Clephas, C.; Stergiou, P.; Tyreman, H.; Katz, L.

(Predicting Olympic success by regression modeling in sport - an analysis of the beginning of the 21st century)

Competition results analysis are an increasingly important tool for national federations to compare and understand the performance of their own athletes compared to their competitors all over the world. Because of this, performing competition results analysis is becoming more important. The aim of this study was to establish if a connection existed between World Cup results and performance at the Olympic Games to predict performance based on number of medals won per nation. In a retrospective analysis of competition results for 14 winter sports between 1998 and 2010, we analyzed the predictive power of world cup results in connection to the Olympics. The results showed that there was a very strong relationship (r2 = 0.91) between the outcomes during the four-year time span before the Olympics compared to the results at the Olympic Games. Analyses showed that it was possible to predict the medal outcome by a nation, depending on the sport. It may also be possible, to a lesser degree, to predict which athletes won medals. Additionally, this study made reasonably accurate predictions over the entire data set and the various Olympic cycles. Future research will include all individual sports in this data set to test the predictive power of the presented models.
© Copyright 2023 13th World Congress of Performance Analysis of Sport and 13th International Symposium on Computer Science in Sport. IACSS&ISPAS 2022. Veröffentlicht von Springer. Alle Rechte vorbehalten.

Bibliographische Detailangaben
Schlagworte:
Notationen:Naturwissenschaften und Technik
Tagging:Regressionsanalyse
Veröffentlicht in:13th World Congress of Performance Analysis of Sport and 13th International Symposium on Computer Science in Sport. IACSS&ISPAS 2022
Sprache:Englisch
Veröffentlicht: Cham Springer 2023
Online-Zugang:https://doi.org/10.1007/978-3-031-31772-9_13
Jahrgang:1448
Seiten:60-63
Dokumentenarten:Artikel
Level:hoch