data and use different methods to display it.
Subjective sports scoring and officiating are often biased because they are human. An example is the 2022 Salt Lake City Olympics figure skating scandal, where the scoring investigation resulted in judges being dismissed. Most sports scoring/officiating is generally fine (like who is the fastest in swimming/track, who wins tennis matches, etc.), but more so for sports with an artistic component, it is often biased because of the subjectivity of the human factor. These biases can include factors such as politics, gender, faith, etc. Our database and analysis of the data are designed to bring awareness to such bias and encourage discussions regarding ways to improve scoring. Specifically, we analyze whether a particular judge will always score a player of a particular country lower than other judges. If this is the case, is it possible to reduce the weight of the judge's score in the overall scoring to achieve a fairer score? This was pointed out during the 2022 Winter Olympics in Beijing (source: https://twitter.com/SkatingScores/status/1573349500810567680?cxt=HHwWgIDR9Z7C1NUrAA AA) , but his analysis and visualization of the data only showed the results for one skater's data analysis. What we want to do is to combine all the scores of each judge to determine if the judge may have bias.
Our data sources are currently decided as three data sites, the first one is https://skatingscores.com/, the second one is https://github.com/BuzzFeedNews/figure-skatingscores/tree/master/data/json, and the last is https://github.com/BuzzFeedNews/2018-02-figureskating-analysis/tree/master/data . This data includes information about the athletes, the judges and detailed scoring. All we need to do is read the JSON file and insert it into the corresponding table, e.g. grab the judges' information and insert it into table Judges. Skating scores' data is highly visualized, so we can better understand how to analyze each part.
| Info | Description |
|---|---|
| TeamName | Mango |
| Captain | Athena Zheng |
| Captain | athenaz2@illinois.edu |
| Member1 | Zepei Li |
| Member1 | zepeili2@illinois.edu |
| Member2 | Murat Altindag |
| Member2 | murata3@illinois.edu |
| Member3 | Victoria Buszek |
| Member3 | vbuszek2@illinois.edu |