Linear Regression - Sports Analytics Methods скачать в хорошем качестве

Linear Regression - Sports Analytics Methods

https://www.agilesportsanalytics.com Take a free sports analytics assessment: https://www.agilesportsanalytics.com/... Linear Regression Victor Holman, The Sports Analytics Expert, presents his Sports Analytics 3 Minute Drill - Linear Regression. Linear regression is a linear approach to modelling the relationship between a scalar response (or dependent variable) and one or more explanatory variables (or independent variables). The case of one explanatory variable is called simple linear regression. For more than one explanatory variable, the process is called multiple linear regression. This term is distinct from multivariate linear regression, where multiple correlated dependent variables are predicted, rather than a single scalar variable. Drafting Errors and Decision Making Bias in the NBA Draft This is a review of the NBA draft research conducted by Daniel Sailofsky, applying linear regression modeling. Every year NBA teams struggle to determine who their draft choices should be. This study focuses on the choices that teams have made in order to analyze decision making biases and errors. The data for this research came from the NCAA players who were drafted by NBA teams between the years 2006 and 2013. Information regarding the draftee's -court statistics -conference they played in -the year they were eligible to be drafted -the year they were rookies Data was collected regarding their performance after they joined the NBA including -where they were picked in the draft -their pre-draft physical measurements -their performance -length of their career This information was analyzed to determine how well NBA teams were choosing their draft picks based on their performance after they joined the NBA. The conclusion was that good decisions were not being made when selecting draft choices. The research indicated that most of the factors used to determine who to draft did not relate to the player's future performance. Teams typically want to draft players from well-known colleges that belong to conferences that are traditionally considered strong. It is human nature to avoid change and those making the draft decisions are no different. They are more familiar with the Big Conferences so they feel safer drafting those players. They fail to take into account the strengths of players from other conferences and, as a result, are often not choosing those players who have the best chance of making a difference in the NBA. Research has shown that players who have strong skills regarding ball control and preventing turnovers do well in the NBA. However, these skills have no correlation to the players who are actually drafted. Players with a high rebound percentage and free throw rate in college do better in the NBA than others as well. However, again, these statistics do not relate to those who are actually drafted. The research seems to indicate that those players who make memorable plays in their college days are the ones who are drafted. NBA teams also tend to look at scoring and blocked shots statistics when choosing draftees, but these skills do not actually predict future NBA performance. NBA teams tend to be more interested in the younger, less experienced players. Statistics show that teams are more willing to take on players who have the physique of a typical NBA basketball player even if they lack high-level basketball skills. NBA teams will overlook these players' weaker skills, believing that with proper coaching these players will become better. Unfortunately, size does not guarantee skill levels and no amount of coaching will be able to make these players effective NBA players. Height is positively correlated to draft position but height is not correlated to future performance. Research shows that scoring is positively related to a players' draft position but is actually negatively correlated to future performance. Of course, scoring is a skill necessary to be effective at the NBA level, however, there are other factors that are related to future effectiveness. Star players in the NBA were also strong shooters in their college days, however, they would have strong skills in other areas as well. Teams need to look carefully at how they choose who to pick in the draft. They need to determine what characteristics are related to a college player becoming a strong player in the NBA, and leave their biases behind. Other analytics methods used in this research: linear regression model, probability, percentage-based statistics, variables