Regression Analysis For Mba

Baseball is a game of extreme statistics, especially in the MLB. Sabermetrics is becoming a very popular cultural term for average Americans. After the release of Moneyball a film that showed the importance of statistical analysis in baseball, the culture was hooked on Sabermetrics. This project will analyze earned run average (ERA) and some of the variables that play a pertinent role in determining a great ERA. ERA is the most popular metric used to decide whether a pitcher is deemed great, good, or bad. ERA is the mean of earned runs given up by a pitcher per nine innings. ERA is similar to golf in the fact that the lower the pitcher's ERA is the more effective he is. You must include the innings pitched because you have pitchers who will pitch for seven innings and pitchers who pitch for one inning in a game. Pitchers have different styles to be effective, sticking out a lot of hitters versus allowing hits but getting ground balls so the defense can make a double play.

In this model, I want to analyze what variables had the largest impact on a pitcher's ERA. Specifically, I am interested in testing which individual statistics play the biggest role in a …show more content…
I imported data for 30 starting pitchers in the year 2016, all pitchers pitched in the MLB in the year 2016. Next, I decided what independent variables I wanted to include in my multiple regression. I included strikeouts per nine innings (K9), walks per 9 innings (W9), if the pitcher through a complete game (CG), hits per nine innings (H9), and WHIP. WHIP is a statistical measure to determine how many runners get on base versus a pitcher, a lower WHIP is good. The variable complete games, is being used as a dummy variable, either a pitcher pitched a complete game in the season or they did not. I ran a multiple regression on ERA using the five independent variables listed