Wednesday, January 22, 2014

Basketball on Paper

          When given the opportunity to analyze the girl's basketball game stats, Margaret and I jumped at the chance. Especially since I am on the basketball team, this was a unique opportunity for me to have. Our goal was to find the overall efficiency of the team using areas of statistics that we thought were valuable to the team. Although we encountered many obstacles and potential errors throughout the process, we were able to come up with a fairly accurate analysis of the team as a whole as well as each individual player.

          This idea originated from the book "Basketball on Paper". We were able to read bits and pieces of this book to gain an understanding of the overview of basketball stats and what is important when it comes to making a team. We consulted with Mr. Bernards, as he is doing the same efficiency rating with his eighth grade team. Mr. Pocius helped us get started by giving us his own player efficiency rating formula. He formulated this based on what he thought was needed in each player. The formula is below:

[Points + Rebounds + (2 x Assists) + (2 x Steals) + (2 x Blocks)] / [(2 x FG Missed) + FT missed + (2 x Turnovers) + (2 x Fouls)]

           At first, Margaret and I started to watch each game and take our own statistics so we knew for sure that they were right. After about three games though, we realized that it was a tedious job and we weren't even sure if our stats were as accurate as we thought they were. We resorted to using the stat sheets that are taken on the bench during the game. Although there could have been a couple mistakes, we know that there is a small margin of error, and it was a lot less busywork for us. We took the first eleven games and did a team efficiency rating and a player efficiency rating and put it all on a google spreadsheet. Part of it is shown below.

          We then did confidence intervals for each of the players as well as the team. This worked better than averages because it eliminates outliers. There were many girls who would have either a really good game or a really bad game, and their averages would be skewed by it. Using a confidence interval helped get rid of this problem. Along with the formula from Mr. Pocius is a ranking scale (poor to excellent). We were able to analyze the team and how we changed from game to game. One thing we found was that there tends to be a pattern in the games. We will have a really high efficiency game (a peak) and then the next two or three would be a significant drop. One question that arises from that is why that pattern is so prevalent? This is something that our project did not answer, but would be something interesting to find in the future.  
          I would love to go deeper into this project if there was more time, but for now, I am content with what we have here. I learned a lot about how stats really do matter and how much more you can do with them than just looking at them individually. It really goes to show that everyone contributes to the team, and if the player efficiencies are overall pretty low, the team efficiency will also be low. We had a lot of fun with this project and although it was a challenge at time, it was definitely beneficial to do!

No comments:

Post a Comment