Data driven approach to sports management : a case study using major league baseball

Baseball is considered to be the national sport of the USA but its popularity has declined in the last few years, mostly due to people’s interests in other sports. Not many sports come close to baseball regarding statistical analysis where everything concerning the sport is carefully registered. There is one statistic variable who has gained more attention in the later years. This variable combines all the statistics of a player into one number which depicts how many wins that player adds to the definite minimum of a replacement. This variable is called wins above replacement or WAR. This project endeavors to see if the possibility to use WAR, exists to predict if a team reaches the playoffs or not. It subsequently attempts to see if this variable is equipped to create optimization model, which should simplify coaches’ decisions in signing players. Data from 1969 up until 2014 were used to create a database where players had been connected to the team they had started each season with. Statistical data for upcoming season were minimal in processing the data. The study’s results indicate the possibility to use WAR up to a certain point to predict whether a team qualifies to the playoffs or not. A logistic regression provided a model with a TPR close to 60% and an accuracy of approximately 70%. These are rather high percentages when there were several factors which limited the value of the calculations. Eliminating those factors would result in decreased errors and more accurate calculations.

Author dc:creator

• Jón Ragnar Guðmundsson 1985-

Contributors dc:contributor

• Háskólinn í Reykjavík

• Rekstrarverkfræði
• Tölfræði
• Bestun
• Hafnabolti
• Tækni- og verkfræðideild
• Meistaraprófsritgerðir
• Engineering management
• Optimization
• Statistics
• School of Science and Engineering

Language dc:language.iso

en

Handle dc:identifier.uri

http://hdl.handle.net/1946/22334

OAI identifier oai:identifier

oai:skemman.is:1946/22334