Filtering by
- All Subjects: NBA
- Creators: WPC Graduate Programs
- Creators: Casas-Arce, Pablo
- Creators: Matejka, Michal
- Member of: Theses and Dissertations
- Resource Type: Text
Home advantage affects the game in almost all team sports across the world. Due to<br/>COVID and all of the precautions being taken to keep games played, more extensive research is able to be conducted about what factors truly go into creating a home advantage. Some common factors of home advantage include the crowd, facility familiarity, and travel. In the English Premier League, there are no fans allowed at any of the games; furthermore, in the NBA, a bubble was created at one neutral venue with no fans in attendance. Even with the NBA being at a neutral site, there was still a “home team” at every game. The sports betting industry struggled due to failing to shift betting lines in accordance with this decreased home advantage. With these leagues removing some of the factors that are frequently associated with home advantage, analysts are able to better see what the results would be of removing these variables. The purpose of this research is to determine if these adjustments made due to COVID had an impact on the home advantage in different leagues around the world, and if they did, to what extent. Individual game data from the past 10 seasons were used for analysis of both the NBA and the Premier League. The results show that there is a significant difference in win percentage between prior seasons and seasons behind closed doors. In addition to win percentage, many other game statistics see a significant shift as well. Overall, the significance of being the home team disappears in games following the COVID-19 break.
Sports analytics is a growing field that attempts to showcase interesting aspects of a sport with the use of modern technology and machine learning techniques. This thesis will demonstrate how the NBA has progressed in the past decade by comparing the performance have five teams (SAS, OKC, PHO, MIN, and SAC). It will also provide key insight on what an NBA team should focus on to build an optimized NBA team composition, which will better their performance in the league, which will improve their chances of making into the playoffs. These teams were chosen after conducting extensive analysis on all NBA teams. These five teams were chosen because of the variability in performance (two successful and three less successful teams). Two successful teams, SAS and OKC, and three less successful teams, PHO, MIN, and SAC, were chosen to exemplify the different approaches of teams in the NBA and to distinguish what an NBA team should consider build an optimized team composition to better their performance in the league stage.
The National Basketball Association is one of the most popular and most profitable sports leagues in the entire world, and with stars like Giannis Antetokounmpo, Nikola Jokić, and Luka Dončić, it continues to expand its international reach. In the past decade this has meant that the salary cap has continued to increase considerably. From 2013 to 2017 the salary cap increased by about $40.5 million from around $58.5 million to $99 million meaning there was an extra $1.2 billion worth of cap space to fill. All this new cap space created a perfect storm for numerous players to be overpaid. Many saw the performance of these overpaid players as a part of the contract year phenomenon where a player performs better before a new contract and then after receiving their new contract, their performance deteriorates. The purpose of this research is twofold. First, it looks at whether the contract year phenomenon has been present in the NBA since 2015. After that it looks to find what statistics are the best predictors for performance based on their positions. This was done through various statistical analysis techniques such as T-tests and piecewise regression. Box score statistics like point, rebounds, and assists as well as advanced metrics like player efficiency rating, usage percentage, and true shooting percentage were utilized in this study. The results indicated that the concept of the contract year phenomenon was present in the players sampled. However, rather than contract year only being for players who increased their performance in the previous year, it is a more general phenomenon. Also, there was major differences in the statistics that predicted performance. The biggest of these was the importance of usage percentage rather than points and that centers had the least predictors, most likely due to the evolution in the play of centers.